NꙮW WITH MꙮRE MULTIꙮCULAR ꙮ

Open Thread 75.25

This is the twice-weekly hidden open thread. Post about anything you want, ask random questions, whatever. You can also talk at the SSC subreddit, the SSC Discord server, or the Cafe Chesscourt forum.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

Leave a Reply

797 Responses to Open Thread 75.25

  1. Scott Alexander says:

    If impeachment charges for Trump came before the House, would the House be biased against impeaching since they’re majority Republican? Or would they be biased in favor of impeaching since they’d prefer an orthodox Republican president like Pence to a loose cannon like Trump?

    Would the Republican Party probably whip its members to agree with whatever Paul Ryan decided? Or could a faction of pro-Pence Republicans appear and unite with the Democrats to push impeachment charges through regardless of the party line?

    • andrewflicker says:

      I think it depends partly on how much damage Trump appears to be doing to A) individual republicans and B) the republican brand against GOP-supporting voters. If Trump still has 80+ popularity with self-described Republicans, there’s little incentive for GOP congressmen to break ranks with him and join the Democrats’ effort.

    • Wrong Species says:

      For better or worse, Republicans have decided to stick by Trump. Something much more substantial has to come up before they decide to dump him.

    • HeelBearCub says:

      Absent something that convinces enough Republican base voters that Trump deserves impeachment, Republican House members are disinclined to impeach Trump.

    • BBA says:

      I’m in the “zero chance of impeachment” camp, at least until the tax reform bill passes. The whole Republican party has to be on board or else the split between Trump and anti-Trump factions would cripple the congressional agenda for the rest of the session.

      Even if the Democrats win the House in 2018 (a tall order for several reasons) there’s no way they get 2/3 of the Senate, so the same analysis applies.

      (There is also zero chance of Trump resigning, since quitting is for pathetic losers. Sad!)

    • meh says:

      House republicans will not vote to impeach unless they know for certain senate will convict. No one will be willing to vote to impeach if there is a possibility of him remaining in power.

      • alamesage says:

        I’m not sure that that is true. If the House passes articles of impeachment, then the President has to offer a defense in the Senate for several weeks, which is great negative publicity. See: Bill Clinton.

    • Conrad Honcho says:

      You can’t be impeached just because “we don’t like you.” In your scenario, what high crime or misdemeanor is Trump accused of, and how strong is the evidence?

      • BBA says:

        You can’t be impeached just because “we don’t like you.”

        Sure you can. Andrew Johnson was.

        • Conrad Honcho says:

          No, Johnson was impeached for intentionally violating the Tenure of Office Act.

          We don’t have a parliamentary system where the executive derives his power from the legislature and serves at their pleasure and can simply be deposed by a vote of no confidence. The executive is its own branch, derives its legitimacy from power granted by the states, and the chief executive must actually do something criminal to be removed. Article II Section 4:

          The President, Vice President and all civil officers of the United States, shall be removed from office on impeachment for, and conviction of, treason, bribery, or other high crimes and misdemeanors.

          So, very relevant to “would the congress vote to impeach” is “what are you impeaching him for and how strong is the evidence?” If you’ve got video of Trump shooting a man on 5th Avenue in cold blood, it’s going to be hard to vote not to impeach. If the charge is jaywalking and the evidence is “some Democrat says they saw him do it”…eh not so much.

          • HeelBearCub says:

            I agree that there have to be some charges to hang the impeachment upon. Although, this is really a constrained only in that it is a norm not to use the process frivolously. ETA: And I think that norm is largely maintained because 2/3rd’s of the Senate is such a high bar to clear. The odds of a sitting President having 2/3rd’s of the Senate being of the opposition party roughly zero.

            But it’s fairly immaterial to Scott’s original question, as he assumed that such charges existed:
            “If impeachment charges for Trump came before the House”

          • Conrad Honcho says:

            @HeelBearCub

            That seems to imply the vote would be independent of the facts of the case. With that much cynicism any answer is meaningless.

            While I think there are a lot of Republicans who don’t like Trump, the idea of giving a win to the hated Democrats so they could crow about an evil impeached Republican president would be worse. I’m reminded of the example in “I can tolerate anything except the outgroup” of the Muslim who wouldn’t concede a point to Bill O’Reilly because screw you Bill O’Reilly I’m not giving you ammo.

            So, Republicans would be motivated to vote against impeachment and Democrats obviously for. Which means it comes down to the strength of the case and plausible deniability. The cold blooded murder on tape gives you no plausible way to vote against impeachment, but the jaywalking charge is an obvious political witch hunt.

            Meaning the strength of the charges is critical.

          • BBA says:

            Passing a blatantly unconstitutional law meant to cripple this particular President, and then impeaching him for inevitably violating it, rounds off to “impeached because we don’t like you” in my book.

          • HeelBearCub says:

            @Conrad Honcho:

            That seems to imply the vote would be independent of the facts of the case. With that much cynicism any answer is meaningless.

            No, I’m saying there isn’t anything de jure to keep this from happening.

            Posit a counter-factual world where all that is needed to remove a President is a majority vote in both houses. I think eventually that system devolves to simple votes of no confidence.

            But needing to have members of the other party, and substantial numbers of them, to actual remove the President forces that the spirit of the Constitutional clause be upheld. There is a de facto expectation that impeachment conviction votes in the Senate should be meaningful (although, look at Clinton’s impeachment and tell me this really holds true? Regardless of how you feel about Clinton?)

            In any case, this is still immaterial to Scott’s original question. The assumption was that we had charges.

            You can, of course, ask the question “Under what circumstances would The House Committee on the Judiciary in a Republican House actually allow impeachment charges to proceed?”

            I mean, Jerry Nadler already attempted to move impeachment forward

          • Conrad Honcho says:

            @BBA

            Why would the president sign into law a blatantly unconstitutional bill, that he would then personally break?

            If you mean he’d then enforce the unconstitutional law, that’s not one of the criteria for impeachment. If he tried to enforce and unconstitutional law, it would be challenged in the courts, the Supreme Court would rule it unconstitutional, and then it would no longer be enforceable. We’ve had lots of laws the court has struck down. The president operating in good faith implementing those laws before they’re stricken isn’t committing any crime and has never been impeached for it, nor could he be.

            Remember, impeachment is not for “the President is doing a bad job” or “the legislature doesn’t like the job the President is doing” or “polls show people don’t like the president.” It’s to remove a president who has actually committed crimes. You need to show what law the president violated and provide evidence. You can’t just put “for being a big dumb meanie racist sexist literally double mega hitler” on the articles of impeachment. That’s not a crime.

          • Brad says:

            The Tenure in Office act was passed over Johnson’s veto. At the time the Declaratory Judgment Act was not in existence and there would have been no way for Johnson to test the Constitutionality of the Tenure in Office Act other than by contravening it.

          • Conrad Honcho says:

            @HeelBearCub:

            In any case, this is still immaterial to Scott’s original question. The assumption was that we had charges.

            And I’m saying the assumption doesn’t give you enough information for a meaningful answer because it depends entirely on what the charges are.

            “Assume you’re driving a car. How fast are you going?” Just giving someone an assumption doesn’t mean you can provide a meaningful answer. Kind of depends on where you’re driving, why you’re driving, what the other drivers are doing…

            You can, of course, ask the question “Under what circumstances would The House Committee on the Judiciary in a Republican House actually allow impeachment charges to proceed?”

            That’s a better question! And my answer is it would have be something undeniably wrong and evil. And a grave matter (not just lying even under oath about sex). Even Republicans who don’t like Trump don’t want him impeached because that would be a nuclear explosion win for Democrats that they would lord over Republicans for all time. It is not going to happen for something that isn’t airtight and undeniable.

            I mean, Jerry Nadler already attempted to move impeachment forward

            Did you read it? Maybe seems a little more like political grandstanding than putting forth an actual charge of any kind?

          • HeelBearCub says:

            (not just lying even under oath about sex)

            I’m confused.

            Given that this has already happened, on what leg do you stand here?

            Remember, you originally said:

            the executive is its own branch, derives its legitimacy from power granted by the states, and the chief executive must actually do something criminal to be removed.

          • Deiseach says:

            But it would still turn on “what are the charges?” Trying to impeach Clinton for having adulterous sex (and lying about it, hence the obstruction of justice charge) wasn’t good enough, so trying to impeach Trump for “we don’t like him” certainly won’t be good enough. He has dodgy financial dealings? Do any politicians want to open that cupboard and see the skeletons tumbling out, because there are plenty more besides Trump getting favours and doing business.

            Besides, with the sore losers having fevered fantasies of impeaching Trump even before the inauguration took place, you would need to show very convincingly that this impeachment attempt wasn’t simply trying to overturn the result of an election that didn’t go your way:

            We launched this campaign on Inauguration Day based on the President’s choice to profit from the presidency at public expense, in violation of the United States Constitution. Over 950,000 people joined the campaign in its first 110 days. Now we are expanding the impeachment campaign in response to the President’s apparent interference with a criminal investigation by firing FBI Director James Comey.

        • cassander says:

          I actually think America would be a better place if Johnson had been convicted, but he wasn’t.

          • Gobbobobble says:

            Oh? It’s been some years since APUSH, most of what I remember was that Johnson was a lame duck anyway. What would removing him have accomplished other than demonstrating the ability to have disliked Presidents removed?

          • cassander says:

            @Gobbobobble

            It’s not so much that I think Johnson was terrible (I mean, he was, both of them actually, but that’s not the point) but that it would have established a precedent that presidents serve, to a degree at least, at the pleasure of congress and made the american system more parliamentary, which I think would be a good thing institutionally. Presidentialism has one or two good features, but they don’t come close to outweighing the bad features

          • Gobbobobble says:

            All right, thanks. So I was close with “demonstrating the ability to have disliked Presidents removed” 🙂

            Now I’m curious what debates the Founding Fathers (praise be unto them) had for Presidential vs Parliamentary. Had the English Parliamentary system fully evolved by that point? I know they still had a monarch with actual power back then, so maybe it’s a case of their current system came after ours and was able to improve on it?

          • BBA says:

            @cassander: Agreed on that. In any sane system, Obama and Biden would have both been forced to resign after the debacle of 2013 and John Boehner would have become the first orange president.

            Really, the Reconstruction-era Congress should have dispensed with the idea of trying to wedge a major restructuring of federalism into the Constitution through the 14th Amendment and just rewritten the thing entirely, with the preceding decade’s worth of constitutional crises in mind. But Founding Father ancestor-worship was strong enough to prevent it even then. Now it’s even stronger.

          • Gobbobobble says:

            Gobbobobble says:
            May 11, 2017 at 6:12 pm
            […] the Founding Fathers (praise be unto them) […]

            BBA says:
            May 11, 2017 at 6:13 pm
            […] But Founding Father ancestor-worship was strong enough to prevent it even then. Now it’s even stronger.

            I find this coincidence highly amusing

          • hlynkacg says:

            This is not a coincidence, because nothing is ever a coincidence.

          • Nyx says:

            > Had the English Parliamentary system fully evolved by that point? I know they still had a monarch with actual power back then, so maybe it’s a case of their current system came after ours and was able to improve on it?

            By the late 1700s, monarchs did not have much power to reject Parliament and never refused Royal Assent to any Act of Parliament that passed both houses. They had lost their financial powers. The Crown had decisively lost to Parliament in both the Civil War and the Glorious Revolution, which deposed Charles I and James II respectively. Parliament manipulated the succession (so as to avoid any Catholics) and ended up putting it in the hands of a line of German aristocrats who had poor grasp of English and little talent for governance, under whose reign the powers of the Crown further diminished. The transition was not total until the 19th century.

            Secondly it’s difficult to say that the English were “inspired” by the Americans. As you describe it aptly, the British system evolved according to crisis and circumstance, with the balance of power already being decisively in Parliament’s favor by the 1780s. But the system of checks and balances that is so critical to the Constitution is nowhere to be found in British law. A simple parliamentary majority is all that is required to enact huge constitutional change (and that not requiring a massive preponderance of votes), with naught to oppose them but a supine Crown and the mostly defanged House of Lords (with every government threatening to cripple it further).

      • Iain says:

        A hostile Congress could pretty easily put together a case using the Emoluments clause.

      • mtraven says:

        Um, have you been asleep this week when Trump basically confessed in a TV interview to obstruction of justice?

    • J Mann says:

      A useful model (I can’t remember if I heard it from a professor or on EconTalk) for legislator behavior is “Assume that legislators are driven primarily by two goals. The second most important is that the legislator wants to do the right thing as she sees it. The first is that she wants to keep her job.”

      In this case, the question on impeachment is whether the legislators are concerned about being challenged and losing – either losing to a populist-right candidate if they vote to impeach, or by a left candidate if they don’t.

      My guess is that unless there’s a public perception among independents and conservatives that Trump is not just Trump-y but actually seriously corrupt, most GOP legislators would be more in danger if they voted to impeach than if they didn’t.

    • Deiseach says:

      Is impeachment a good idea, though? So early in his presidency and really based mainly on “The Democrats think Hillary should have won”?

      I know the Russian rumours are serious but until we see actual evidence of Trump taking orders from Putin, him simply being exactly the blowhard vulgarian outsider that his campaign presented isn’t good enough.

      What will they put in place? The Dems don’t want President Pence, will they demand a new election? Can that happen legally? Notions of fudging up “Pence takes over and then profit happens mysteriously” aren’t much of a plan.

      • Eric Rall says:

        The Constitution is pretty explicit that the VP becomes President for the remainder of the current term if the President dies, resigns, or is removed from office. It also explicitly gives Congress latitude to decide by law what happens if the Presidency is vacant and there’s no Vice President. Current law is that the Speaker of the House becomes President, or (if that’s also vacant, or if he’s ineligible or refuses), the line of succession is the President Pro Tempore of the Senate, and then the members of the Cabinet in order of the seniority of their departments (State, then Treasure, then Defense, etc).

        But in the past, there have been provisions for a special election if there’s no President or VP. In theory, we could see a special election if Trump is removed from office, Pence is also removed (or agrees to resign, or dies in office before Trump is removed or at least before Pence can appoint a new VP and get him confirmed by Congress), and Congress changes the law in advance of removing Trump. I can imagine circumstances where this might be plausible, but those circumstances seem spectacularly unlikely.

        • John Schilling says:

          I’m not sure it is plausible at all. Current statutes don’t allow a special election for President In order for such an election to occur, both houses of Congress would have to pass a law calling for it. But, in the absence of such a law, the presidency (when both POTUS and Veep are absent) goes to the leader of the ruling party in the House and/or senate.

          Under what circumstances would it be plausible for the ruling party in congress to pass a law calling for a presidential election that their party might lose, when the status quo guarantees that the presidency goes to their party?

          • Eric Rall says:

            The spectacularly unlikely scenario I was imagining was post-midterm-elections:

            1. Dems have a majority in the House (I’d guess this has about a 30% chance of happening), so the Speaker is a Democrat.

            2. Pence has died or resigned (I think this has less than a 1% chance of happening in the timeframe we’re considering), and Trump has failed to appoint a successor whom Congress is willing to confirm (maybe 40% chance of happening given a Dem majority in the House). Thus, the Speaker is next in line under current law if Trump is removed.

            3. Trump’s facing impeachment charges bad enough that many (enough to combine with Dems to make a 2/3 majority) Senate Republicans would prefer a special election to keeping Trump in office, but not so bad that they’d be willing to bite the bullet and remove him in favor of the Speaker. This is a very specific level of bad.

            4. Dems would have to want Trump gone badly enough that they’re willing to cut a deal with Senate Republicans on the succession bill, rather than making Republicans own an unpopular vote to block Trump’s removal.

            5. The crisis would have to come to a head early enough before the 2020 election that cutting a deal for a special election would be worthwhile rather than just running out the clock until the regular election. This is a fairly narrow time window given that we’re assuming this happens post-midterms. Maybe the first six months of the new Congress.

            6. Enough House Republicans would have to be of the same mind as the Senate Republicans to combine with Dems to override Trump’s inevitable veto of the succession bill.

            7. Trump doesn’t decide to resign before the revised succession bill is voted into law, which if he did would make the Speaker the President under the existing succession law. I could very well see Trump doing this, as a parting shot to Congressional Republicans for supporting his removal.

        • Deiseach says:

          My understanding was that Pence is hated by a certain proportion of the Democrat voters (or at least those representing themselves as Democrat supporters and speaking to an audience of the same), so would impeaching Trump to replace him with Pence really be an appealing vista for them? I saw scripts and directions for “call your local representative about voting on the healthcare bill” all over the place, with “threaten Republicans to vote them out of office”. I’m sure that a similar “ring up your Democrat representative and threaten to vote them out of office” campaign would be organised if there was a chance Pence would take over from Trump, so for the Democrats to push this, they need to win something, and President Pence does not seem like a victory to me, when he’s going to establish torture camps for gays as soon as he gets into power:

          his favorite movie is … oh Jesus, get a load of this: The Wizard of Oz. Of COURSE it is. Of course a man who wants to take money from AIDS programs and use it to fund ex-gay torture camps loves the most iconic gay movie ever made. Of course a man who wants to withhold medical care from women loves a movie with strong female leads.

          So that’s why I asked could they demand a new election. What about if they try impeaching Pence as well as Trump, for whatever grounds? Would they be happy to risk “the leader of the ruling party in the House and/or senate” becoming the new president?

          • hyperboloid says:

            The constitution is very clear on presidential succession.

            To quote article two section six:

            In Case of the Removal of the President from Office, or of his Death, Resignation, or Inability to discharge the Powers and Duties of the said Office, the Same shall devolve on the Vice President, and the Congress may by Law provide for the Case of Removal, Death, Resignation or Inability, both of the President and Vice President, declaring what Officer shall then act as President, and such Officer shall act accordingly, until the Disability be removed, or a President shall be elected.

            The constitution is silent on succession after the vice president, but as a matter of federal law, the line passes through the speaker of the house, the President pro tempore of the Senate, the Secretary of State, then through a list of cabinet members which can be found here.

            Now, could the democrats pass a law allowing for a special election after impeaching both Pence and Trump?

            No, almost certainly not, as article II clearly states that congress must select an existing officer of the federal government.

            What about if they try impeaching Pence as well as Trump, for whatever grounds? Would they be happy to risk “the leader of the ruling party in the House and/or senate” becoming the new president?

            When you understand the workings of the US constitution that question doesn’t really make any sense. To remove a president requires majority of the house to impeach, and a two thirds majority of senate to convict. If the Democrats had such majorities then the the presidency automatically goes to the democratic speaker of the house, if they do not, then how are they going to get the Republicans to impeach Pence on presumably trumped up (no pun intended) charges?

            I don’t see why would they want to impeach pence anyway, given that the earliest they could do it would be 2018, and President Pelosi (or whoever) would almost immediately have to face the judgment of voters in the 2020 election. Given that this is a scenario where the Democrats have just won a landslide majority in both houses of congress, why take the risk? Just let Pence serve out the rest of Trump’s term as the lamest of lame ducks, and defeat him in the general election.

            As for your claim that Democrats believe that Mike Pence, Donald Trump, or Republicans in general want to round up gays and put them in camps; you’ve said this repeatedly, and I’m glad you’ve finally cited a source. And what a source it is.

            A blog post on the local member of the gay press named Matt Baume on the Seattle Stranger website; which is if, you don’t know, an “alternative” counter culture newspaper that used to print, among other things, a regular sex advice column written by a professional dominatrix.

            What Baume is talking about, in very unfair and hyperbolic terms, is the allegation that Mike Pence supports providing federal funding to providers of so called gay “conversion therapy”.

            Here is a link to a story in which a gay activist named Samuel Brinton describes ugly such “therapy” can sometimes be.
            To quote from the article:

            Britton said they were 11 years old when their parents sent them to see a conversion therapist.

            “There were seven King James Bibles on a stack on the coffee table,” Brinton said, recalling the conversion therapist’s small office in an Orlando, Florida strip mall.

            Brinton said the so-called therapist used extreme and abusive techniques, including covering Brinton’s hand in ice, wrapping it in conductive wires to apply heat, and even electroshock therapy, while forcing Brinton to look at images of men touching. Brinton said the therapist stopped the techniques when he showed Brinton images of men with women.

            “First you’re supposed to learn that it’s painful to touch men, then you’re supposed to learn that it’s not painful to touch women. Then you’re supposed to learn that it’s excruciating if you’re have sex with men,” Brinton explained.

            What you may not know about our fine country, is that we have a small industry that operates prison like private facilities that claim to “straiten out” troubled teenagers. All you have to do is sign your kid over to a boarding school/re-education camp, pay a huge “tuition” fee, and all your parental discipline problems will be solved. Gay conversion actually a tiny percentage of the market for these places, and vast majority of them deal with drug and alcohol abuse, mental health problems, or general bad behavior. Here is an article in the American Conservative about them; they are basically offering child abuse for hire, have actually killed kids in some cases, and should be banned.

            Now, does mike Pence actually support any of this?
            No, as there is no evidence that pence ever supported funding therapy of any kind for unwilling minors.

            Baume is basically playing the same game that the right plays when they accuse the Democrats of wanting to murder babies when they support funding for planned parenthood’s (non abortion related) women’s health services.

            The one important difference being that free cervical cancer screenings are probably a good thing, whereas spending federal money to help people overcome there homosexual urges is one the stupidest wastes of taxpayer dollars I have ever heard of.

            In summation:

            Nobody thinks Pence is going to round up gay people. Any realistic attempt to impeach Trump is an attempt to make Pence president, and can only succeed with republican support in the senate. If the democrats do this it will because they genuinely believe Trump is a unique threat, in a way that an ordinary republican is not.

            And finally, you have a history of going off on kulturkampf inspired rants about things that you claim American liberals believe, but would strike most people on the mainstream American left as very strange. When you do provide citations for these views they are, almost with out fail, from obscure fringe sources or you’ve misunderstood what there saying, or both.

    • Nyx says:

      It would be electoral suicide for House Republicans to reject Trump. It would confirm the suspicions of many conservatives who are skeptical of the Republican party structure and see it as a stooge for corporate interests. They aren’t a majority of the voter base but they are a large enough demographic that Republicans cannot afford to lose them.

    • Kevin C. says:

      Per impeachment of Trump, I note that on one hand I see things like:
      -Rod Dreher fulminating across multiple posts how Trump is a dangerous, lying madman unfit for office and leaking vital secrets to the Russian Enemy, and that if the Republican party wants to have any credibility and future, they must impeach him now.
      -Fellow TAC writer Noah Millman writes in The Week that Republicans must oust Trump in order to prevent a military coup. More specifically, that if Congress doesn’t remove him, our military leadership, in order to fulfil their duty to protect national security, will have to step in and do so.
      James Fallows writes in the Atlantic about “Five Reasons the Comey Affair Is Worse Than Watergate“.

      So plenty across the political spectrum insisting Trump Must Go. What do I see for the opposite argument? The “Jim” position that if Congressional Republicans in want to suvive they’ll uniformly support Trump, because if Trump is ousted, it means that every single last Republican in Congress — even those who “crossed the aisle” to help with the impeachment — will be literally murdered by Leftists.

      So it seems pretty clear which way things look to be trending.

      • Kevin C. says:

        As a follow-up, today at the New York Times, Ross Douthat calls for Trump’s removal via invocation of the 25th Amendment.

        • Edward Scizorhands says:

          I used to think talk that Trump wouldn’t make it through his term were silly.

          Now I’m starting to doubt it myself.

          • LHN says:

            Removal via the 25th amendment is a higher bar than impeachment. Impeachment+conviction requires a majority of the House and 2/3 of the Senate. 25th amendment requires a majority of the Cabinet, 2/3 of the House, and 2/3 of the Senate.

            Given that, I think if he fails to complete his term, impeachment is much more likely. It “only” requires 19 Republican senators and 25 or so Republican representatives (depending on how House vacancies shake out) to decide he has to go. (Vs ~95 Republican reps and half the cabinet for a 25th Amendment determination of incapacitation.)

      • The Nybbler says:

        It’s basically the Fourth Estate versus the Executive. The ever more breathless claims about Trump saying a piece of intelligence came from Aleppo are really getting insane. Even if what the leaker said is true, as the New York Times admits, it’s not unprecedented for Presidents to run their mouths too much in meetings with foreign officials. Damaging? Perhaps; certainly it’s possible the fact that the intelligence came from Aleppo could lead to the Russians discovering the US’s source was Israel. If, that is, the leakers hadn’t sent that information to the New York Times, which is a much easier way for the Russians to find it out.

        The idea that the military would engage in a coup over that is ludicrous.

        • Kevin C. says:

          It’s basically the Fourth Estate versus the Executive.

          You mean the Fourth Estate versus the Chief Executive, with plenty of the “rank and file” in the Executive on the other side. Given this matchup, it looks pretty clear which side is stronger.

  2. Scott Alexander says:

    It seems like Comey’s decision not to press charges against Hillary last summer suggested he was either honest or biased toward Hillary. And then his decision to re-open the case just before the election suggested he was either honest or biased toward Trump. And then his decision to continue investigating Trump’s Russian connections suggested he was either honest or biased against Trump.

    From this I conclude he’s probably honest, since no single bias could account for all of his decisions. Does anyone disagree with this assessment?

    • John Schilling says:

      I lean towards “probably honest” myself, but his recent comments about why he reopened the investigation can be plausibly read as having expected that Hillary had the election in the bag and not wanting her legitimacy (or his own, or the FBI’s) to be undermined by the perception of coverup when Wienergate (Wienerghazi?) erupted a few weeks after the election.

      • Deiseach says:

        Yeah, that evaluation seems about accurate: everyone believed Hillary would win, some people had reason to believe that yet another scandal associated with her and her entourage was going to erupt, and better to get it out of the way and handled during the election than casting a shadow over the inauguration (and ironically in view of what we’re discussing here, provoking calls for her impeachment).

        Anthony Weiner – why has nobody tied concrete blocks to his feet and dropped him in the harbour, rather than letting Huma (who continues to be Hillary’s good pal) take him back yet again?

        I love this conspiracy rant about the machinations to get Hillary via the email thing – see below about ‘baiting’ Weiner to fall for a honeytrap, an act that I imagine is about as difficult to pull off as waving a piece of meat in front of a starving dog and making it try to snatch it out of your hand:

        • An operation to bait Anthony Weiner, the controversial husband of Clinton’s top aide Huma Abedin

        • A successful effort, perhaps from within the FBI, forcing director Comey to utilize the Weiner allegations as a basis to reopen the Hillary Clinton email investigation

        • CatCube says:

          Yeah, Weiner would fuck a snake if somebody held it out straight for him. Why go through the trouble of trying to bait him instead of just waiting for him to do what he does and capitalizing on it?

    • andrewflicker says:

      Honest and idiosyncratic, and perhaps not the most wise political mind.

    • Tom Bartleby says:

      I think there’s another option: he was very focused on appearing to be honest–so focused that he let that become a higher priority than actually being honest. Specifically, I think Comey was optimizing for maximizing the FBI’s perceived legitimacy in the event of a presumed Clinton victory.

      I think his problem is that he (like a lot of people) was way too confident that Clinton would win. Think about it, if Clinton had won despite the scandal, then there’d be a lot of people out there who are inclined to think that Clinton only won because The Establishment went easy on her (not making a claim either way about the truth value here; just that a lot of people would think it). And if the FBI had done things the more normal way, it would have been really easy to paint it/Comey as part of that problem, especially if it later came out that the FBI had found some sort of lead right before the election but kept it quiet.

      Comey’s actions were, I believe, part of a strategy to get people to say that the FBI wasn’t biased in favorof Clinton. In this hypothetical, you’d still have some on the left criticizing the FBI for being biased against Clinton, but since she won anyway, I bet Comey figured they’d move on pretty quickly. Or at least that having them grumble wouldn’t be as bad an outcome for the FBI’s legitimacy as having Fox news say that Clinton was only president because the FBI softballed the investigation.

      All this was based on the assumption that he could be as hard on Clinton as he liked (without actually indicting her) and she’d still win. But then Trump won.

      And that was a real disaster for Comey, in my view. Hed just spent all this time trying to make sure no one would blame him for Trump’s defeat, and all he’d managed to do was make it look like he was to blame for Clinton’s loss–and he’d done so by breaking a lot of informal rules. By trying to maximize the perceived legitimacy, he’d actually crashed it.

      If that’s all right, then his next step is obvious: be super hard on Trump to repair the damage that he did. If he’d stayed in office and had been really hard on Trump too, then his legacy is of someone who was tough on both sides and he rehabilitates the legitimacy of the FBI.

      Which seems to be exactly what he did, leading to Trump firing him.

      In a way, Comey is a tragic figure: he wants to make sure no one thinks he’s biased, so he bends over backwards to avoid the appearance of going easy on the future President. Except he miscalculates, and she looses. Then, he tries to make up for his miscalculation by being hard on Trump, figuring it would be too impolitic to fire him. Except he miscalculates again, and gets fired.

      None of this really expresses an opinion about whether he is honest or not. But I think his highest aim was in trying to chase the appearance of honesty as a higher value than the substance.

      • Very nice analysis. It sounds totally plausible to me. I think you’re right.

      • hlynkacg says:

        My own take is essentially the same as John Schilling’s above but I feel like you may be on to something here.

        • John Schilling says:

          And my own take strongly overlaps Bartleby’s. Comey seems to have wanted to maximize the post-election legitimacy of A: the inevitable Clinton administration, B: James Comey, and C: the Federal Bureau of Investigation. It isn’t clear how he ranked these three, because all three were served by the same actions at every stage.

          Note that there is nothing wrong with these goals; quite the opposite, they were and are laudable ones for Comey to pursue. They do not justify dishonesty or deviation from established norms in law enforcement, and the first does not justify damaging the legitimacy of the potential or actual Trump administration, so if there’s room for criticizing Comey that’s where it will be found. I think he comes out looking pretty good, but not perfect.

          • Paul Brinkley says:

            I think he comes out looking pretty good, but not perfect.

            My take as well, with one addendum: if he had guess the election outcome correctly, he would have ended up looking perfect, in the same way Nate Silver looked perfect for guessing every state outcome in 2012.

    • Protagoras says:

      I suppose the default should be to assume honesty, but evidence like this is very far from being conclusive; a dishonest person who is pro-X can do anti-X things (or vice versa) in a deliberate attempt to conceal their bias, perhaps in cases where they expect (rightly or wrongly) that doing the apparently pro-X thing won’t really benefit X and doing the apparently anti-X thing won’t really harm X in that instance.

    • AnonYEmous says:

      from what I hear, in response to some of the things he said:

      he knew Hillary was a criminal and knew his superior in Lynch wouldn’t do shit about it

      he was forced into reporting the re-opening because he’d promised congress he would do it

      and then he investigated trump’s russian connections because duh

      in other words I’d say he’s biased more towards being stuck between a rock and a hard place. But I still think he fucked up regardless.

    • meh says:

      I think he is biased toward appearing not biased. If you go through his history back to Bush, he enjoys being the non partisan defender of democracy.

    • shakeddown says:

      My impression is he has a republican bias in the same way the NYT (supposedly) has a liberal bias – he cares enough about the idea of being honest that he won’t do something blatantly dishonest like charge Hillary for a non-crime or just let the Russia thing go, but he will go out of his way to make Hillary look bad (saying she did bad or sending that letter to congress) in ways he wouldn’t to Trump.

    • Brad says:

      From this I conclude he’s probably honest, since no single bias could account for all of his decisions. Does anyone disagree with this assessment?

      He’s honest, but that’s not the only virtue.

      He’s also arrogant, self righteous, and high handed. Whether or not there was an ulterior motive to fire him (there was) the deputy AG’s memo was accurate — Comey’s decision to testify before Congress and then hold a press conference (!) regarding a FBI investigation that did not lead to a prosecution was an unjustifiable departure from the rules and norms of the Justice Department.

    • ThirteenthLetter says:

      Counter-suggestion: never attribute to honesty what can plausibly be blamed on incompetence?

      Comey’s flaw was less how he decided to open and close investigations and more how he decided to do it on the front page of the New York Times, which means all the errors and stumbles are on the front page as well. He had to embarrassingly walk back his conclusion about Hillary right before the election, and then a week or so ago walk back the walkback. The Trump/Russia investigation has been no better, with endless leaks and no clear explanation of what precisely it was investigating. Dude was way overdue to take a walk.

      That said, Trump being Trump he found the most chaotic way to make it happen.

      • nimim.k.m. says:

        Counter-suggestion: never attribute to honesty what can plausibly be blamed on incompetence?

        Many are of the opinion that being honest is stupid. Starting from the weird “How-are-you-oh-I’m-fine-thank-you-how-are-you?” ritual to the political savvy.

    • tmk says:

      Biased towards Trump, but pressing spurious charges against Hillary would get him smacked down by the court. Instead he did not press charges while using the attention to make an anti-Clinton speech. Then he re-opened the investigation just before the election, when there was not enough time for him to get in serious trouble.

      However being pro-Trump didn’t save him, in the same way that being pro-Stalin didn’t save anyone.

    • hls2003 says:

      I would go with “biased towards Hillary over Trump, but more biased towards his own status than either one.”

      I agree with the posters above who suppose that Comey’s actions are almost entirely based around what could either be defined (most charitably) as “optics / reputation of the FBI” or (less charitably) “optics / legacy of James Comey’s tenure as FBI director.” At any given point, it is probably reasonable to guess he must have thought Hillary would win, because practically everyone did. He didn’t want to be “the guy” who elected Donald Trump by recommending felony charges against Clinton; so he didn’t. But he didn’t want to be “the guy” who invented a scienter requirement in a strict liability statute merely to damage Donald Trump; so he declined to press charges in the most negative-towards-Hillary way possible. Similarly, when the Wiener stuff came up, he didn’t want to be “the guy” who covered up for Hillary and thus beholden to her; so he raised it. But he didn’t want to be “the guy” who elected Donald Trump, so he quickly followed up with his “nothing new” letter. Continuing the Russian investigation allows him to be both anti-Trump and pro-James Comey’s FBI simultaneously, so that data point is less telling.

      I’m reminded, Scott, of your prior post on how to write persuasively. When you are trying to communicate an unpopular opinion with an outgroup, you use language and style to signal really hard that you are part of their tribe. If you’re trying to communicate with liberals that SJW stuff is unfortunate, spend time bashing Trump and making mocking reference to religion, then slip your point in quietly after that, e.g. “these bozos don’t deserve our time in censoring them on campus, we’re better than this and better than them.” If you ‘re trying to communicate with conservatives that NIH funding is important, spend time slagging stupid government regulations and excoriate wasteful spending before quietly segueing to “stupid liberals would fund everything, so it’s up to us smart conservatives to employ money usefully for once.” That seems to me similar to Comey’s M.O. throughout. When he was taking a pass on the Hillary indictment, he did so accompanied by negative statements about her carelessness. When he was discussing the Russian investigation, he did so accompanied by statements that Trump himself was not under investigation. Unfortunately for Comey, neither version of signalling worked; instead of signalling membership in both teams, he signaled membership in neither.

      It also occurs to me that Comey’s firing is evidence that the Russian investigation has no substance, and specifically that Trump is not worried about it. Sticking with FBI directors, think about J. Edgar Hoover. Why did he survive so long? Because he had the goods on everybody. No President was going to cross him because Hoover had damning information about them. Nobody knows what James Comey has in the Russian investigation thus far; if it were anything substantial, it probably would have leaked by now, but even if kept under wraps, a guilty Trump wouldn’t know one way or the other what Comey had on him. If Trump had a guilty conscience about the Russian thing, he wouldn’t dare fire Comey, especially since there are five (5) parallel investigations going on in addition to the FBI investigation that will remain ongoing.

      I’m not suggesting Trump is a Machiavellian genius on this point; quite the opposite. I think he’s emotional and reacts instinctively with self-interested cunning, which suggests to me that his knee-jerk “fire Comey” reaction is a sign of a clear conscience on the Russian stuff. Trump knows he’s not guilty, and so he’s angry that the investigation is continuing to be in the news; he’s also wary of Comey’s track record of going public with “On the one hand, on the other hand” pseudo-exonerations like Comey did to Hillary. So in that respect, I really do think the firing is all about Hillary – Trump doesn’t want Comey to come out with an attempt to virtue-signal to Democrats even as he formally clears Trump, similar to the Hillary non-indictment’s virtue-signalling to Republicans even as he formally cleared Hillary.

      • Brad says:

        Pretty good comment overall, but to pick a nit this sentence:

        But he didn’t want to be “the guy” who invented a scienter requirement in a strict liability statute merely to damage Donald Trump; so he declined to press charges in the most negative-towards-Hillary way possible.

        is wrong. Scienter is for civil law, mens rea is for criminal law. And there were no strict liability crimes that Clinton potentially violated, every one of them had a mens rea element of one sort or another.

        • hls2003 says:

          Fair enough. I was trying not to use too much legalese and was imprecise. There’s always a mens rea requirement in criminal law, if only to intend to do what you did (even in a strict liability crime like, e.g., statutory rape, you couldn’t be prosecuted if you were under anesthesia the whole time). The more precise way to say it would be “he didn’t want to be ‘the guy’ who invented an ‘intent to harm the United States’ element in a statute which criminalized ‘gross negligence’ regardless of intended outcome…”

          • Brad says:

            I feel like I’m descending into pedant madness, but I’m compelled to object that voluntariness is a part of actus reus not mens rea.

            That aside your more precise version is much better. I’m not sure I agree as a substantive matter, but it is arguable.

            Edit: pendent to pedant

          • hls2003 says:

            @Brad: In my opinion, you are using both terms incorrectly, and also are incorrect that scienter only applies in civil cases. However, since we understand the underlying point, we can agree to disagree.

            EDIT: Correction noted! Also, don’t you mean “pedant” or “pedantic” madness? 🙂

        • BBA says:

          The one that confused a lot of people is the “gross negligence” statute, which Comey refused to recommend prosecution under despite Clinton’s “extreme carelessness.” Aren’t those the same thing, ask the armchair lawyers? (Not quite. There’s also the question of whether Gorin v. United States ruled that element of the Espionage Act unconstitutional; certainly, nobody has been prosecuted under it since.)

          • hls2003 says:

            I don’t read Gorin as calling into question this statute. Constitutionally, it looks to me like they were dealing with a definiteness challenge regarding the definition of “information respecting the national defense,” said to be potentially so broad as to make it impossible to determine guilt (are crop yields information respecting the national defense? how about road conditions? etc.). I would think that concern would be ameliorated in the case of 793(f) because it is already self-limited to classified information and the disposition thereof.

    • cassander says:

      If Comey was really concerned about honesty above all else, why did he not announce his intention to resign regardless of who won? Doing that would have effectively placed him above suspicion. That he didn’t suggests that he had some angle besides honesty. Maybe he really liked the job and wanted to keep it, maybe he didn’t trust either candidate and didn’t want them picking his successor. I have no idea what it was, but there was definitely something going on besides honesty.

  3. bean says:

    I’m going to do another edition of ‘questions I get as a tour guide’ today. (Series index)

    There are two questions I get a lot as a tour guide which might be of interest here:
    1. Does the ship move sideways when you fire the guns?

    No. For those who don’t want to follow the link (which has all the math), if the ship was on a frictionless surface and fired a full broadside at zero elevation, the resultant sideways velocity would be approximately 6 inches per second. As it is, the wall of water the ship runs into (about 850 feet long and 35 high) stops it pretty quickly. The apparent sideways movement in some pictures is the result of muzzle blast interacting with the wake.

    2. Could the ship be reactivated?
    Theoretically, yes, but in practice it won’t happen. There is no situation where the resources required wouldn’t be better spent on something else. Even in a total war, we’d be better off using our shipyards for other things.
    As I’ve commented on before, the Iowas were not reactivated to fight the Kirovs, or to provide gunfire support, but as cruise missile platforms. Today, there would be no reason to reactivate them in that role (particularly as ships routinely deploy with partially-empty VLS), which only leaves gunfire support. However, the opposed amphibious landings that the 16” gun would be the appropriate weapon to support will not happen. While modern anti-ship weapons are not lethal enough to instantly kill all of the carriers, they are more than deadly enough to render the conventional amphibious operation impossible. Instead, the Marines have spent the last 30 or so years focusing on how to land where the enemy isn’t, which means that any future fire support system will need much longer range than the 16” gun has with the existing ammo. So any reactivation will need to develop new ammunition, and we have to look at the alternatives.
    The most obvious alternative is the railgun, but that’s still some years away from service. The Zumwalt and the 155mm AGS was supposed to replace the Iowas, but that turned into one of the worst-run defense programs in US history. A better alternative is probably to create a vertical-launched version of the Small-Diameter Bomb, which is quite cheap and suitable for close support of ground troops. There is a surplus of VLS cells in the fleet today, so these can be carried by all existing destroyers and cruisers without impacting other roles.
    There are also practical problems with reactivating the Iowas. I haven’t found a good figure for the cost of the reactivation, it would certainly be above $1 billion per ship, and probably significantly more. Many of the original systems in use on the ship were still in service on other vessels in the 80s, but the vessels in question have long since been retired. For instance, the propulsion plant of the canceled USS Kentucky, BB-65, was split up and used on two high-speed replenishment ships, USS Sacramento and USS Camden. They provided the expertise to reactivate the plants in the 80s, but were retired over 10 years ago. Likewise, mechanical fire-control systems were still in use on cruisers up until the late 70s, providing the cadre for the engineering crews on the Iowas. The 5”/38 caliber guns were still in use in the 80s, but left US service in the early 90s. New Jersey herself had been reactivated for Vietnam, and had only been in mothballs for 13 years before she was recommissioned. All of the supporting infrastructure, stored projectiles, powder, spare gun barrels, and the like, has been scrapped over the past 10 years. And all of this ignores the difficulty of environmental remediation, and bringing the electronics fit up to modern standards. And the manning issues will remain as they were in the early 90s. By this point, only Congress thinks that bringing them back might happen, and only a few crazy people think that it is a good idea.
    (And before you ask, if reactivation is seriously proposed, I will fight it as hard as I can, because it’s a terrible idea. If it happens anyway, my resume will be the first one on the desk of whoever is in charge of making it happen.)

    • orihara says:

      As a note, the ground launching of Small Diameter Bombs has already been done. They’re using old MLRS motors. Boosts it up to altitude, and then tells the bomb that it’s been dropped and to do it’s thing. They can even do neat things like having it go around an inconvenient mountain.

      The tricky part is getting it compatible with the Mk41 and approved for USN use. So far, it’s just been test shots from a MLRS launcher, under a manufacturer funded test series.

      • bean says:

        This is true, and I’m not sure why I didn’t mention that. There would probably be a bit of weapons safety work that would need to be done (the navy is a lot more paranoid than the other branches about making sure its munitions are safe because, unlike on land, running away is not a sound strategy), but I have seen speculation on this approach in print.

        • John Schilling says:

          Well, except for the part where they allows missiles to use DOT class 1.1 explosives (if catch fire, can detonate TNT-style) as fuel, whereas the Air Force insists on class 1.3(if catch fire, burns real good because duh rocket fuel). I suppose the logic is that if an SLBM tube or VLS array catches fire you’re going to lose the ship and crew even if it’s “just” fifty tons of high-intensity fire, and the rule since Jutland is that we make damn sure the magazines don’t catch fire. And that much, at least, the Navy is still properly paranoid about so, yeah, has to be requalified against e.g. stray microwaves coupling into the igniters.

          Also, if we’re going this route I think we can fit SDBs on top of Sea Sparrow motors and put them in the Mark 48 VLS or quad-pack in the Mark 41. May be important if LCS 2.0 winds up with Mark 48 (plausible) and gets tasked with part of the shore bombardment mission (ditto).

          • bean says:

            Interesting. I did not know that the navy lets 1.1 onto the ships, while the AF insists on 1.3. Do you know if this applies to airborne weapons as well? Most of my knowledge of this comes from Ignition, and all the hassles the navy made them go through looking for better propellants.
            I did know about the non-fire-resistant pits on the W88, as opposed to other warheads.
            I think my source on VLS SDBs looked at twin-packing. I’m not sure about quads. It would be helpful to have Mk 48 capability (no, not the torpedo), as that expands the number of platforms you can hang it on. I’d be more interested in having it on LCS 2.0 just to have a medium-range land attack missile than for amphibious fire support specifically.

          • Protagoras says:

            I had gotten the impression that unused fuel was an intended supplement to the damage-causing power of anti-ship missiles, which would mean the fuel would need to be volatile enough to perform that function.

          • bean says:

            Unburned fuel is very important to ASMs, but not so much to SLBMs or SAMs (and yes, I know that SAMs have largely taken over the ASM role, but I doubt that influenced fuel standards many years ago). For that matter, all US ship-launched ASMs have run on jet fuel.

          • John Schilling says:

            Not sure about the rules for air-launched weapons, but at this point I think every air-launched missile with a large solid motor is a joint service weapon and I think they are all class 1.3. Now I want to look this up.

  4. keranih says:

    I wonder if people could identify members of what they considered their outgroup who are…emmmm…sound-thinking people of good heart who should get more attention. (Even if you think everyone already is listening to them.)

    I’m going to go with Chris Arnade who is an Ivy grad Wall Street rich guy. And Jon Haidt – progressive humanities proff in the Ivies. And Paul Farmer – another self-righteous over educated progressive New Englander.

    (I am debating recommending in-group thinkers as well, but I’d rather hear from other people.)

  5. Wrong Species says:

    How would we go about giving an AI an IQ test? Apparently, it scores similarly to a four year old but it’s obviously inferior to a four year old in many ways:

    What’s more, many of the wrong answers are entirely unlike those that children would give. For example, in the word reasoning category, ConceptNet 4 was given the following clues: “This animal has a mane if it is male,” “this is an animal that lives in Africa,” and “this a big yellowish-brown cat.”

    But its top five answers were: dog, farm, creature, home, and cat.

    So what would be a better way to measure intelligence in AI in comparison to humans?

    • Jiro says:

      A big yellowish-brown cat is a cat, by definition.

    • rlms says:

      That is a perfectly good way of measuring one facet of intelligence: understanding and reasoning about natural language. In humans, that facet is very tightly correlated with other skills like pattern matching and arithmetic, and we call skill in all areas intelligence. The correlation breaks down occasionally (so we talk about verbal/spatial/mathematical intelligence) but usually it’s strong enough for “intelligence” to be a coherent term. In terms of the field of current AI, the correlation is much weaker: AIs can pattern match and play chess very well, but not reason verbally. Until this changes, I don’t think it makes sense to talk about the IQ of an AI.

      • Wrong Species says:

        An AI could be good at pattern matching, arithmetic and reasoning verbally and yet still be less intelligent than a four year old. Like you said, different facets of intelligence are correlated in humans but not necessarily so in machines. When does a collection of specific intelligences cross the line in to general intelligence and how would we know if it did?

  6. hlynkacg says:

    At the risk of doxing myself, (and getting banned for making tasteless jokes) I once gave Dan Rather a helicopter ride and as a result of this comment in Neutral vs. Conservative: The Eternal Struggle have discovered that I appear multiple times in the first page of a google image search for “Dan Rather” + Sumatra, but not at all in a search for my own name.

    Has anyone else here had a similar brush with celebrity?

    • keranih says:

      …I won’t tell my stories, but I will tell a friend’s –

      She was staying in the same hotel as Bill Clinton and Al Gore back when they were running for the WH the first time. She was an avid runner at the time, and a couple times met them on the way out to a run. One morning, rather early, she actually rode down the elevator with Gore & his SS guy, all three of them in running shoes, etc. Noting that Mr Gore was missing his usual jogging partner, my friend said – to break that awkward elevator silence – asked, “So, you’re running without Bill Clinton now?”

      She said Gore looked extremely confused and nearly panicked. The SS guy caught on first, hid a snicker, and my friend finally figured out that Gore thought she had meant running for office, this year instead of going jogging, this morning?

      Pre-caffine mornings, man, they really suck.

    • dodrian says:

      I went to the Edinburgh Fringe festival a few years ago and was walking late at night with a friend when I saw someone coming towards us who looked very familiar. Staring, because I couldn’t quite place who it was he looked up, noticed me staring and gave a quick nod before hurriedly pressing on. As he passed us it clicked, and I turned to my friend whispering a bit too loudly: “That was Al Murray!”

      It was only half a minute later when I realised that it wasn’t Al Murray who had nodded at me, but was in fact Dara Ó Briain. I really hope he didn’t hear me!

    • Conrad Honcho says:

      I was in New Orleans for the 1996 Sugar Bowl, drunk in the French Quarter when out of a restaurant walks Nature Boy Ric Flair. I said “Mr. Flair!” and shook his hand.

      Never washed this hand since.

    • HeelBearCub says:

      I’m not sure what you are looking for, really. People who can have their brushes with celebrity validated by a google search?

      But, since we are just talking about brushes: I played pickup basketball at the local Y against Phillip Seymour Hoffman when he was in town filming “Patch Adams”. I knew who he was, and really admired the work he had done up to that point, but I have an internal tick that says “treat celebrities like you would anyone else” so I didn’t even acknowledge that I knew who he was.

      I’ve often wondered whether he would have rather been pointed out.

    • Back when I was a graduate student, I was on a ship crossing, I think, from Greece to Italy. There was a man there with a very attractive wife and, I think, some kids. We got into a conversation. At some point I asked him his name, and he said “Vladimir Ashkenazi.” I’m not very interested in music but I thought I recognized it, so I asked him if he was famous and he said “moderately” or something similar.

      Then there was the time, in a Moroccan restaurant in the L.A. area, when I noticed a diner at an adjoining table with extraordinarily long legs. It turned out that he was Wilt Chamberlain.

    • The Nybbler says:

      Random brushes? I happened to pass Yao Ming at LAX once. My name appears on a document mentioning Moldbug (OK, perhaps not a real celebrity, but should count for SSC purposes) linked from his Wikipedia page. I’ve eaten a few tables away from Terry Hogan, but I think that’s not unusual (I’ve run into others who have also; the man likes to eat I guess).

    • LHN says:

      My parents were visiting town, and we went out for Sunday brunch with them at a downtown restaurant. Around the middle of the meal, my mom (whose extraversion and gregariousness I unfortunately didn’t inherit) started interacting with a baby at the next table, then making light conversation with the parents.

      I smiled at the baby for a few seconds, then went back to my meal. I didn’t really register the couple at all.

      As we were leaving the restaurant, my wife turned to me and said, “Wasn’t that George Lucas?” Answer (confirmed with a quick Google Image search to show him with his wife and new child): yes, yes it was.

      As it happens, early in the brunch I’d had occasion to go off on a minor rant about the Star Wars prequels. Fortunately (since whatever I think of the movies, I’m sure he’s heard way too much on the subject and doesn’t deserve to have his meal ruined) that was over before they’d arrived.

    • BBA says:

      I once sat on an airplane next to Dee Snider. I wasn’t quite sure who he was, but figured that anyone who looked like that and was flying transatlantic first class had to be a rock star.[*] At one point I caught a glimpse of him reading emails on his tablet, which confirmed his identity.

      I didn’t say anything, because I’m not really into hair metal.

      [*] Or was getting a heavy discount on the seat, like I did.

    • Spookykou says:

      I am not sure if this counts.

      While working at an upscale grocery store I cashiered for Sandra Bullock and Andy Roddick (not at the same time). Nothing interesting happened.

  7. Controls Freak says:

    I found a role that MealSquares could play in my diet, so I decided to give them a shot. Just got my order this week. I found them quite palatable. Like a cross between a sweet bread and a breakfast cookie. It’s not too sweet, but that works for me, because I don’t eat many sweet things in general, so I find it rather pleasant instead of bland. It’s got the kind of somewhat dry mouth feel that is similar to a lot of gluten-free breads I’ve had (celiac in the family). I don’t find it too off-putting (probably because of that history), but it’s pretty manageable if it’s eaten alongside a drink.

    Overall, I’m pretty pleased, but I can see how my particular situation mitigates some of the downsides which might turn off others.

    • sandoratthezoo says:

      Me too! They hit a sweet spot for me of “good tasting enough that I don’t feel repulsed by eating them,” but “not good tasting enough that I’m tempted to eat past satiety.”

    • BBA says:

      If they could make them without the sunflower seeds (or maybe grind them up?) I’d be an enthusiastic convert. As it is, they’re just okay to me. It’s a texture thing.

  8. Wrong Species says:

    When it comes to machine learning, when the AI comes up with an answer to a query, is the process that it uses to come to that decision understandable in theory and simply too complex to explain to humans or is it something much more opaque?

    • MNH says:

      Your question is a little underspecific. “Machine learning” is a big broad term that is often defined in ways that include methods as simple as linear regression. I’m going to guess that you’re referring to learning using neural networks in particular?

      Could you also clarify what does and doesn’t qualify as “understandable in theory”? A human could certainly replicate the network’s decision process by hand, but I really doubt that most could be reduced to anything compact and intuitive. I’ve trained a few NNs but I’m no expert, though, so maybe an actual professional would know better.

      What’s your math background like? If you’ve ever learned linear algebra, Andrew Ng’s Coursera course on machine learning is extremely easy and highly informative and you only have to do the first half (or if you’ve learned about linear and logistic regression before, only the second fourth of it) to learn the basics of what neural networks are and how they work. (lin alg is not a prereq to the course but it will probably actually involve difficulty if you are not familiar)

      • Wrong Species says:

        Same question that I asked Ian:

        So we have two choices X or Y. We use the neural networks to pick one. And then to check it we get someone like Terrence Tao to check the results. He does the math and doesn’t find anything wrong. Would Tao say that the machine was right but he simply doesn’t have an intuitive understanding of why? Or would he say that the math was right but he still doesn’t know if it’s the right decision?

    • Iain says:

      It depends what you mean by understandable.

      To massively oversimplify, machine learning works by creating a model from an initial training set, then querying that model with new data. There are many different kinds of possible models, but roughly speaking they all involve defining a function from your input data to the set of possible answers, and then tuning the weights in that function to get the best fit with the training data. A couple of examples:

      Support vector machines
      try to draw the best possible line between two categories. (Technically, since you are typically working with high-dimensional data, they try to draw hyperplanes.) Not all divisions are going to be nice and linear (they are not linearly separable), but you can do some clever pre-processing of the data (using a kernel function) to try to map it into a different space where your data behaves better. Training a support vector machine means finding the best kernel function and hyperplane. To query a trained SVM, you feed your data point into the kernel function, and check to see which side of the hyperplane it’s on.

      Neural networks
      create a bunch of abstract “neurons”, typically arranged in layers. The neurons in the first layer are linear combinations of your input data: if you have input values x, y, and z, then one neuron might be 0.1x + 0.4y + 0.2z, one neuron might be 0.2x – 0.6y + 0.4z, and one neuron might be 0.3x – 0.1y (with an implicit +0z). The output is typically clamped to (0,1) using something like the sigmoid function. The second layer of neurons are the same, except they use the first layer as their input data. This is repeated for some number of layers, then you have a set of output neurons that read the final layer to give you an answer. Once you fix the structure of your neural network (how many layers, how many neurons per layer, and so on), then training the network is just a matter of finding the best weights for each of your neurons.

      These mechanisms are not at all mysterious. Given the model, you could sit down and crunch through the math by hand, although it would take a long time and I would not recommend it. The process of answering a query is not complicated. Furthermore, the process of training the model is also relatively easy to understand: all you’re doing is minimizing the error of your model on the training data.

      But that’s all math. There is no guarantee that you can come up with a high-level explanation of why the weights ended up the way they did. It’s entirely possible that the machine learning algorithm picked up on an unintended distinction in your training set: here’s one example, in which a neural network to identify camouflaged tanks instead noticed that all the pictures of real tanks had been taken on a cloudy day, and learned to identify the weather.

      There are some techniques that you can use to try to peek under the covers. For example, you can try running a neural network in reverse; instead of asking “is this a picture of a dog?”, asking “what tiny changes would be most effective at making this picture look more like a dog?” This is how all those Deep Dream images that I’m sure you’ve seen floating around were created. Similarly, there are techniques for figuring out which parts of your input are most important to the final answer, by systematically modifying them to see how much it affects the result. To the best of my knowledge, though, this is still an active area of research.

      In short: it is trivial to understand the basic low-level functioning of a machine learning algorithm, but much less straightforward to map its results onto high-level concepts.

      • Wrong Species says:

        Let me be a little more specific:

        So we have two choices X or Y. We use the neural networks to pick one. And then to check it we get someone like Terrence Tao to check the results. He does the math and doesn’t find anything wrong. Would Tao say that the machine was right but he simply doesn’t have an intuitive understanding of why? Or would he say that the math was right but he still doesn’t know if it’s the right decision?

        • Iain says:

          The latter.

          You don’t need a brilliant mathematician to redo the calculations. (You could more or less do it yourself. It would just be tedious.) Getting the math wrong is not a significant source of error for machine learning.

          Machine learning algorithms, like all other algorithms, can only work with what you give them. The math behind machine learning is easy; finding the right training data is hard.

          Think of it this way. Training a neural network is like walking up to it with a wheelbarrow full of examples and saying “I want you to figure out the best possible generalization of these examples that can be expressed using a neural network with X neurons and Y layers.” To simplify a bit, there are three ways that this can go wrong. In increasing order of importance:

          1. The neural network gets stuck in a local maximum. You can think of this as a polished version of a strategy that is not the best possible strategy. There are some problem domains in which this is not an issue; in cases when it is an issue, there are techniques that can help avoid it.

          2. The neural network starts overfitting. You can think of this as just memorizing the input set, and focusing too much on that. We also have a bunch of techniques for dealing with overfitting.

          3. Sometimes your training data is bad. This is what happened in the tank example: the data from which the model learned (sunny days without tanks, and cloudy days with tanks) was not a good match for the real world, in which tanks do not vanish when the sun comes out. In other words, sometimes you train a machine learning algorithm to do an excellent job on the wrong problem. This is not the sort of issue you can reliably catch just by checking the math. You need an understanding of the problem domain.

          • roystgnr says:

            Wait, how does gradient descent prevent you from getting stuck in local maxima? The gradient at a local maximum is either zero or (at least in the Fréchet sense) undefined, no?

          • Wrong Species says:

            I think I get it. Essentially, you can check the output but there is no way a priori to know if your input is any good. Garbage in, garbage out.

            So that’s mean that if we ever have neural networks running something important we would be screwed right? Since we can’t check our result to make sure everything is right we’ll never know if it’s making a decision at a higher or lower level than we would.

          • Marshayne Lonehand says:

            The AI heuristic that “neurons don’t convey meaning” implies that new neurons, and new layers of neurons, can be added to a net, with a view toward eliminating suboptimal local minima, without any preconceptions in respect to new meanings that the new neurons and new layers will convey.

            Indeed, within the “New AI” paradigms, questions of “meaning” commonly don’t arise at all!

            What’s transformative for the pace and span of AI development is the emerging appreciation that “neurons needn’t convey meaning” carries with it tremendous advances in the celerity and efficacy of AI development — because instead of improving a neural net by the tedious distillation of more facts linked by deeper Bayesian logic, it suffices instead simply to allocate more neurons that are organized in more layers.

            The lack of ratiocinative rigor in this otherwise satisfyingly effective heuristic is deeply disconcerting to many AI researchers.

            With an explicit view toward remediating this perceived deficiency in AI theory, Boaz Barak and David Steurer have created an on-line course (which is marvelous, as it seems to me) which is available as the free-as-in-freedom textbook “Proofs, beliefs, and algorithms through the lens of sum-of-squares” (SOS).

            Aye, SSC laddies and lassies, courses like SOS are a cutting-edge of ratiocinative cognitive science! 🙂

          • lycotic says:

            Wait, how does gradient descent prevent you from getting stuck in local maxima?

            It doesn’t. What’s interesting is that, for sufficiently interesting networks, it doesn’t appear to matter, usually.

            See The Loss Surfaces of Multilayer Networks.

          • Iain says:

            @roystgnr: Sorry, I was being lazy with that link. I intended to imply: we train neural networks using gradient descent; we have spent a long time using gradient descent and coming up with techniques to help us out of local maxima; those techniques are applicable to training neural networks. I’m thinking of things like simulated annealing, which in retrospect is not quite the same as gradient descent, so maybe it is better if you just pretend I never included that link at all. Point is, it’s a problem, but a tractable one, and can often be handled by just throwing more math at it.

            @Wrong Species:
            As I said in my first reply, there are ways that we can extract information from neural networks to give us a slightly better idea of what sorts of things their decisions are based on. It’s not an insoluble problem, but it is the sort of problem that (for now at least) requires careful human judgment.

        • rlms says:

          It’s not that kind of maths. An analogy: imagine you have a primitive computer that approximately multiplies big numbers using a complicated algorithm where each step is pretty trivial. You don’t know the degree of approximation, but usually it’s pretty good. Suppose further that there aren’t any non-primitive computers that can do the calculations accurately. You give the computer some inputs, and it spits out an answer. You can check the the maths in a sense, by going through the algorithm yourself (anyone can do that, it’s just calculation). That allows you to confirm that the computer has followed the algorithm correctly, but it doesn’t tell you how accurate the answer is. In machine learning, as well as the answer being inaccurate there is also the possibility that you are asking the wrong question anyway.

        • Marshayne Lonehand says:

          A more realistic “X”-scenario is this one.

          Terry Tao has two students, Alice and Bob, and a grader, Eve. A homework assignment requires that Alice and Bob prove statement “X”.

          • Alice’s intuition conceives an algebraic proof of “X”.
          • Bob’s intuition conceives a geometric proof of “X”.
          • By a purely ratiocinative process, the grader Eve verifies both proofs.
          • In a subsequent lecture, Professor Tao surveys further applications of both kinds of mathematical intuition, with particular attention to (unproved) postulates “Y” that might serve to unify Alice’s algebraic intuitions with Bob’s geometric intuitions.

          Note that only the grader Eve can rely solely upon ratiocination — everyone else’s mathematical processes are guided (or at least initiated) chiefly by intuition — which is where notions like “beauty” enter into the discourse.

          Here all four parties are collectively and cumulatively improving their cognitive capacity — the somewhat mysterious, somewhat controversial, much-discussed, largely non-ratiocinative, yet amazingly powerful, and professionally crucial, cognitive capacity — that is called “mathematical maturity.”

          Does there exist an analogous professional capacity in psychiatric medical practice, that might be called “psychiatric maturity”? — a cognitive capacity that might synthesize, for example, (synapse-altering) pharmaceutic psychotherapies with (connectome-altering) behavioral psychotherapies?

          These parallels between mathematical practice and psychiatric practice — which are entirely natural (as it seems to me) — are why an SSC essay comparing and contrasting “mathematical maturity” with “psychiatric maturity” would be read with interest by many (definitely including me).

    • Marshayne Lonehand says:

      An ultra-short answer — which overlaps compatibly with Iain’s answer immediately above — is “deep learning yields occult understandings.”

      A slightly longer and largely equivalent answer is that the USMC maxim “Fake it until you feel it” works remarkably well as the basis of deep learning algorithms.

      An unforeseen (by pretty much everyone except maybe Marvin Minsky) practical consequence has been that 21st century “superintelligent” programs (like AlphaGo) can feel “what’s right” — right Go-moves, right language-translations, right lane-changes, and right stock-trades — while having no capacity to explain “what’s right.”

      For nuts-and-bolts narratives, some recommended readings are:

      • Wikipedia’s “Backpropagation

      • Matt Mazur’s hacker-friendly on-line essay “A Step-by-Step Backpropagation Example

      • Chapter 2 “How the backpropagation algorithm works” of Michael Nielsen’s new on-line textbook Neural Networks and Deep Learning — Mike Nielsen was coauthor with Isaac Chiang of the celebrated (among physicists and computer scientists) “Mike and Ike” textbook Quantum Computation and Quantum Information (2000).

      This SSC comment provides some futurological projections in regard to the AI / psychiatric / Ted Chiang-class implications of the 21stf century’s new deep learning / zero ratiocination algorithms.

      That machines evolve human-level cognition via human-type processes of “march until you dance” (in psychotherapeutic CBT/DBT vernacular) — and that AIs generically exhibit, in consequence, a human-like incapacity to explain their own cognition — is immensely reassuring to many people … including me. 🙂

      • rlms says:

        Sadly, reliably detecting John Sidles remains beyond the reach of deep learning (for the moment at least).

      • MNH says:

        Do you have any advice for if I want to learn the nuts and bolts of something other than backpropagation?

        Am I responsible for your narrow choice of recommended readings since I recommended supplementary backprop reading for the Andrew Ng course? Or is it just a coincidence?

        Is the process by which John Sidles decides to ask questions and embed links understandable in theory?

        • Marshayne Lonehand says:

          For philosophical nuts-and-bolts, it’s tough to better Mavin Minsky’s well-respected AI classic Society of Mind (1986), read in conjunction with the concluding essay “Appendix: Is there a simple algorithm for intelligence?” of Michael Nielsen’s (above-referenced) Neural Networks and Deep Learning .

          For computational nuts-and-bolts, it is necessary to build upon a foundational understanding of undergraduate-level classics like Numerical Recipes (by Press et al.) and Matrix Computations (by Golub and van Loan).

          Then a computation-centric, abstraction-heavy, yet applications-oriented journey into “Yellow Book” mathematics might start with Ideals, Varieties, and Algorithms (by Cox, Little, and O’Shea), Tensors: Geometry and Applications (by Landsberg), Introduction to Smooth Manifolds (by Lee), Cartan for Beginners (by Ivey and Landsberg), and Invitation to Algebraic Geometry (by Smith).

          To gain a mature appreciation of the essential mathematical unity of these texts, it’s sound pedagogic orthopraxy to read the “Preface” and/or “Introduction” of all of the above-mentioned texts, prior to tackling the body of any one of them.

          For a bridge back to philosophy, some computation-oriented students will enjoy too Conceptual Mathematics: A First Introduction to Categories (by Lawvere and Schanuel) … William Lawvere being the most-celebrated PhD student of none other than the rigor-loving Clifford Truesdell! 🙂

          Caveat: these books originate in disparate 20th century mathematical communities, and consequently, they don’t much reference one-another — so it’s up to 21st century students to say (among themselves) “Hey, these folks are all talking about the same things”. In consequence of this essential unity, acquiring a working understanding of any one text, helps greatly with understanding the others … fortunately for everyone. 🙂

          Lawvere and Schanuel’s above-cited Conceptual Mathematics explains how the mathematical traits of “naturality” and “universality” — which are formalizable attributes that are becoming so central and useful as to play a central role even in student-level textbooks — are inculcated in practice.

          A sobering-yet-awesome feature of the modern STEAM-enterprise, is that for each STEAM-text named in the preceding paragraphs, there exist a dozen-or-more texts that are comparably excellent. Yikes! 🙂

          • Marshayne Lonehand says:

            PS  Also on the list of undergraduate-level pre-“Yellow Books” should have been stochastics-oriented texts, for example Understanding Molecular Simulation: from Algorithms to Applications (by Frenkel and Smit) and Handbook of Stochastic Methods (by Gardiner).

            Existing undergraduate texts in quantum physics — including in particular The Feynman Lectures — regrettably are poorly matched to the Yellow Book literature. It’s fun to speculate that this mismatch helped Ed Witten to prosper as a quantum theoretician — Witten took no undergraduate physics courses, and in consequence, his physics education began with Yellow Books! 😉

            Consciously seeking to unite “practical” cognition (as inculcated by the aforesaid undergraduate texts) with “unversal” and “natural” cognition (as instilled by the aforesaid Yellow Books) provides young people with no bad start (as it seems to me) towards a creative, mathematically mature career in multiple STEAM-disciplines and enterprises.

            If the systematic cultivation of mathematical cognition that is practical, universal, and natural sounds a lot like psychiatric CBT/DBT orthopraxy … well … at higher levels modern mathematical practices *do* overlap considerably with modern psychotherapeutic practices.

            Come to think, shared orthopraxy between mathematics and psychotherapy isn’t all that surprising (is it)?

            Concretely, mathematical pedagogy and psychiatric therapy alike seek to instill and extend harmoniously cognitive compatible capacities in respect to the inculcative “marching” that is ratiocination and the instilled “dancing” that is intuition and empathy, isn’t that evident? 🙂

            Even in respect to pharmaceutical practices — coffee (and Adderall) for mathematicians and their students, Prozac (and Adderall) for psychiatrists and their clients — the parallels are striking … and to no small degree, even discomfiting, aren’t they?

            No one expects that “mathematical maturity” can be instilled solely by instructing students to drink coffee and/or take Adderall … so how plausible is it, that pharmaceutic interventions alone should reliably instill “psychiatric maturity”?

            The world wonders — wonders at these tough, naturally overlapping challenges and open questions of mathematical and psychiatric orthopraxy! 🙂

          • Marshayne Lonehand says:

            PPS  To encapsulate above appreciations within an even smaller compass, try going to Michael Nielsen’s website Cognitive Medium and asking:

            How would Michael Nielsen’s essays change, if they were not about AI and deep learning, but rather about human cognition and psychotherapy?

            The most natural, beautiful, wonderful, and awesome answer (as it seems to me) is set forth in Wallace McRae’s cowboy poem “Reincarnation”, whose concluding line is “You ain’t changed all that much!” 🙂

          • Incurian says:

            It’s stuck in a loop.

          • Marshayne Lonehand says:

            Assuming, Incurian, that your name is nominatively indeterminative, you might find that your cognitive capacities are expanded by William Thurston’s friendly and accessible Foreword to John Hubbard’s friendly-and-inaccessible text Teichmüller Theory and Applications to Geometry, Topology, and Dynamics, Volume I: Teichmüller Theory (2006).

            Passionate engagement is contagious. It shows through in his [Hubbard’s] writing.

            This book develops a rich and interesting, interconnected body of mathematics that is also connected to many outside subjects.

            Mathematics is a paradoxical, elusive subject, with the habit of appearing clear and straightforward, then zooming away and leaving us stranded in a blank haze.

            Why?

            It is easy to forget that mathematics is primarily a tool for human thought. Mathematical thought is far better defined and far more logical than everyday thought, and people can be fooled into thinking of mathematics as logical, formal, symbolic reasoning.

            But this is far from reality. Logic, formalization, and symbols can be very powerful tools for humans to use, but we are actually very poor at purely formal reasoning; computers are far better at formal computation and formal reasoning, but humans are far better mathematicians.

            The most important thing about mathematics is how it resides in the human brain.

            Mathematics is not something we sense directly: it lives in our imagination and we sense it only indirectly. The choices of how it flows in our brains are not standard and automatic, and can be very sensitive to cues and context.

            Our minds depend on many interconnected special-purpose but powerful modules. We allocate everyday tasks to these various modules instinctively and subconsciously.

            Hmmm … Thurston talks like pretty much any science-grounded rationality-minded skills-trained CBT/DBT psychotherapist, doesn’t he?

            Thurston has written Forewords to many celebrated mathematical texts; these Forewords are heartily commended to SSC’s psychoanalytic and AI-minded readers … Daina Taimina’s Crocheting Adventures with Hyperbolic Planes (2009) includes a good Thurston Foreword.

            Does the 21st century’s Consilient Enlightenment pursue Thurston-class ideas with unrelenting passion? The plain answer is “yes”

            Hopefully, Thurston’s essays will help the SSC’s SF/Fantasy fans to appreciate that’s there’s no shortage of interesting company on (what might be called) “the 21st century’s hell-bound progressive train” … a train of thought and dreams that a diverse crowd of folks (me included) — “the drunks and the sinners, the gambling folks and the grifters, the big-time spenders and the skirt-chasers, and all the jolly crew” — so  greatly enjoy riding, on a never-ending journey! 🙂

  9. axiomsofdominion says:

    So I’ve been working on a fantasy world/politics simulator for a couple years and I’m currently doing the espionage/surveying/information gathering interface. I was curious if people would be willing to operate an intelligence network with a complexity similar to the combat/conquest aspect of modern strategy games. I compare it to that since that dominates most 4x, grand strategy, and general turn based games.

    • Incurian says:

      Depends, how would it work?

      Edit: Also, when you say “intelligence network,” what kinds of intelligence did you have in mind?

      • axiomsofdominion says:

        In the current implementation you are deploying funds and units to various tasks. Scouting/surveying/reporting on province data like buildings built, troop movements, goods transport info, populations, and other stuff. Keeping track of the ideology of the populations. Tailing characters, that is relevant enough persons to be represented separate from populations. Gathering intel on those characters, maybe killing them or learning their secrets. When you unapply resources to a location, population, army, character, or opposing intelligence unit you lose detail over time so your assassination chance drops, you no longer can see the buildings or populations in a province, etc.

        The theme/genre/setting is fantasy.

        There is a detailed logistics and economic system, very complex magic, an ideology system for the political stuff and a propaganda system that affects ideologies as well as opinions that various characters and populations have of each other.

        You have intelligence units as well as equipment and then you assign funding. There’s no system for not “burning sources” as another poster mentioned. More like counter intelligence can get a general idea that someone is doing stuff and then investigate. But it wont be like, oh the king got caught with his mistress must have been that chamber maid he molested since she has a reason and access to pass the info on.

        There may be a double agent or unreliable source thing going on though. Depends on how much extra work that is. Setting up an interface for the player wouldn’t be too hard but making the AI understand it could be a pain.

        • Civilis says:

          I’m interested in this from a game design perspective, because I’m always a little annoyed by the espionage components of 4X style games. What follows is a slightly filtered stream of thought brainstorm.

          Since this is a game, you need a goal. It sounds like you’re eventually going to work this in as a component of a 4X game, where it will inherit the objectives of the larger game. Since this is just a foreign espionage sim, I’d imagine you get issued orders you have to fill that serve as a proxy for what you’re going to do with the espionage in the larger 4X game. For example, (a) “find out the military strength of X’s army in these border regions”, (b) “kill the governor of territory Y and spark a revolt”, and (c) “acquire magic researcher Z for our kingdom”. I’d probably want to rate the objectives with some form of Victory Points, perhaps influenced by the difficulty of the objective, the time taken to complete them, and the thoroughness of the way they were completed. Victory in the game is then rated by VPs accomplished, something like ‘get X VPs’ or ‘have the most VPs when time expires’.

          Since it’s the most complex, let’s look at mission (c), because it’s also seems like the most fun. First, you have to find the target and what protection he has. You send agents with ‘gather information’ skills to do their thing in territories most likely to have a magic researcher until you find him, and then you concentrate on that territory until you think you know enough to actually get him. This is going to be a balancing act; the more you know, the easier it is, but the more people you have the higher chance their activity will be detected, increasing the security level.

          Once you think you have enough info, you make a choice. Option 1: take your agent(s) with the best ‘direct action’ or ‘black ops’ skills, give them all the magic combat items, and tell them to grab the guy. This is quick (more VPs) but also not subtle (less VPs). Option 2: send in your subtle agents, undermine and weaken the local security forces and subvert the researcher (perhaps some characters have the ‘honey trap’ skill), then fake his death (‘magic potion of feign death’ item) and spirit him across the border (either requiring you to know the security of the route to the border or perhaps have the rare ‘teleport stone’ item). This is not quick (less VPs) but more subtle (more VPs).

          Working what happens like this tells us what information we need to have for the game. We need a map of discreet territories. For each of those territories, we need some values, most likely eventually determined by the larger 4X game (economic info, military info, etc.). We also have some values determined by the espionage component itself: security level, spies in territory, etc. And those values need to have some values themselves, most importantly, ‘Does the player know this information?’

          Since it sounds like you want to put this in a larger 4X game, you may want to work out the counter-intelligence system at the same time. It may be as simple as assigning resources and posture to each defensive territory (security forces, money, and magic). Still, you might want to tie it in to the game more. Perhaps each territory has a ‘magic research rate’ which generates the amount you can spend on magic items to equip your agents. You can assign your magic researchers to the high rate territories, but that makes them easier to find for enemy spies to assassinate or kidnap.

          • axiomsofdominion says:

            You can look at some of the others aspects of the game here:
            http://axioms-of-dominion.wikia.com/wiki/Axioms_Of_Dominion_Wiki

            Different articles have more or less detail depending on how far along implementation is and also how much they diverge from standard 4x stuff that doesn’t need much explanation.

            You may want to check the sections for Ideology, Secrets, Research, Magic, Espionage, Propaganda, and Opinion. Probably Conspiracies as well.

          • ADifferentAnonymous says:

            Skimming the front page, I got surprising far in the page list thinking it was a sentence.

            “Huh, ideology integration buildings, interesting idea… But that seems like an odd place to house the climate religion secrets, shouldn’t those just be in the climate temples? Oh, this text is blue for a reason.”

          • Civilis says:

            Looking over the wiki, one of the things I’m wondering about which has a direct bearing here is the scale of the game and the level of abstraction.

            In Civilization 6, the most recent 4X game I played, a spy unit costs 300 production, about as much as a unit of cavalry or artillery of roughly the Napoleonic Era, or just under half the cost of a modern armor unit. Given the one unit per hex limit and the map scale, a unit seems to approximate roughly a corps-sized formation. Each spy unit is named, which gives it the illusion of being a single agent, but it seems more likely to represent some number of agents and the necessary support, management and analysis infrastructure.

            Your espionage game seems to be at a scale where you’re micromanaging individual agents, but it’s really hard to run the more conventional 4X military at that scale, and in a realistic world, it’s hard to manage both scales at the same time. If I’m limited to buying military force by the brigade, the cost of a spy and gear is going to get lost in the change. Now you have a fantasy world, so I could accept a system where it costs as much to hire a level 20 adventurer and buy him a stack of +5 quality gear as it does to hire a brigade of level 1 mercenaries with mundane gear. The biggest issue with scale at that point becomes how the two scales interact, if they can interact. If my level 20 adventurer can defeat an endless wave of mercenaries, then I’m going to be hiring adventurers and not mercenaries.

            The other problem with the difference in scales is the asymmetry between spy and counter-spy. If my options as the spying player are ‘pick a territory and pick a type of target (eg, subvert morale, assassinate governor, steal research, sabotage production, etc.)’, I’m going to be annoyed if my only option as the defending player amounts to ‘how much do you want to spend on general counter-espionage?’. If I suspect the spy is after my research, I better have an option to focus my counter-espionage on defending my research.

          • ADifferentAnonymous says:

            I don’t think the scale problem is so intractable–if spies are fungible not with units but with unit commanders, then balance is roughly achieved.

          • axiomsofdominion says:

            @Civilis

            As far as costs go a lot of that can be ameliorated by “funding” which I’m not sure is described in detail on the current wiki page. You can employ both characters, which are also merchants, mages, generals, and province owners, and “units” which are sort of the like the units in the “army” and cost a similar amount discounting gear. Spying is a bit more limited in “agency” than the military aspect. You can wardec anyone you want but maybe your opponent lacks any sort of significant secret you can use against him.

            An example I like to use for the more interesting side of secrets is this:
            A sort of tribal/warrior/steppe type of nation worships dragons. Their leader, in secret, has located a dragon, magically imprisoned it, leeched its attributes, and proclaimed himself as blessed by the dragons. Thus he has gained the authority to sort of unit the tribes and pursue larger enterprises. You, assuming you have the capability, are trying to figure out how to stop him and your spying turns up his secret. You then have a variety of options on how to proceed. There is a less detailed version of this story on the Secrets page of the wiki. You can blackmail him for favors, out him to cause chaos for various purposes, free his dragon, and other things.

            Counter espionage is indeed more detailed, once you located a threat. So if someone is tracking your guy and you find out, or even guess that its happening, you can apply resources to trick, catch, subvert, or kill the guy or w/e.

        • Trofim_Lysenko says:

          EDIT: wrote this and posted it before your Wiki link.

          A few suggestions:

          1) Top-Level Information Security/OSINT: trade and international commerce, since you have that economic/logistics system, should also allow for a baseline level of spying with minimal investment. For example, if you’ve spent $RESOURCE to build your first tier “Black Chamber”, then just engaging in normal commerce with Territory A should give you some basic details about Territory A’s military spending, resource distribution, leadership, their cultural/legal “settings” if the game has them, etc. You should be able to counteract this with a coarse counterespionage and information security setting, but it should cost…my first thought is slow down troop movement of military units, and decrease the profit from and frequency of trade. Security measures that make it harder to travel, that get merchants questioned or termporarily detained because they asked too many questions about the big ships being built in the harbor, etc will have a chilling effect on your economic activity.

          2) “Rules Of The Game”: Decide how you want to treat spying in this setting. Is it a gentlemanly pursuit in which agents are captured and exchanged in return for considerations, or is anyone caught spying turned over to the Inquisitors or Her Majesty’s Pain Technicians? How are you going to handle AI reaction to spying on allies, and how does spying/information sharing within and between alliances work? I think this is especially relevant if it’s your classic style of 4X game where the making and breaking of alliances and the timing of same can have a large impact on the course of play. “Your Spies Report: Nation A is courting Nation B, trying to woo them away from you” and “Your Spies Report: Nation B has accepted Nation A’s offer and will terminate their alliance in X Turns” could provide a decisive advantage.

          3) Once is Happenstance, Twice Is Coincidence…: In addition to stealing technological/magical research (speeding up your unit and building development time), espionage assets should be able to -delay- both research and development. But here’s the rub: If you want to make this a big feature, I strongly suggest that it be at least theoretically possible for this to be disguised from the AI and player alike. However, that would require you to implement something like a per-turn random event table that could throw stuff like “Your Stone Masons are Striking for Better Pay, all Building Construction paused for X turns unless you Spend Y Resource” at random, to allow a player or AI with espionage units to hide their sabotage, agitation, etc in amongst those events. That’s potentially a lot of work.

          • axiomsofdominion says:

            Different nations of different ideologies have different rules for the consequences of spying. Some might execute and some might not. You can get information about people trying to subvert allies. Especially if you spy on said allies as well as the relevant enemy. You can see their direct opinion of you and there is also a “secrets” system as described at the wiki link. The game is pre industrial fantasy, for the most part, so you wouldn’t have strikes. But some sort of accident or something would work in place of that. Research, depending on which form, already has some randomness in there. So it wouldn’t be too hard to hide interference of some sort. Counterespionage mostly costs direct money but there is an ideology related to borders and trade so if you closed borders or something you’d get less trade and more security. Its not a direct feature of the espionage system, though.

        • Incurian says:

          This sounds similar to the way the Total War series handles various “agents.” Have you looked into the way they did it? I didn’t hate Total War spies, but they were a little shallow.

          • axiomsofdominion says:

            Total War has individual agents without equipment. They have limited numbers of interactions all of which are incredibly simplified. They also have no funding, unless you count unit upkeep, and can’t coordinated or increase unit count or anything. They are generally just econ boosts/nerfs in unit form though they can also “slow army” or something. Depends on the game.

          • cassander says:

            @axiomsofdominion

            the latest total war game is a sort of masterpiece in how it makes the strategic level game looks extremely sophisticated while actually being incredibly shallow.

          • axiomsofdominion says:

            Can you go into detail for that? I’m pretty curious. I played that game a lot, often with diplo and other map stuff heavily modded. Steam says almost 200 hours which is a lot for me lately, since I played games less than in the past.

          • Spookykou says:

            I really enjoyed the agents in Shogun Total War I and II. I could be remembering this wrong, but you had to build out specific buildings to get access to agents/higher level agents and they all had passive and active abilities. They leveled up and had a talent tree allowing for specialized agents good at particular tasks. Minor, but they had names which I really enjoyed because I could really get attached to my master assassin who took out a Daiymo and his four sons. The total number of agents you controlled was proportional to agent buildings so if you wanted a massive ninja network you needed to build out a massive ninja infrastructure. The tactical importance of being able to stop armies was huge, and so are high ranking generals, the ability to assassinate top generals/control the map made a ninja focused strategy viable, and I won a few games that way.

            The only new Total War game I played was the Warhammer one, two things I love coming together like mint and orange juice.

          • cassander says:

            @axiomsofdominion

            Eh, there’s not much to go into. I didn’t play it all that much. I just remember being pleasantly surprised at how sophisticated it looked when I started, with resources and trade flows and what not, only to quickly realize that none of it really mattered much. I’m sure there are mods that do better.

          • axiomsofdominion says:

            Okay, I understand. All the different resources just deal with what buildings and don’t mean much plus the percentile shit plus trade is just get any deals you can.

    • beleester says:

      I’d be interested, but like Incurian said it depends how it’s done.

      There’s certainly enough material to work with, like developing different types of intelligence sources (sigint, humint etc), counterintelligence, double agents, unreliable sources, how to act on intel without burning your sources, and so on. It’s really fascinating to read about, but I’m not sure how you’d make it into something as intuitive as pushing an army around a map.

    • ADifferentAnonymous says:

      Sounds fun, I have a lot of scattered thoughts.

      1) The payoff needs to be commensurate with the effort required. Manipulating nations’ politics seems like a promising way to realize that.

      2) I imagine the quality of the characters you use will be one of the most important factors. Where do you get them? Is it a matter of having a roughly fixed number of elite characters that you have to decide how to allocate? Or is there a more involved system for producing them? I’m sort of imagining being able to implement things like civil service exams or devshirme.

      3) Other aspects of the game can feed into espionage in eats being the obvious. When your nation gets a reputation for dishonesty, your citizens will be immediately suspect in other lands and find spying much harder. You might train “illegals” to conceal their nationality, but this won’t work if you and your rival nation are ethnically distinct. But if you can conquer a population of their ethnicity, and win its loyalty, you can draw spies from them…

      • axiomsofdominion says:

        There is a population system and an ideology system. There is also a knowledge system that covers language/culture/history. Characters can be trained in those. This also affects espionage and the opinion system. Leaders have trust/fear/etc scales. I haven’t finalized the list. Different races have different innate magical aptitudes and traits. So your spies could be half regular people of the same ethnicity and half shadow elves that have good stats for spying. Or w/e you like.

        I have not 100% finalized the system for how characters come to exist. There’s an aspect of bloodlines for the rulers/nobles. But for merchants/democracies/mages I’m not totally sure how it will end up.

        Particularly in combination with diplomacy and ideology and conspiracies espionage can have an extremely powerful impact.

    • massivefocusedinaction says:

      In most 4x games, there’s so much information available to the player via the map and units (especially if a decoy fleet can enter combat) that spying too often devolves into a nice lottery for accelerated tech developments rather than an intelligence source.

      I think it could be interesting, but to make it interesting you probably have to get players to accept that absent their intelligence findings they won’t be able to tell much about their opponents from the usual tools (like the map, comparison charts, even battles).

      • axiomsofdominion says:

        There’s no global Paradox style “ledger” and no leader board or anything. You just have the knowledge you discover. The average player would probably be starting from the grand strategy style existing world rather than the Civilization style start. A lot of systems really only kick in after a few hundred to a few thousand years of game time. Thus they’d have the knowledge the character/nation they chose to play as had gathered as an AI.

        • massivefocusedinaction says:

          That sounds pretty interesting. You might want to look back at a game like Covert Action (not a great game) that had some multi-round arc information to learn in order to progress (at the higher levels).

  10. Well... says:

    Last Open Thread someone recommended I read “The Truth of Fact, The Truth of Feeling” by Ted Chiang.

    I did, and was moved to write a review on my blog. (Contains a minor spoiler near the very end, if anyone really cares.) I was going to just write the review here but decided it was far too long (for my tastes anyway) to be a comment on a thread.

    • Marshayne Lonehand says:

      There was a flurry of Chiang-related SSC comments in September 2015 (here), October 2015 (here and here), and November 2015 (here and here) … comments that include, moreover, links-to and quotations-from from Ted Chiang’s various nonfiction essays.

      Not every SSCer found Chiang-class cognitive inquiries to be congenial, so it’s interesting to see steadily increasing levels of Chiang-respect now becoming evident among the SSC commentariat (eighteen months later).

      Since 2015, Chiang has given a bunch of thoughtful, occasionally SSC-heretical, essays that (as it seems to me) are well worth reading (for example, here and here).

      • nimim.k.m. says:

        It might a combination of things.

        Foe example, more people might have heard about him because a big successful science fiction film based on his text (Arrival) hit theaters not long ago. (That’s how I learned about his existence.) And there are network effects (say, you didn’t know about the film, but your philosophical science fiction friend did hear about it and is going to talk about Chiang more.)

        Or it is the other way around: the film got made because he is becoming more prominent author.

  11. R Flaum says:

    I just realized that the plot of The Last Hero is basically the same as the plot of the Dragonlance Legends trilogy. A former hero decides to kill the gods, but unbeknownst to him this will destroy the world. Another group sets out to stop him, but in the end (spoiler warning) the former hero must give his own life to defeat the peril he himself created. Weis and Hickman should totally sue the Pratchett estate.

  12. shakeddown says:

    Reminder that Wednesday 11pm EST is the Discord Schelling point

  13. Nancy Lebovitz says:

    I keep seeing in discussions of the US corporate tax that it’s high, but that doesn’t matter because corporations don’t pay the high rate.

    Do they have to go to significant amounts of cost and/or trouble to avoid the high rate?

    • Brad says:

      Companies that are mostly domestic have to expend a lot of effort to get their rate down really low. Multinationals that have significant sales abroad can defer taxes indefinitely of their foreign fairly easily. If at any point in the future there’s a repatriation holiday they can turn that deferral into avoidance.

      • bean says:

        I think that you’re sort of missing the point. Yes, multinationals can keep their international profits overseas, but it’s pretty rare to see a US multinational without extensive US operations. The ones that didn’t have them moved to Ireland. So Google or whoever still has to pay US taxes on their US profits.

        • Brad says:

          No, I acknowledge that. But if you take an overall corporate rate you get a weighted average of the effective US rate for domestic sales and a much lower weight (or zero) for foreign sales. Any significant sales abroad is going to bring that overall number down below 35% (leaving aside the domestic loopholes like games with deprecation and R&D which is the trickier part to pull off).

          • bean says:

            So? The point is that the tax rate for the US portions of the business is 35%, and if they want to use any of the money from overseas at home (which I know includes investment, and probably also includes things like dividends) they have to pay 35% on that, too. Yes, if we take total taxes/total earnings, it’s lower than 35%, but that’s not necessarily the metric that we want to optimize for.

          • I agree with Brad and don’t understand your point, Bean. I think Nancy’s initial point was that even though US corporations have a high tax rate, it appears that they don’t pay that high a tax rate. Brad answered that by saying that it is multi-nationals that have a lower rate, and that is because they are subject to other countries’ taxes as well as the US, so it averages lower than the US rate. I think Brad is correct.

            When I was tax manager of a medium sized firm in the ’90’s (without significant international operations), I was proud of the fact that I was able to get about a 32% effective tax rate for several years in a row. The main areas that brought down the rate of my firm was the R&D Credit and the FSC export benefit. FSC is now gone, but there is similar credit today for domestic manufacturing. By the way, depreciation has no effect on these rates, because it only affects timing, and not permanent differences. Effective tax rates refer to the tax on the P&L statement, which ignores timing differences.

            Nowadays one sees many effective tax rates in the ‘20%’s or below. That is because of the much lower tax rates outside the US, and it is only multi-nationals that now achieve such low rates.

          • Douglas Knight says:

            It is widely claimed that the effective tax rate on domestic corporations in America is the same as in Germany, the more deductions canceling the double rate.

          • Nancy Lebovitz says:

            Mark V Anderson

            My point was to check on whether corporations have costs related to lowering their effective tax rate– they might be better off if the US tax rate was closer to the world average.

            Also, it’s reasonable to assume that whatever corporations pay to lower their taxes is a dead loss– it does no actual good to anyone.

        • massivefocusedinaction says:

          If your firm does any trade with itself across borders, it can usually recognize any profits it wishes in whichever domicile it chooses, even profits that would otherwise be recognized from US operations. The common way is via intellectual property licensing.

    • episcience says:

      Tax lawyer here (though not a US lawyer, so I might be slightly broad-brush in places). The short answer is that it’s not hard to move profits offshore if you’re a multinational company, especially in the US, but it’s becoming harder.

      First, the US is unusual in having a full exemption for foreign profits earned by US companies. Most countries tax the global earnings of companies tax resident in their jurisdiction. But the US provides an exemption if funds are “indefinitely reinvested” offshore. Given the globalised market, it’s not difficult to make good use of that cash without ever paying US tax on it.

      Second, there are a lot of common strategies companies use to decrease taxable profits. The most prevalent is debt. Companies get tax deductions for interest they pay, and it’s easy to reallocate funds within a multinational group. The result of these two facts is that companies push a lot of debt into higher-tax jurisdictions and have the interest payments being made to a company resident in a lower-tax jurisdiction.

      Similar tricks can be used with royalty payments for intellectual property such as trademarks or “know-how” (which is what Starbucks did/does), and with pricing goods and services between companies in your group. Again, by placing expense in a high-tax country and the corresponding income in a low-tax country, the global tax bill is reduced. This strategy is generally called “transfer pricing”.

      Third, a lot of “American companies” are not headquartered in the US, or have their parent company registered in a different jurisdiction. Strategies covered well in the media (like the “double Irish”, or corporate inversions) are used to make sure profit that otherwise would have arisen in a US tax-resident company arises somewhere else.

      So it’s not hard, as long as you have (a) enough money to hire one of the big four accounting firms to do the work for you and (b) enough global scope that it’s relatively convenient to create a few new companies in various jurisdictions.

      • Hi, I’m a tax accountant in the US. Please explain your comment about the US being unusual in receiving a full exemption for foreign profits. IT is my understanding that almost all countries outside the US use the territorial system, which means that these countries tax only those earnings that are earned in that country. All global earnings are taxable to any US individual or entity. The only way to avoid taxes on these earnings is to have a foreign entity (owned by the US parent) earn the income in a foreign country. I certainly don’t understand income taxes outside the US as well as US taxes, so please let me know if I am misunderstanding these other systems. It seems to me that the US always taxes MORE income than countries with territorial systems.

        • episcience says:

          Most countries outside the US have a “dual basis” system, so they tax (a) worldwide income of residents, and (b) local income of non-residents. I’m not aware of many countries that are explicitly territorial outside of low-tax jurisdictions like Hong Kong.

          The US foreign profit exemption I was thinking of is exception from the CFC rules (“controlled foreign company” rules, for non-tax people reading). Most OECD members will tax CFC profits on an accruals basis, the US (as far as I’m aware at least) doesn’t tax CFC profits as they accrue if they’re being reinvested offshore.

          • Anonymous says:

            Most countries outside the US have a “dual basis” system, so they tax (a) worldwide income of residents, and (b) local income of non-residents. I’m not aware of many countries that are explicitly territorial outside of low-tax jurisdictions like Hong Kong.

            Some make due with just (a).

          • Most countries outside the US have a “dual basis” system, so they tax (a) worldwide income of residents, and (b) local income of non-residents. I’m not aware of many countries that are explicitly territorial outside of low-tax jurisdictions like Hong Kong.

            This is certainly different from my understanding. That sounds essentially like the US system — since the US only charges tax on income earned in the US on non-US persons. That’s why inversions.

            I don’t know what accruals basis means. Most accounting is done on an accrual system, so it sounds like you are saying that if a France company has a 100% owned CFC in the US, that all of the the CFC’s profits are taxable in France?

            I need to educate myself somehow on corporate income taxes outside the US. I am depending too much on what I hear from random journalists or commenters — which is quite often too partisan to be trusted.

          • Strange, it didn’t give me the option to edit after my posting. I wanted to say that it just occurred to me that episcience could be considered a random commenter, but my posting was not aimed at him. It somehow seems that SSC is more trustworthy, and also episcience doesn’t appear to have any partisan agenda.

        • Glen Raphael says:

          @Mark V Anderson:

          I’m a tax accountant in the US.

          Hey, maybe you can help me on this. I heard/read a claim that Apple’s thirst for unconventional materials in the Apple Stores and Apple Park started as a tax avoidance strategy. The claim is that Apple has lots of profit trapped outside the US which effectively makes it cheaper to spend money outside the US than inside the US. So if they can somehow use Apple’s outside-the-US profits in Europe to buy expensive European materials which get shipped to the US to make fancy buildings, this is a way to repatriate earnings in physical form. A specific example is the architectural use of giant sheets of glass SO big and special they can ONLY be made at a particular factory in Germany.

          I don’t know where I got this claim from, and I’m not sure the math works. Care to comment?

          (If it did work, this strategy would be a bit like the way Brits in the 98% income tax bracket used to buy Rolls Royces. Their choice was either (1) put money in a bank account and earn a piddling insignificant amount of post-tax income, or (2) buy a Rolls and it probably depreciates…but you get the use value of owning a Rolls, which is worth more than the foregone income stream.)

          • Nornagest says:

            heir choice was either (1) put marginal earnings in a bank account and earn a piddling insignificant amount of post-tax income, or (2) buy a Rolls and it probably depreciates…but you get the use value of owning a Rolls

            I don’t understand. Is this pre-tax money?

          • Glen Raphael says:

            @Nornagest

            [about the Rolls Royce story] I don’t understand. Is this pre-tax money?

            I just tweaked the wording to be less confusing. I’m actually talking about what to do with post-tax savings of a high earner. Suppose you are in the 98% marginal tax bracket. Maybe you’re a pop star, or a banker. You’re already making so much income from your regular job that any extra income you make will get taxed at 98%. (this was pre-Thatcher; it’s not like this now).

            Imagine you have $100,000 sitting around in savings. How should you invest it? Let’s generously imagine your investments are paying 10% risk-free interest, so you could keep that money in the bank and earn $10,000/year in interest…but then after taxes all you get to keep of it is $200/year. Thus, the income stream you get by investing $100,000 is $200/year.

            If you use the same money to buy a $100,000 automobile, the opportunity cost of owning that car to you is $200/year. Thus, rich people blow their money on things like ludicrously expensive cars that they can at least get some use value out of, since they get almost no use value from saving it.

          • Nornagest says:

            That makes sense, thank you.

          • Hey, maybe you can help me on this. I heard/read a claim that Apple’s thirst for unconventional materials in the Apple Stores and Apple Park started as a tax avoidance strategy. The claim is that Apple has lots of profit trapped outside the US which effectively makes it cheaper to spend money outside the US than inside the US. So if they can somehow use Apple’s outside-the-US profits in Europe to buy expensive European materials which get shipped to the US to make fancy buildings, this is a way to repatriate earnings in physical form. A specific example is the architectural use of giant sheets of glass SO big and special they can ONLY be made at a particular factory in Germany.

            If I am understanding this right, it would not work. Are you saying an outside US (OUS) affiliate of Apple would invest in some fancy German product to be used in the US? It sounds to me that the OUS affiliate would still own the product. There is a rule that when an OUS CFC (affiliate) invests anything in the US, that is considered subpart F income to the extent that CFC has US untaxed profits. Subpart F income is fully taxable in the US.

            They need to have this rule to avoid the OUS CFC’s from lending money to the US company, which has the same cash effect of a dividend back to the US.

    • The Nybbler says:

      Do they have to go to significant amounts of cost and/or trouble to avoid the high rate?

      Most of the crazy accounting tricks in the news have been to avoid _EU_ taxes. US corporate tax is another thing entirely, and there’s no One Weird Trick (like the now-banned Double Irish with a Dutch Sandwich) to cut it, though there are more ordinary shelters.

      What multinationals tend to do is realize profit overseas and then keep the money there (if they repatriate, they get taxed, though they can deduct the overseas taxes at that point also). They also do things which appear to be silly, like borrowing money at interest while sitting on a load of cash, to avoid having to repatriate. In a sense they’re still using the money as they wouldn’t get as favorable terms if they weren’t. Some companies (like Apple) are sitting on really enormous overseas cash reserves; I think, therefore, it’s very likely the tax matters a great deal. Note also that keeping money overseas doesn’t help companies which make the lion’s share of their money here, so companies serving the US market are at a disadvantage.

      • Aapje says:

        The EU tax evasion schemes can be used to avoid US taxes:
        1. Ensure that the subsidiary that makes the profit is in a tax haven country
        2. Ensure that the profit is made in a way that is taxed minimally by the tax haven

        Step 1 can be done by making the US company pay to use the patents or other IP of the subsidiary, at a cost that is suspiciously close to the profit. If the subsidiary is in the EU, the profit is now by a EU instead of US company.

        • random832 says:

          Are royalties on US patents not taxable if the owner is a non-US company? Seems like an obvious loophole.

          • Are royalties on US patents not taxable if the owner is a non-US company?

            Royalties are taxable in the country where the patents are located, regardless of who owns them. Being an intangible, the location of patents may be an issue. I don’t know what constitutes the location myself. But if the R&D to create the patents are done in the US, then the patents are in the US. These patents can be be moved outside the US after that point, but the US does have a rule that that tax is then due based on the fair market value of the patents.

          • Aapje says:

            @Mark V Anderson

            The issue is that IP can be created for the express purpose of tax evasion and valued at a level that is optimal for that purpose.

            For example, let’s say that I own a coffee shop company called Pequod. I can create a new design for my cups in the US, send the file to The Netherlands, have someone in the IP subsidiary wave their hands over it and now make Pequod pay 100 million to the IP subsidiary to use that design on their cups.

            The design really has no market value without the link to Pequod, so it has no liquidity. This means that the value is merely in the price that Pequod is willing to pay to the IP subsidiary. Of course, since they are both owned by the same people, the price is set by collusion.

          • @Aapje. Maybe you can do that, but only if you hoodwink the US tax authorities. There are laws against it.

            When you send the new design to the Netherlands, you have to pay US taxes on the fair market value of the design. Later, when Pequod pays its Netherlands subsidiary royalties, these royalties too must be based on fair market value. This will probably be the initial fair market value of the design divided by the expected numbers of units sold over the life of the design, adjusted by a discount rate to account for the time value of money. Actually the original fair market value of the design is probably calculated using this same method. Companies selling to themselves across international borders definitely do NOT have a free hand in setting the prices they charge.

            So if the tax authority is doing its job, the royalty deductions of Pequod/US over the life of the design is exactly equal to the income they originally declared when they sent the design to NL. As I’ve said before, the large multi-nationals may still come out ahead just because they can afford smarter lawyers and economists than the IRS (US tax authority), so they probably can hoodwink them to some extent. But the US laws themselves theoretically don’t allow the gaming you suggest.

          • Aapje says:

            @Mark V Anderson

            Imagine that Pequod would use a ‘beauty contest’ to let designers from across the world make a new design for them. A thousand companies create a design. The value of these designs are very small until one is chosen by Pequod, at which point that one design becomes very valuable and the value of the other 999 goes to zero dollars. So having a design chosen by Pequod greatly increases the value.

            It seems to me that this same mechanism can be used in my scenario. The design that is sent to the Dutch subsidiary can be valued at a rather minimal value (say $100k), yet when ‘chosen’ by Pequod, it becomes valued at $100M.

            I would argue that the only way for the US to penalize this, is if they have evidence that the design had already been chosen at the time of the IP transfer. Of course, the US could argue that since only one design was transferred, this is evidence that the game was rigged. But then Pequod could evade this by having a ‘beauty contest’ where other companies then their IP subsidiary can submit designs (although those will never be picked, of course).

            However strict the US may be, you can pretty much always just add levels of obfuscation so Pequod is safe.

          • random832 says:

            These patents can be be moved outside the US after that point

            How? They can’t cease to have been issued by the US Patent Office. Being owned by a non-US entity shouldn’t mean that the patent is not a US patent.

            That’d be like saying that property tax should not be due on land owned by foreign (or out-of-state, since property taxes are mostly state taxes) entities. It makes no more sense to be able to move a patent than to be able to move a piece of land.

          • episcience says:

            How? They can’t cease to have been issued by the US Patent Office. Being owned by a non-US entity shouldn’t mean that the patent is not a US patent.

            The patent is sold to a non-US company. Then the new patent-holder licenses the patent to another group company in return for royalties. The licensee receives tax deductions in the country where it is tax resident, which might be a non-US country. If it is, then the patent-holder is a foreign company which has made its profit from the use of that patent outside the US, so the US doesn’t tax that income (despite it still being a US patent granted by the USPO). Even if the licensee is resident in the US and uses the patent in the US, tax treaties will usually mean the US doesn’t get to tax that income.

          • @Aapje. Okay now you are theorizing a situation where the development of the patent does not take place in the US. I think the value of the patent occurs in NL, as you say, and thus royalties should be paid out of NL. Your supposition isn’t about gaming the system — it just brings up a very unusual case where the taxes work out well for the company. In most cases, the value of IP belongs in the country where it is developed. Thus the deductions of creating the IP is in the same country that ultimately earns the royalties. It is symmetric and not particularly tax beneficial. They can sell or transfer the IP to another country, but the US at least has laws that cover that situation.

            @random

            These patents can be be moved outside the US after that point

            How? They can’t cease to have been issued by the US Patent Office. Being owned by a non-US entity shouldn’t mean that the patent is not a US patent.

            You seem to imply that the location of the patent is in the US because that is where it is valid. That actually makes sense now that I think about it, but I don’t think that’s how it works. A company invents a new product in country A. The company then gets patents in every country where they think it is worth doing so, which might be 30 different countries. All those patents are based in country A because that is where it was developed, not in the 30 countries in which it is licensed. The company can then sell or transfer the patents to any other country as desired, but they have to follow the resulting tax rules of country A and the recipient country.

          • Aapje says:

            @Mark V Anderson

            Okay now you are theorizing a situation where the development of the patent does not take place in the US.

            I explicitly described a situation where the IP was developed in the US, but became worth more after being sold to a Dutch subsidiary. Such changes in value are exploitable.

            In my scenario, the IP is used specifically to ensure that the profit is made in a different country, in a way that is taxed minimally by that country. You can keep claiming that this is not true, but it is a fact that what I’ve described has been a crucial part of tax evasion schemes.

          • You can keep claiming that this is not true, but it is a fact that what I’ve described has been a crucial part of tax evasion schemes.

            I read your link, and I have no idea why you compare that to your strange scheme of making the IP valuable only when it moves to NL. The link didn’t talk about how they got around the issue I discussed, which was to move the IP outside the US with minimal US taxes. And in fact they didn’t explain anything in enough detail to understand what they were doing. Mostly it was hand waving “exploit a loophole.” So no I don’t see your method working in real life.

          • Aapje says:

            @Mark V Anderson

            Let me try once more:

            Do you agree that the valuation of IP can change? For example, imagine that I get a patent for a boopdiebooper. Today there is no one that came up with a way to use a boopdiebooper in a useful product. Then tomorrow, Google comes up with such a product that every human wants to have and where my boopdiebooper is a crucial part. Suddenly my patent became hugely more valuable.

            My point is that you can create such valuation changes in a controlled way, where the value goes up after selling the IP to a foreign subsidiary, by lying to the government in ways that are undetectable to them.

          • We are using this thread far beyond its expiration date. So I’ll just say I agree with you that there are ways to fool the government to come out ahead (that was my point about the smart lawyers and economists, after all). I don’t think it is as easy as you imply, but my expertise lays mostly in knowing and working with the law and regulations, not in planning ways to profit from them. So maybe your idea is a workable plan.

  14. Jaskologist says:

    Mulling over the Getting High post, my main response to “why don’t facts work anymore?” is “because you guys keep telling us things which aren’t so.”

    How could we quantify this? I feel like I see one of these at least every fortnight now, and I’m not even actively looking. Is there some metric we could set up to measure how often the media reports to us “facts” that turn out not to be, and which side those “facts” favor? Anybody up for this project?

    • Aapje says:

      You first need to define what counts as a falsehood for your metric. Lots of articles have mistakes, where the seriousness ranges from to an insignificant factoid to the evidence being presented a little stronger than is warranted to one piece of evidence being made up to the central claim being false. Where do you draw the line? If you count different cases of different severity, do you give more weight to the more severe cases?

      How do you deal with bayesian mistakes? If the claim is ‘X is most likely true’ and the evidence suggests that ‘X has 50% chance of being true,’ do you count this the same as a situation where the claim is: ‘X is true,’ while the truth is ‘X is 80% certain to be false?’

      • Anonymous says:

        You first need to define what counts as a falsehood for your metric.

        Crucially: Does reporting about true events that confirm a preconceived conclusion and suppressing true events that deny that same conclusion count as falsehood? Because AFAIK, that’s how newspapers such as the NYT work.

    • Conrad Honcho says:

      It sounds like you want PolitiFact fact-checker for news, NewsFact. This inevitably results in the same bias PolitiFact has. Take a speech where the candidate you don’t like makes 95 statements of true fact and 5 statements of questionable validity. Ignore the 95 true statements and take the least charitable interpretation possible of the other 5 and declare her a HUGE LIAR. For the candidate you do like, ignore their 5 questionable statements and “fact check” 5 of their true statements. Declare him PARAGON OF VIRTUE.

      The other reason facts don’t matter is because the facts support lots of different narratives. If on the same day a white cop shoots an unarmed black kid and an illegal immigrant rapes a white college girl, Breitbart devotes 99% of its coverage to the rape and 1% of its coverage to the shooting and CNN does just the opposite, no one’s lying. And if you’re super mad about cop shootings, well the facts are on your side. And if you’re super mad about illegal immigrant crime, well the facts are on your side. And it’s that stupid, fact-hating Other Tribe that just won’t listen to the facts! The problem isn’t the facts. The problem is that different people legitimately care about wildly different (but still true) sets of facts.

      I think the actual way forward is to recognize that bias is an inherent part of individual humans and organizations, take everything you hear with a grain of salt, and use a variety of sources so you know what facts the other side cares about. Read Breitbart, read Current Affairs and Mother Jones, and only watch CNN when you want to know what they’re lying about this time.

      • Jaskologist says:

        To clarify, what I’m really hoping for is that some of the left-leaners around here will join in and we can all pre-commit to a metric to try out for the next few months. I realize the problem space may be way too complex for that.

        • Conrad Honcho says:

          I think the problem space is too complex. Even ignoring the (much bigger I think) problem of lies of omission,

          1) What is the granularity of a “fact?”

          2) The egregious lies are provably false, but most narrative-driving facts are of the form “sources inside the White House say…” followed by something entirely unverifiable and immediately contradicted by other sources inside the White House.

          3) Semantic arguments, strategic equivocation, isolated demands for rigor make determining the truth value of facts very difficult. “Black Lives Matter” but “Race doesn’t exist.”

          Ultimately I think the problem is people thinking we’re arguing about facts when really we’re arguing about values and the facts are merely a proxy war. We can completely agree on the facts and have 180 degree opposite ideas of what to do about them.

          I don’t think impartiality is possible, I don’t think it’s possible to be unbiased. I would rather have an adversarial media system with clear, stated partisanship rather than anyone trying to claim impartiality. CNN, ABC, CBS, etc, are flying flags of neutrality while smuggling contraband.

          • Aapje says:

            It’s also a shit ton of work, if you want to do it well and in a relatively objective way (not just fact checking the stories that stand out).

            I doubt anyone wants to commit the time.

        • Paul Brinkley says:

          Lots of work, but you could at least explore the terrain together and acquire a shared understanding of the challenges.

          I used to work in semantic modeling for several years. Some of the challenges were mentioned above. There are others.

          (1) Establish what counts as a fact, as Conrad Honcho says. More formally, you want what philosophers would probably call a proposition. Propositions are the primary truth bearers. “1 + 1 = 2” and “1 + 1 = 6” are both propositions. (Ignore semiotics here.)

          (2) Decide whether you want to try to handle this manually or programmatically. If the former, accept that this will be an extremely academic exercise in which you go through maybe one or two articles as case studies; you won’t have the energy for more. Otherwise, continue.

          (3) Establish a canonical representation of propositions. This has several sub-problems.

          (3a) Basic stuff that looks like propositional calculus. “Haley visited the UN” might translate to “visit(Haley,UN)” where the tokens are appropriately specified (see below). W3C’s OWL 2 is as yet insufficient, so if you use it, be prepared to build at least another layer on top of it.

          (3b) Establish what your entities are allowed to be. In other words, ontology. You’ll probably have people, organizations, locations, and several dozen subtypes of each for your first article. These types will populate a heterarchy – acyclic subsumption with multiple ancestors permitted. (I.e. you need to be able to say Haley is an official and female.) Your relations between entities will also populate a heterarchy.

          (3c) Temporal representation and event calculus. You’ll need a way to say “Haley visited the UN on Thursday” in a way that lets a checker (human or program) say this happened before anything Haley did on Friday, or at the same time as anything anyone did on Thursday or May 11, 2017. Your logic must therefore tolerate time zones. (I suggest translating everything to UTC internally.) Your entity and relationship heterarchies must accommodate temporal logic, too: you will need to be able to say that people are not always officials, but that officials are always people, etc. Note that temporal logic by itself will require you to use a reasoning engine mathematically more powerful than any relational database.

          (3d) Modal logic: you will need to be able to reason over instances of propositions. For example, Sean Spicer said that Haley visited the UN on Thursday, or that according to Gallup, 35% of those surveyed answered “foo” to question “bar”, or that Gary Johnson didn’t know where Aleppo was, or that the fact that Uruguay has orbital mind control rays is classified.

          (4) Establish a mapping from propositions as presented in news to propositions as presented by whatever canonical system you settle on. In other words, you will need to solve the parsing half of the NLU problem.

          (5) Establish a deductive system for concluding new propositions from older ones. If you’re doing this manually, this will probably be how you run out of energy after one article.

          (6) First-order and second-order logic rules that keep you from concluding silly things like Haley being at both the UN and Raleigh-Durham at the same time, or that Trump signed a bill after it became a law, or that Uruguay has orbital mind control rays.

          Addressing these challenges should be enough to get you started.

  15. Atlas says:

    Does anyone have any thoughts they’d care to share about Neal Stephenson’s novel the Diamond Age? (I read it recently and thought it was wonderful.)

    • Incurian says:

      It was pretty good and you should keep reading Stephenson because they’re all good. Also, you should read The Culture series by Ian M Banks. The individual novels are of variable quality, but on the whole it’s pretty amazing. Actually, the quality of the writing is uniformly good, but I enjoy some stories much more than others.

    • ThirteenthLetter says:

      Like a lot of books from Stephenson’s middle period, great ideas, middling execution, terrible ending. I’m a much bigger fan of his very early stuff (The Big U, Zodiac, Cobweb) and his most recent stuff (well, I enjoyed Reamde, anyway, and really need to make another attempt at Anathem sometime.)

      Edit – I think a commonality about the Stephenson books I like is that they’re set in the modern day. He’s really, really good at painting pictures of the modern world and filling them with unique details. It’s harder for me to get on board with his fantastical or historical worlds.

      • Nancy Lebovitz says:

        I liked the historical parts of Cryptomonicon much more than I liked the modern bits. The modern bits were one-note satire and I got tired of it.

        • HeelBearCub says:

          I liked Cryptonomicon well enough as I read it, but by the end I was so unsatisfied with the thing as a whole, that I haven’t tried any other Stephenson book.

          For some reason, the protagonist standing on a rock in the middle of a river of molten gold was the last straw for me. Can’t quite put my finger on why.

          • John Schilling says:

            Stephenson is notoriously bad at endings, and my biggest problem of Cryptonomicon is that the major conflict of the plot remained unresolved, and near as I can tell the major adversary was not only undefeated but clearly positioned to claim the river of gold for his own faction. But, yes, fun while it lasted, and that can be good enough.

            Hmm, which of Stephenson’s works have satisfactory endings?

            The Baroque Cycle I think, absolutely, but with 2500 words he damn well better have found room for an ending.

            Diamond Age is perhaps jarringly abrupt at the end, but we can see how the main plot is going to finish playing out, so OK.

            Seveneves, the near-future section ends reasonably well and the second part serves as an extended epilogue, but introduces new characters and new plotting that deserve a better ending than they get.

          • Rob K says:

            @John Schilling I think my biggest problem with Stevenson’s endings is that he so often promises (or at least gestures at) a much grander scale than the ending eventually operates on.

            So, in Cryptonomicon, we get these forays into novel ideas of how currencies could function, etc, which makes the eventual outcome of “some guys we like get a bunch of gold” seem “smaller”, so to speak, than it otherwise would.

            Seveneves, although I agree about the oddness of the short far-future section, at least offers endings to both sections that play on the scale that the setup promises.

          • Incurian says:

            Agree the that the Baroque Cycle had an excellent ending, but yeah it was like a whole book. Reamde wrapped up nicely too, though.

            So, in Cryptonomicon, we get these forays into novel ideas of how currencies could function, etc, which makes the eventual outcome of “some guys we like get a bunch of gold” seem “smaller”, so to speak, than it otherwise would.

            The eventual outcome is Snow Crash and Diamond Age!

          • Well... says:

            Hmm, which of Stephenson’s works have satisfactory endings?

            Books I’ve read by him, ranked by how good I found the endings (meaning the non-dramatic stuff that happens after the final climax):

            1. Snow Crash
            2. Diamond Age
            3. Zodiac
            4. Reamde
            5. Anathem
            6. Seveneves*

            *Though, the ending to the near-future part was very good (if I’m remembering it right), and had that been the end of the book I’d put Seveneves at 1 or 2.

          • Vermillion says:

            I thought Seveneves was 2/3rds of a great book and then blurghglhlylghllglgy. And then I learned it was originally going to be the backstory of a MMORPG and that made a lot of sense but it didn’t make me like it anymore than before I had learned it.

          • Well... says:

            I still really liked the second part of Seveneves. Its crappy ending only took up a few pages at the end, which I’m happy to forgive Mr. Stephenson.

          • Douglas Knight says:

            Vermillion, what do you mean by “learned”? Did you encounter some outside evidence? Or did you just consider the hypothesis and think it correct?

    • Well... says:

      In some respects (though not in others) this is my favorite of the 6 or 7 novels I’ve read by him. I thought it contained some of his very best writing (though the proofreading evidently wasn’t as good as his other books–maybe it was rushed?), the concepts were compelling and fascinating, the imagery was incredible (if it were a movie, James Cameron would have to really stretch to even begin to do it justice), and it had a surprisingly hard-hitting emotional impact (for me anyway; I have a daughter who is just like young Nell–though not abused or neglected!). Unlike most of the other Stephenson novels it also had a non-corny ending.

      On the downside, I wanted to know how some of the other plot lines resolved. E.g. we just stop hearing about Judge Fang after a while; what happened with him? What role did he play in the revolution of the Fists, or was he not part of it? Etc. And, some parts of the book felt rushed.

      This book actually changed the way I approach parenting a bit. Results aren’t in yet but so far things look good.

      • Brad says:

        Straining my memory a bit here, but as I recall there were two characters with legit development arcs. The little girl was pretty much standard wish fulfillment. She grew older, stronger, wiser, and better but the mid-level engineer arc was less hackneyed. Did any of the books have a better character development arc than that second one? Maybe some character in the Quicksilver series?

        • John Schilling says:

          All three primary characters in the Baroque Cycle get strong arcs. Jack Shaftoe from two-bit con artist to pirate captain, Daniel Waterhouse from second-rate scientist to player in the Great Game, Eliza of the unpronounceable surname from harem slave to genius financier, and all three of them to decent human being by the end (though Waterhouse perhaps started out there).

          Also, Stephenson deserves a nod for demoting Johnny Depp’s amusing but overrated character to only the third-greatest fictional pirate to go by the name of “Captain Jack”.

          • Well... says:

            This is a tangent, and I live under a rock so maybe my perception of his reception is off, but why have so few people noticed that Johnny Depp is not a great actor?

          • HeelBearCub says:

            Johnny Depp was an absolutely amazing actor.

          • Aapje says:

            It regularly seems to happen that actors just start phoning it in/hamming it up* when they are famous, probably because loads of people are content just to see a famous actor and can’t suspend disbelief when recognizing the person anyway.

            This makes it extra painful when someone like Seymour Hoffman dies, because he stayed at a high level.

            * See De Niro and Pacino

          • Well... says:

            When was Johnny Depp an amazing actor? The one role I thought he was perfect for was Edward Scissorhands, in which he basically played an animated mannequin. (I.e. any actor who could follow his blocking could have done just as well.)

            His two most expressive roles–Hunter Thompson and Jack Sparrow–were both basically decent impersonations of people who are, for whatever reason, very easy to impersonate well (Hunter Thompson and Keith Richards).

      • Nornagest says:

        IIRC, there’s a throwaway line somewhere towards the middle of the book saying that Judge Fang disappeared into the Middle Kingdom. It may be inferred that he didn’t play an active part in the revolution, though he might have been behind the scenes somewhere.

        I’d like to have seen more of him, but Stephenson was clearly straining to write even as much as we see of the Confucian side.

    • Jugemu says:

      I really liked it overall, though there was some gratuitous weirdness. I also think the ending is fine, albeit abrupt.

    • J Mann says:

      I loved it – I think it’s more accessible and tighter than his later works, but serves as a good canvas for his creativity.

      If you want to check out a little more plot/action-oriented nanotech book, I liked Walter Jon Williams’ Aristoi quite a bit.

      • Nancy Lebovitz says:

        Aristoi is entertaining, but oh God it leaves out the amount of administration you’d need to manage nanotech in order to have a fantasy of really cool person in charge.

        • John Schilling says:

          I give it credit for acknowledging the management structure needs to exist, and showing us glimpses of it in action. Specifically, the meritocratic small-a aristocracy that is privileged to develop nanotech but has to run their final designs past the RCPiC for approval before the fabs will process the build.

          Hard to see how you could do much more in a plot/action-oriented book.

    • Nornagest says:

      Diamond Age is probably my second favorite Stephenson book, after Cryptonomicon. It’s in an awkward place between early- and middle-period Stephenson, though; it’s an enormously ambitious concept and probably needed a book the size of Cryptonomicon or REAMDE to really do it justice. Instead, the second half feels inadequately explored, even accounting for the fact that everything’s going to hell in-universe and the characters feel just as lost and rushed as we do. This is mitigated somewhat on rereading but it’s still a problem.

  16. sketerpot says:

    Why do we seem alone in the universe? Where are all the alien civilizations we would expect to have arisen before us? My new favorite hypothesis is that the universe only looks empty because the Old Ones haven’t yet emerged from their sleep of aeons.

    If a civilization wants to maximize computation it appears rational to aestivate until the far future in order to exploit the low temperature environment: this can produce a 10^30 multiplier of achievable computation. We hence suggest the “aestivation hypothesis”: the reason we are not observing manifestations of alien civilizations is that they are currently (mostly) inactive, patiently waiting for future cosmic eras. This paper analyzes the assumptions going into the hypothesis and how physical law and observational evidence constrain the motivations of aliens compatible with the hypothesis.

    That is not dead which can eternal lie: the aestivation hypothesis for resolving Fermi’s paradox.

    • shakeddown says:

      This is basically a version of the “pass into transcendence” theory, if I’n reading it right. Which still leaves the question, why are there no civilizations that rose between us and the last cleansing?

      (For the record, my theory is that interstellar space exploration is not technically feasible).

      • Le Maistre Chat says:

        How common do you believe extraterrestrial intelligences are, and why don’t we see more stars that only radiate in the infrared or appear partially, uh, Dyson-ized?

        • sandoratthezoo says:

          Stipulating that interstellar space exploration is not technically feasible, there might easily be several thousand currently extant highly technological species in the Milky way, doing things that would be immediately obvious if we closely examined their stars, and we just haven’t noticed because we’ve never looked closely at one of the few thousand stars that they’re on.

      • sketerpot says:

        Check out section 7, “Interactions with other civilizations”. Starts on page 16.

      • andrewflicker says:

        I still haven’t heard a compelling excuse for why traditional slow unmanned probes aren’t all over the place, even if more “exciting” interstellar travel is impossible.

        • sandoratthezoo says:

          How about “because why”? If traditional slow unmanned probes are possible, but yield results centuries later with no prospects of those results leading to anything other than fairly mild astronomical observations, well, maybe a given highly technological civ might send a few to a few nearby star systems just for kicks. But who sends a million of the things given those returns?

          • John Schilling says:

            A post-singularity civilization for whom sending a Daedalus probe is the rough equivalent of a middle school science project?

            Yes, yes, the One True Path for post-singularity civilizations is to condense into a heavily protected fist-sized lump of computronium inside of which a trillion sentiences contemplate their navels with minimum latency. It’s always a weakness to Fermi-Paradox explanations when one has to assert that every single advanced civilization will behave in one particular way.

          • sandoratthezoo says:

            There are no middle-school projects that have a time-scale of centuries (or millennia. Or hundreds of millennia).

            Like, I just don’t buy that “we could pretty much afford this” means “we will do it to the nth degree no matter how little it benefits us.”

            (By the way: the assumption that of course I think that there are post-singularity societies that are in a fist-sized lump of computronium speaks to the kind of weird culture here. You have not correctly identified the direction of my disagreements.)

          • andrewflicker says:

            Mostly the idea is that you build a self-replicating probe, to make it easier to explore nearby star systems where the results will still matter to your grandkids (or whatnot), and then those can easily expand throughout the galaxy. I think these are often called Von Neumann probes, and seem to only be a handful of decades away from our own capabilities (arguably we might be able to do it with current tech if you had enough governments willing to devote an absurd amount of money and engineering resources to it).

            And yes, the extension of this argument is that even if the first “wave” of probes has no benefit to even the great-grandkids of the launchers, it eventually gets cheap enough that some people/governments/civs do it anyway because of an attention to matters beyond their lifespan (after all, many people on Earth are environmentalists that care about the state of the Earth hundreds of years from now!).

          • sandoratthezoo says:

            Those aren’t traditional slow unmanned probes, they’re “space colonization.”

            So, look, here are the facts on the ground. One or both of these things are very likely true:

            a. It is practically impossible for technology to advance so far that you can create a machine intelligence that can go for hundreds of years to someplace and then create an entire technological infrastructure out of found materials and a very small probe.

            and/or

            b. We are the only intelligence which has ever evolved in the Milky Way because something between “planets in the habitable zone” and “able to build computers” is incredibly unlikely.

            All the other answers to the Fermi Paradox are, like, “WELL, what if the universe were set up incredibly conveniently for the purposes of my sci-fi dreams.”

        • Thegnskald says:

          The Industrial Revolution required that organisms evolved cellulose eons before an organism evolved the capacity to digest it. Among many other smaller things.

          Maybe most intelligent life just don’t have the hydrocarbons to get out of orbit.

          • Nornagest says:

            The early Industrial Revolution was wood-powered, and if you really need the energy density it’s not hard to make charcoal, wood gas or biodiesel. I think fossil fuels are overrated as a limiting factor; we might have had a slower takeoff, but it seems quite plausible to me that we could have limped along with wood and hydro until we got to solar and wind.

    • roystgnr says:

      “We want to save most of our negentropy for when it will be 30 orders of magnitude more efficient” sounds like a brilliant idea, except:

      A) I think assumptions 4 and 6 in that paper may be incompatible. If your civilization is letting all the stars you control spill useful energy into space at enormous indistinguishable-from-natural rates, you may not be able to defend those stars from another civilization who isn’t so limited.

      B) Even without competitors, isn’t there an unstated assumption here, that aestivating civilizations are happy to let most of their stars’ output go to waste? Why would that be the case? If you’re patient enough to wait a few hundred billion years and spend your energy in a cooler universe, why not gather more energy to spend while you’re at it?

  17. onyomi says:

    This short video is the best take I’ve yet seen on a question which has long perplexed me: why do people hate communism less than Nazism (relevant question: many people on my Facebook still seem to insist that fascism as an ideology is so bad that physically attacking people who supposedly espouse it is ethically justified; I’ve never seen anyone claim that physically attacking a fellow American for espousing communism (or something arguably akin to it, which is the relevant point of comparison, since the “fascists” Antifa attacks rarely self-identify with that term) would be justified).

    Among the reasons he cites, one is one I had long noticed, though I still don’t have a great explanation for it: namely people are more horrified by the slaughter of outsiders than of one’s own countrymen. This is kind of surprising, given that most ethical systems seem to reserve a special place in hell for traitors.

    The best explanation I can think of is one he does mention here, which is that fellow countrymen-slaughterers have a high overlap with national founders and/or “great leaders”: Lenin, Stalin, Mao. The Germans have thoroughly repudiated Hitler and Nazism, but the Russians and Chinese have not done so for Lenin, Mao, or even Stalin. In the Chinese case, especially, the ruling party, as much as it has diverged from Mao’s original vision, derives some degree of legitimacy from the revolution Mao led, and so cannot easily denounce him without hurting their own legitimacy.

    It would be interesting to imagine an alternate history in which the Nazi party retained nominal power under different leadership after WWII, and whether or not in such a case they would still have so thoroughly repudiated Hitler and his philosophy. However, such scenarios seem not often to take place because few parties/ruling ideologies can survive massive military defeat at the hands of outsiders. Put another way, a regime can seemingly kill huge numbers of its own citizens and maintain an air of legitimacy so long as it isn’t perceived as responsible for the sort of “national weakness” which leads to a military defeat at the hands of outsiders.

    Another factor which strikes me as accurate, but also a little inexplicable, is how the American war to defeat Nazism is widely, and especially on the left, perceived as the last “good” American war. American wars to defeat Korean and Vietnamese communism, on the other hand, are widely viewed as mistakes, especially in the latter case (though maybe that would be less the case today if Vietnam had turned out more like North Korea).

    I have always tended to share this general perception, especially of Vietnam (unlike most Americans, however, I also think our involvement in WWI was ill-advised), though I’m not sure this is a very consistent position for me to take. In retrospect, the government of Vietnam has not turned out to be as bad as Nazi Germany, but the government of the DPRK arguably has. Assuming we share the general perception that Vietnam was a mistake, how were we to have known, at that time, that Vietnam wouldn’t turn out like Stalinist Russia, East Germany, or the DPRK? Also ironic, it seems like a lot of the opposition to Vietnam resulted from our attempt to fight the war in a manner that would be broadly more acceptable to the left (defensively), whereas we fought WWII in a way which seems more broadly right-wing (offensively). At the end of the day, is it possible that the left hated Vietnam but loved WWII really just boils down to WWII being a war against what was perceived as a right-wing threat, while Vietnam was a war against a left-wing threat?

    • shakeddown says:

      Theory: People used to hate communism more, but then people on the right used it to attack people on the left as communist sympathizers, and it got to the point where someone expressing distaste for communism was probably just signalling his hatred of the American left.

      • onyomi says:

        This also seems right to me, though I’m not sure whether “communist” as a slur against the American left has lost its sting due to overuse, or due to the end of the Cold War. I lean more toward the latter, but if that is the case, it doesn’t well explain why fascism is today more toxic than communism, since WWII is also over.

        I guess Antifa would say that the risk of America becoming fascist is today higher than the risk of America becoming communist; maybe, but I think Hitler has been perceived as more evil than Mao for a long time, though that may also be about Westerners caring more about the slaughter of fellow Westerners, whereas mostly only third-world agrarian nations have previously become Communist (Russia, of course, having always existed somewhat on the periphery).

        • shakeddown says:

          If Antifa catches on long-term, I suspect fascist will lose some of its meaning.

          There is also an element of westerners caring about fellow westerners, but it’s not just an abstract fellow westerners thing – a lot of people in the west personally know/knew holocaust survivors/WW2 veterans, while few people in America know survivors of Mao or Stalin’s oppression (and even fewer who talk about it), so it’s a lot more distant and far-mode.

          • J Mann says:

            IMHO, the left has already played out “fascist” – these days, I hear most people escalating behavior they don’t like all the way to “nazi.”

          • Conrad Honcho says:

            a lot of people in the west personally know/knew holocaust survivors/WW2 veterans, while few people in America know survivors of Mao or Stalin’s oppression (and even fewer who talk about it)

            Do you think that’s a PR issue? There’s a lot of effort that goes into Holocaust remembrance, but not so much Holodomor/gulag remembrance. There is no National Holodomor Museum. Any time someone on reddit says something nice about communism somebody will angrily respond that they grew up under soviet communism and launch into horror stories of their youth. I think they’re out there, just nobody’s putting a camera in their faces.

            I’d like that to happen more because I really don’t like the results of Communism. Everybody knows how awful Nazism is, but they seem to treat Communism like it just had some implementation details wrong. I see people wearing hammer & sickle t-shirts like it’s nothing, but that gives me the same chill as if I saw someone wearing a swastika.

            This is why I like capitalism. We capitalists still kill you, sure, but it’s slowly, via heart disease from too many cheap hamburgers.

          • Nancy Lebovitz says:

            It’s not just the acceptance of communist symbolism, it’s the folks (and not a few of them) who say “smash capitalism” as though it’s an obviously good idea.

          • Civilis says:

            a lot of people in the west personally know/knew holocaust survivors/WW2 veterans, while few people in America know survivors of Mao or Stalin’s oppression (and even fewer who talk about it), so it’s a lot more distant and far-mode.

            A lot of people in the West know veterans of Vietnam or Korea or refugees from Vietnam or Cuba, though. (see also http://www.mercurynews.com/2017/02/23/gop-state-senator-a-vietnamese-refugee-removed-from-california-senate-floor-after-criticizing-late-senator/)

            The biggest issue is that there’s no way to isolate the political ideologies from their unique histories.

          • Deiseach says:

            If Antifa catches on long-term, I suspect fascist will lose some of its meaning.

            “Fascist” already has lost a lot of its meaning; now it’s a boo-word to signify “You hold conservative/right-wing ideas I don’t agree with and don’t like and I think you are a horrible person”, as shorthand for “this is an evil person, do not interact with them, do not engage with their arguments”. It’s the slightly milder form of “Nazi” which means simply “horrible and horribly evil person, punch on sight”.

            Same as “racist” has morphed from, while still retaining the unfavourable connotations of, “person prejudiced against another human on the grounds of skin colour or ethnic origin; typified by slave owners” to “person who benefits from ‘structural racism’ even if they personally are of a lower socio-economic class or status in society than the person of colour pointing out their racism to them”, hence why there is no such thing as ‘reverse racism’ and POC can never be racist – because ‘structural racism’ and holding power, and invisible knapsacks of privilege, you see.

      • DrBeat says:

        Counterpoint: America during the 1960s and 1970s, when basically everyone on the entire Left was in a race to crawl inside Communism’s asshole.

    • Nancy Lebovitz says:

      It gets on my nerves that capitalism is blamed for people being poor, while communism, which has done a great deal more to impoverish people, doesn’t get blamed.

      Perhaps it’s partly that preventing value from being created just isn’t as visible as not distributing wealth, though I think the “nice words” theory from the video also has a lot going for it.

      I keep saying that if people are willing to risk death to go from one place to another, and especially if this keeps happeninig for decades, then it’s reasonable to assume that the place people are leaving is worse than the place they’re going to. I’m not sure that this argument has convinced anyone.

      • The Nybbler says:

        Anyone who remained convinced of the superiority of Communism in the latter years of the Soviet Union is probably beyond convincing. Anyone who remained convinced of the superiority of Communism after the fall of the Soviet Union played out is almost certainly beyond convincing.

        Those convinced of the superiority of Communism who weren’t around for the fall of the Soviet Union… well, they were probably taught by those in the groups I mention above.

    • Aapje says:

      @onyomi

      Communism (at least theoretically) killed people for their beliefs. Using violence to suppress the belief in an ideology which used violence against people to suppress their beliefs is hypocritical.

      Nazism killed people for reasons unconnected to their beliefs. A Jew who believed in national-socialism was not safe. Using violence to suppress the belief in an ideology which used violence against people to eradicate a race is not hypocritical.

      • onyomi says:

        I don’t think it’s hypocritical; where I see the double standard is that I think if you’re going to argue it’s okay to attack people physically for professing belief in an ideology, then communism, on its historical record, is at least as punch-worthy as fascism. I don’t think attacking people physically for professing any particular belief is ethically justifiable, however.

        I do think you’re right though that Nazism especially and, to a lesser-extent, fascism in general, uniquely pattern matches to “absolute evil” within our current identity-focused civic religion.

        In theory, at least, capitalists and other class enemies could have chosen not to be capitalists; Jews and Gypsies couldn’t have chosen not to be Jews or Gypsies.

        That said, I’m not sure I agree that killing people for who they are is inherently worse than killing people for what they do.

        Consider the following two cases:
        1. I meet a black person and shoot him because I don’t like black people and I know there’s no way he could have “remedied” his blackness.
        2. I meet a pianist and say to him “I hate piano music, so I’m going to break all your fingers so you can never play the piano again; submit or die!” He refuses and I kill him.

        Is 2 better than 1 because I gave the pianist a “choice”?

      • Anonymous says:

        A Jew who believed in national-socialism was not safe.

        Neither is anyone who believed in Communism safe from Communists. So far as I know, being a Nazi Jew in the Third Reich was actually marginally safer than being a Communist of any notability in Soviet Russia.

        The Soviets could and did purge anyone they disliked, including especially themselves, and put a charge of heresy as a justification either pre- or post-facto.

        “Give me a man and I will find the crime.”

        • Aapje says:

          Neither is anyone who believed in Communism safe from Communists.

          Yeah, but that’s because they were very invested in ideological purity. It was still about beliefs.

          • Anonymous says:

            Maybe I’m cynical, but that’s doubtful. I suspect a great many died because they were inconvenient, not because they were in any objective way ideologically impure.

          • Conrad Honcho says:

            I’m not sure which is more existentially horrifying, being murdered for your race or your thoughts. I suppose at least if they’re murdering you for your race you know where you stand a priori and can try to leave. They’re more likely to correctly identify your race, too. Living day to day knowing the state is trying to read your mind and torture and murder you if they don’t like what they think they find is horrifying.

            Can we just agree it’s all incredibly evil and not do any of those things?

          • Aapje says:

            @Anonymous

            Weren’t they inconvenient because of their beliefs?

          • Anonymous says:

            Some, certainly. But those would be lower priority targets than people whose crime is existing and holding some kind of material or immaterial resource that the purgers would like redistributed to someone else (like themselves). And that’s not even mentioning the obvious scapegoats.

      • Wrong Species says:

        Communists were the first group of people the Nazis violently repressed.

    • HeelBearCub says:

      A) The biggest thing is simply that Nazi is a term that is far more bounded. Nazi’s existed in Germany for a period of less than 20 years in Germany. Note that Nazi is far more generally used than fascist. Hitler is far more referenced than Mussolini or Franco.
      B) To the victor go the spoils. The Nazi’s were decisively defeated militarily.
      C) The combination of A and B means that the record of the holocaust (which was uniquely evil, as he says) was documented in a way that any given Communist atrocity was not.
      D) The combination of A and B means we have all sorts of examples of non-uniquely-evil communism. You might not like Cuba, but it wasn’t Nazi Germany. Ditto lots of other Communist countries.
      E) The unique evil on top is always going to be over-referenced.
      F) That video is not an explanation of anything. It’s a piece of propaganda to stoke hate of some amorphous left composed of the outgroup.

      • onyomi says:

        I think your F is overly uncharitable. The video suggested some possibilities I hadn’t thought of and at least aimed for a fairly reasonable tone.

        Re. C, I’m not sure I agree with you or the video. What made the Holocaust uniquely evil as compared to e.g. Holodomor, Great Leap Forward, or the Khmer Rouge killing fields?

        • John Schilling says:

          The Holocaust probably represents the largest number of people killed as the result of a single explicit decision to kill a bunch of people for the sake of making them dead. Other incidents with similar or greater body counts had goals that didn’t specifically require megadeaths, but turned out that way due to poor implementation, e.g. Mao would have been happy to have the entire pre-famine population of China come out the far side of the process as happy, industrialized, ideologically purified socialists. And the other incidents that were deliberately murderous, e.g. the Cambodian genocide, have substantially lower body counts.

          Most people are not utilitarians or consequentialists, and so believe that intent matters. “Most deaths due to explicitly premeditated murder” is by definition a unique classification, one that matters to most people, and one that the Holocaust presently occupies.

          • Trofim_Lysenko says:

            I’ve always felt that you don’t even have to be a utilitarian to dismiss “it wasn’t deliberate” as a mitigating factor. I might be willing to grant that the various sorts of “negligent manslaughter”s aren’t as bad as pre-meditated murders, but quantity has a quality all its own. Add in that in both the Soviet and Chinese cases they knew their policies were killing people quite a long time before they decided to moderate or amend said policies…

          • JayT says:

            The Nazi’s originally intended to just exile the Jews, so you could make the argument that the Holocaust was also due to poor implementation of a plan.

          • Nornagest says:

            I think I’m with John here. Whatever the Nazis’ original intent, it’s fairly clear that the Holocaust did come out of a decision-making process the sum of which was “fuck it, let’s kill them all”.

            Comparable decisions were made on a smaller scale by the major Communist powers: dekulakization, the Great Purge. But those killed hundreds of thousands of people each, not the better part of ten million.

          • sflicht says:

            Genghis Khan’s conquest of Eurasia might be a contender.

        • HeelBearCub says:

          It’s easily recognizable as propaganda.

          “until the left [does something] we will continue to live in a morally confused world”

          As to uniquely evil, I meant “with lots of particular idiosyncrasies” (and said idiosynracies being particularly evocative). It’s like saying Jordan was a uniquely gifted basketball player. You can make arguments that he was “the most gifted” but it’s a different argument.

    • Jugemu says:

      Some of it is probably due to the relatively large number of Jews in Hollywood and the subsequent large amount of films and documentaries depicting the Holocaust in emotionally powerful ways, as compared to Communist atrocities.

      • BBA says:

        From the Jewish perspective, Communism really wasn’t that much worse than Czarism. Nazism, on the other hand, was worse than anything before or since. So that framing makes sense to me as a Jew, the question then being why do so many non-Jews seem to buy it? It can’t just be Hollywood.

        • The Nybbler says:

          It can’t just be Hollywood.

          See: “Communist-friendly ideologies, academic and cultural instruments, domination of”, a frequent topic of discussion here.

          • BBA says:

            But I’m not talking about Communist domination, I’m talking about Jewish domination, which unless I’m very confused about my own beliefs isn’t the same thing at all.

    • episcience says:

      I think the key reason is that “communism” doesn’t mean “the Union of Socialist Soviet Republics” (or any particular communist regime you care to name); i.e. that many leftists think that (a) identifying with communism doesn’t mean endorsing gulags or repression and (b) there is room for communist ideologies that don’t align with Stalinism or Maoism. See also the comments by Hannah Arendt on totalitarianism and by leftists generally on what they called Red Fascism.

      “Fascism”, by contrast, is seen as inherently violent or warlike. You cannot support fascism without supporting militarist-dictatorial control of the state and hawkish and violent views against outsiders or particular outgroups, since that is part of the definition of fascism.

      • onyomi says:

        I mean, there are communist regimes with horrifying records and communist regimes with merely awful records; there are fascist regimes with horrifying records and fascist regimes with merely awful records.

        The most salient difference, it seems to me is that communist regimes commit atrocities on their own citizens (class enemies), while fascists commit atrocities against foreigners or those perceived to be outsiders. This seems strange because killing your own people seems, if anything, worse than killing “others,” especially from a historic, tribal perspective (if we were to guess behind a veil of ignorance, based on evolutionary biology, which sort of moral system humans would evolve, I imagine it would be one which much more strongly condemns killing your own tribe, even unintentionally, than killing outsiders, even intentionally).

        This does however give me a new thought that might help explain things: communist atrocities seem to target the strong, while fascist atrocities target the weak. While fascists seek to eliminate the foreigners, the outsiders, the “perverts,” etc., communists eliminate your rich landlord who threatened to evict you last year.

        Humans do have a strong tendency to revile attacking the weak (note how even murderers apparently look down on pedophiles), so, insofar as communism is seen as attacking the strong, but fascism attacking the weak, I can see, maybe, why the latter causes more instinctive moral revulsion in many.

        • episcience says:

          I think you’ve ignored my point: communists today don’t feel that their ideology inexorably leads to violence or repression; people believe that fascism does.

        • Trofim_Lysenko says:

          So do fascists today. They make a point of pointing out peaceful ways of achieving their ends, following the same “Well, THIS time we’ll…” pattern as the communists explaining how they’ll avoid falling into the totalitarian/authoritarian patterns of the past.

          So again, both ideological adherents make that claim, and neither side has good historical support (communist governments have, without exception, been violent and repressive. The ones which have subsequently moderated their violence and repression have done so precisely to the degree that they abandoned their commitment to communist ideology in favor of pragmatic commitment to economic growth, like China and Vietnam).

          And yet on average communists making the claim that their preferred political ideology can be enacted peacefully and without repression are given far more charity than fascists doing the same.

          Your point doesn’t address that.

        • Nancy Lebovitz says:

          Letting governments have a relatively free hand in killing their “own” citizens is parallel to tolerating domestic abuse. It’s a lot of work to do anything about it.

          On the killing people outside your territory side, why is Napolean generally viewed as kind of cool rather than being seen as a mass murderer? Or is this more of an American thing, since he wasn’t our problem?

          • Conrad Honcho says:

            Distant past. Everyone and everything was more violent back then. Same reason people share that statistic about ~2% or whatever of people alive today being descended from Genghis Khan as if it reflects impressively on Genghis. Time + distance makes being all-time world champion of rape somehow worthy of jokes instead of horror.

            Bill & Ted picked up both Napolean and Genghis Khan for a wacky adventure at the San Dimas mall. Not Hitler and Stalin though.

          • Protagoras says:

            I think it’s also partially an American thing; I get the impression Napoleon had a considerably worse reputation in Europe until considerably more recently, though the passage of time is also certainly a huge factor.

          • Aapje says:

            I don’t see how Napoleon was a mass murderer in the same sense as Stalin or Hitler. Napoleon was very prone to go to war and his wars caused deaths, of course. However, Napoleon didn’t seem to go out of his way to cause more deaths than necessary to achieve his military aims.

            When The Netherlands was under Napoleonic rule, we got some very useful reforms, including modernized law. On the other hand, Napoleon wanted to treat The Netherlands as a colony to be exploited for the benefit of France (this caused a conflict with Louis Bonaparte, who was made King of The Netherlands by Napoleon, but who wanted to be a good ruler for the Dutch people, so Napoleon deposed him). Napoleon was defeated before he got his way, but even if he had done his worst, it seems that he at most would have been regarded as ‘bad ruler,’ not as a giant mass murderer.

          • Nancy Lebovitz says:

            War gets a lot of moral latitude, apparently even expansionist wars of choice– though that may only apply to wars before a certain date. I’m not sure how much Napoleon’s wars were defensive in the sense that he had a rational fear of being conquered if he didn’t attack first.

            As for good policies, he did have some.

            https://en.wikipedia.org/wiki/Napoleon_and_the_Jews

            His record about Jews is mixed.

            He lifted a lot of restrictions on Jews– and then imposed some very severe restrictions for three years.

        • Spookykou says:

          It seems reasonable to me that Americans(for example) would be more upset with Germans killing non-Germans*(a group Americans are in) than with Russians killing Russians.

          As for the people in the country in question not being upset about it, two possible explanations jump out at me. First, all other people in their country might not be their in-group/tribe, so if their out-group is dying, that is probably even better than their far group dying. Second, most of the communist countries in question have had gradual changes in government such that they are reluctant to be overly hard on ‘themselves’.

          *They also killed Germans.

          • onyomi says:

            Fascism for me, communism for thee?

            It makes a certain amount of sense, and is in keeping with my idea that fascism is outward-looking (explains Episcience’s perception that it is inherently bellicose, which I didn’t see as a central feature of fascism), while communism is inward-looking.

            That said, as an American, I’d be more afraid of the American government attempting to nationalize all industries than of the American government attempting to aggressively take over Mexico. I’m not saying one is objectively worse, just that the former is a bigger threat to me and people I know.

            That also said, I really do think the more recent upsurge in “Antifa” notoriety really mostly boils down to a calculation which goes: Nazis=Fascists, Nazis=Racists, Fascists=Racists, Trump=Racist (the Clinton campaign told me so), Trump=Fascist.*

            *Also note that I wouldn’t deny the Trump administration has some fascist features, but so, too, have all US administrations since our first fascist president, FDR.

          • Conrad Honcho says:

            @onyomi

            Is racism central to fascism? Obviously in antifa’s thought process it is, but I’m talking about reality here. I seem to recall Mussolini didn’t really understand what Hitler was getting all upset about the Jews over.

          • I seem to recall Mussolini didn’t really understand what Hitler was getting all upset about the Jews over.

            My understanding of Italian anti-semitic restrictions is that they didn’t apply to Jews who had a record as patriotic Italians, most obviously ones who had fought in WWI.

            I think a lot of people in Europe saw Jews as not entirely part of the nations they lived in. Chesterton discusses that in one essay. It’s an attitude I encountered traveling in Europe c. 1970–from European Jews.

            From that standpoint, anti-semitism ends up more nationalist than racist. Which fits the Italian pattern.

        • cassander says:

          >I mean, there are communist regimes with horrifying records and communist regimes with merely awful records; there are fascist regimes with horrifying records and fascist regimes with merely awful records.

          The ratios matter. the only communist regime that didn’t rack up a 6 digit death toll was cuba, and they had 5 digits. There were a few fascist regimes that didn’t get more than 5, and that out of a much smaller pool of countries. Communism was uniquely awful, even more so than fascism.

          >communist atrocities seem to target the strong, while fascist atrocities target the weak.

          Nuns in Spain? kulaks in russia? These were the strong? This is just rehashed “don’t punch down” nonsense. Everyone targets the weak, if you try to target the strong they’ll target you back.

          • onyomi says:

            I’m talking about perception and theory, not reality.

            In theory, fascism is about eliminating the weak, disorderly, heterodox elements from the national “body.” In theory, communism is Robin Hood.

          • Jiro says:

            the only communist regime that didn’t rack up a 6 digit death toll was cuba, and they had 5 digits.

            The Republic of San Marino had the Communist party as the main group in a coalition from 1945 to 1957. Admittedly they tried a coup at the end, but they didn’t kill five figures. Of course, San Marino’s population is only in the five figures anyway.

        • I think you may be missing the obvious explanation for the pattern. People map ideologies onto a simple left/right pattern. Communists are left, fascists are right, and the academy and media are dominated by people who identify left.

          I observe a similar pattern in myself. Le Pen is identified as right. Libertarians, such as me, are identified as right. So my emotional instinct is to think well of her despite the fact that most of her key policies are ones I strongly disagree with.

          • onyomi says:

            Yes, he mentions that possibility in the video. Communism is left, academy is left. What I’m wondering about is the degree to which that also might explain vastly differing perceptions of the Vietnam War (though probably the manner of our prosecution of the war, along with its ultimate failure are bigger factors).

            I agree with you about Le Pen. I’m inclined to like her because I always side with decentralization. But, from an American perspective, her domestic and economic policies look atrocious. I’m not sure how much of that is differences in what constitutes a “right-wing” economic policy in France versus the US. From what I understand, Macron would really be more congenial to most libertarians on that front.

          • Conrad Honcho says:

            Whenever I see one of those political compasses with parties/political figures marked, Hitler and the Nazi party are in the center on the left/right divide and all the way up on the authority/liberty axis.

            Hitler wanted “nationalism without capitalism, and socialism without internationalism.” He nationalized industries and instituted free healthcare, free education, public works projects, etc. These are not right wing policy goals.

          • onyomi says:

            “nationalism without capitalism, and socialism without internationalism.” He nationalized industries and instituted free healthcare, free education, public works projects, etc. These are not right wing policy goals.

            It’s very common for libertarians to argue that fascists are really just liberals/socialists/leftists in disguise because they still want socialism, deny individualism, etc.

            They sort of have a point, but what I think I, and other libertarians have to recognize, is the sad fact that most of the world doesn’t carve political views along the lines we think they should. Right=/=freedom and individualism.

            For practical purposes, both right and left as they exist in most places and at most times imply some level of collectivism (after all, in some sense, politics is, by definition, a collective endeavor; as an ancap, I’m not so much in favor of “individualist politics” as I am anti-politics), just of different kinds.

            For practical purposes there is no “liberty and individualism” end to the traditional political spectrum. The far left (communism) means collectivism with an internationalist, utopian, egalitarian bent and the far right (fascism) means collectivism with a nationalist, traditionalist, hierarchical bent.

          • Conrad Honcho says:

            @onyomi

            Individualism is great, but unless everyone else is an individualist the individualists will be out competed by anyone adopting a collectivist strategy.

          • onyomi says:

            @Conrad Honcho

            Individualism is great, but unless everyone else is an individualist the individualists will be out competed by anyone adopting a collectivist strategy.

            Adopting an individualist approach to political organization doesn’t necessitate an individualist approach to anything else (production, defense…).

            In fact, I’d say that, in most cases, an equal number of individuals voluntarily cooperating to accomplish x will be more effective than the same number working under compulsory political organization. The individualists have the problem of individuals being able to chose to “defect” from contributing to some common good, but, as David Friedman often points out, situations where individual rationality doesn’t lead to group rationality are the exception on the private market, but the rule on the political market.

          • cassander says:

            @onyomi says:

            In fact, I’d say that, in most cases, an equal number of individuals voluntarily cooperating to accomplish x will be more effective than the same number working under compulsory political organization.

            Depends on your definition of “more effective” and “voluntarily cooperating”. In general, yes, decentralized methods will arrive at more optimal solutions given unlimited time and resources. But sometime an a worse solution tomorrow produces a better result than a more optimal solution in year, with war being the classic example.

          • onyomi says:

            @Cassander

            In asserting that government solutions will be faster, if not always better, I don’t think you’re comparing apples to apples.

            If you compare an existing government with e.g. a standing army and a ministry of defense to a mass of heretofore unorganized individuals, then yes, the government will be able to respond to a military threat much faster.

            But that’s clearly stacking the deck in government’s favor. If you start with two unorganized groups of individuals and one is allowed to coercively force people to do things to respond to a threat and the other has to respond to the threat by means of voluntary cooperation, it seems far from obvious to me that the former method will be faster, given that the former group, like the latter group, still has to work out a mechanism for determining who will give orders and make decisions–voting, appointing a king, etc.

          • cassander says:

            @onyomi says:

            In asserting that government solutions will be faster, if not always better, I don’t think you’re comparing apples to apples.

            It’s more that in a certain set of circumstances, faster IS better to such a degree that the advantages of coercion outweigh the usual disadvantages.

            But that’s clearly stacking the deck in government’s favor. If you start with two unorganized groups of individuals and one is allowed to coercively force people to do things to respond to a threat and the other has to respond to the threat by means of voluntary cooperation, it seems far from obvious to me that the former method will be faster, given that the former group, like the latter group, still has to work out a mechanism for determining who will give orders and make decisions–voting, appointing a king, etc.

            for your non-coercive system to be viable though, it has to be able to be able to respond to threats organized by coercive actors, however they arise. this to me is the achilles heel of anarchism, it creates a massive prisoner’s dilemma where everyone is only better off as long as no one defects (starts coercing) but someone always will.

          • Conrad Honcho says:

            @onyomi:

            In fact, I’d say that, in most cases, an equal number of individuals voluntarily cooperating to accomplish x will be more effective than the same number working under compulsory political organization.

            What part of “adopt a collectivist strategy” implies “compulsory political organization?”

            If you have 10 people competing in an every man for himself system, the two who team up will beat the other 8.

            Alternatively, if libertarianism is so strong, why are there so few libertarians and why does libertarianism have essentially no political success?

          • onyomi says:

            @Conrad Honcho

            Okay, if by “adopt a collectivist strategy” you merely mean “cooperate,” then sure, the people working together will do better than the same number of people, each working on their own.

            But I am not comparing a collectivist approach to e.g. defense to an “every man for himself” approach. I am comparing an individualist approach to politics (“anti-politics”) to a collectivist, compulsory form of politics as the meta-level mode of human organization for achieving other goals, like defense.

            The question isn’t “is working together better than working alone?” It’s “is organizing people under threat of violence/coercion more effective than organizing people voluntarily?” I think the answer is “usually not.”

            Re. Cassander’s contention that a group of people who won’t use coercion will be outcompeted by those who “defect” and do use coercion, I also don’t think it follows, because coercion is a tool without its own costs. Government solutions crowd out and often explicitly ban competing private solutions. You’re not going to get an equally strong voluntarily-funded, volunteer army alongside your tax-funded, drafted army as you would if the drafted army didn’t exist.

            Re. why if, libertarianism’s so great, it hasn’t had much success, several points:
            1. Libertarianism, as I said, is, in some sense, a kind of “anti-politics” as opposed to a competing theory of political organization.

            2. If it isn’t that successful electorally it doesn’t necessarily reflect on its overall desirability because democratic elections produce diffuse costs and concentrated benefits while libertarianism is generally offering the opposite.

            3. As a political philosophy libertarianism is really very new and actually has enjoyed a fair amount of success if you consider that e.g. Reagan might never have won if not for the groundwork by people like Goldwater, Friedman, Rand, et al.

            4. Even if libertarianism as an explicit political philosophy known by that name hasn’t yet enjoyed a lot of success, there is a sense in which it has already enjoyed tremendous success–indeed, may be responsible for our incredible wealth today compared to pre-industrial revolution. The ideas of Locke, for example, highly influential among the US founders, were a good deal more libertarian than most earlier political thought (though see also School of Salamanca).

            The success of libertarianism, then, is not best measured by the number of libertarian politicians who get elected, or the number of libertarian laws which get past, but by the extent to which people succeed in keeping things out of the purview of coercive collective action. The Bill of Rights, for example, is arguably a very libertarian document, as it’s all about what the government may not do in the name of the collective, as opposed to most laws, which are about what the individual citizens may not do.

          • cassander says:

            @onyomi

            Re. Cassander’s contention that a group of people who won’t use coercion will be outcompeted by those who “defect” and do use coercion, I also don’t think it follows, because coercion is a tool without its own costs. Government solutions crowd out and often explicitly ban competing private solutions. You’re not going to get an equally strong voluntarily-funded, volunteer army alongside your tax-funded, drafted army as you would if the drafted army didn’t exist.

            My point is that the defectors don’t have to be as strong in the abstract or long run. they just have to be locally stronger for as much time as it takes them to beat you up and take your wallet.

      • Conrad Honcho says:

        But it’s just as impossible to implement communism without violence. It’s called “seizing the means of production,” not “having the means of production freely given to you without coercion.”

    • herbert herberson says:

      I’m not really interested in debating this point further, and I know most people here will disagree with me, (and are welcome to do so at length in the replies without my participation) but I offer it up nonetheless because it’s a relevant component of the full answer and I’m not sure anyone else will supply it: the crimes of communism are far less incontrovertible than those of Nazism. Neither Mao nor Stalin launched invasions of entire continents apropos of nothing. Neither of them ran full-blown death camps (prison and labor camps, sure, but no gas chambers). The vast majority of deaths attributed to both of them are from famines, and famines were not unheard of in either of their countries prior to their rule (although they were after!). The calculation of numbers for those famines is much shakier and more variable. In Mao’s case, there is no question that the famines were unintentional, and in Stalin’s case the level of intention is an open question. It is possible to argue that both of them left their countries better off than when they started.

      Even if one looks at all those ambiguities and sorts them out and decides that all the accusations against communism are true, that fact remains that you have to do some work to get there. That’s simply not true of Nazism’s crimes, which are simple, stark, and obvious.

      • Trofim_Lysenko says:

        Unlike the Nazis, many Communist regimes attempted to obfuscate the nature of their camps. For example, the Gulag system killed vast numbers of Russians, but since they died of combination of exposure, starvation, and overwork it was possible to administratively “release” them just before or just after their deaths, keeping the numbers down.
        Interestingly while this was common knowledge among Russian critics of the Soviet system, and is still remarked on today by a few scholars, it appears to have vanished out of general education on the subject in favor of the narrative “Everyone SAYS the Gulag system was awful, but really it was just bad jail! Why, just look at these official Soviet statistics.

        Or there are the re-education camps of Vietnam:

        A common form of punishment is confinement to the CONEX boxes—air-freight containers that were left behind by the United States in 1975. The boxes vary in size; some are made of wood and others of metal. In a CONEX box 4 feet high and 4 feet wide, for example, several prisoners would be confined with their feet shackled, and allowed only one bowl of rice and water a day. “It reminded me of the pictures I saw of Nazi camp inmates after World War II,” said a physician we interviewed who witnessed the release of four prisoners who had been confined to a CONEX box for one month. None of them survived.

        I will grant that the Nazis killed more Western Europeans in a shorter -timespan- than any of the various Communist regimes, and unlike the Communists made very little effort to conceal what they were doing (if I remember correctly under the assumption that there would be plenty of time to destroy records after the fact if need be, once they won).

        However, that just means that we have been willing to overlook or undervalue deaths of people geographically and culturally more distant to us, not that those deaths were any less severe. Add to that that for the past 25-30 years there has been a sharp decrease of interest in pursuing scholarship on these issues, and prior to that information was both limited due to the cold war and you had active elements within academia going to bat as apologists for communist regimes.

        • Aapje says:

          Unlike the Nazis, many Communist regimes attempted to obfuscate the nature of their camps.

          What?

          The Nazis worked very hard to keep it a secret, including razing camps at the end of the war.

          They just failed more than the communists, probably because they lost, while Stalin & Mao remained in power and thus could hide the evidence more effectively.

          • Trofim_Lysenko says:

            It’s entirely possible I’m underinformed, but my impression was that the attempts to raze the camps and the majority of coverup efforts weren’t undertaken until very late in the war when it finally became impossible to ignore that they were losing, and that as such they were very slapdash, last-minute improvisations.

          • dndnrsn says:

            @Trofim_Lysenko

            Sonderaktion 1005 began at a point when the Germans still looked like they were winning in the East – the crisis of the previous winter had passed, and again they were making significant advances. The sense that the Germans were losing only really started in mid-to-late 1943, after the disaster at Stalingrad over the winter, and after the abortive offensive at Kursk failed to succeed as Barbarossa and Blue had.

            EDIT: Additionally, the concentration camps were not kept secret, not remotely. The death camps, and the existence of facilities for killing at concentration camps and dual-purpose camps (eg Auschwitz was a concentration camp with an attached death camp; those deemed capable of work were sent to the concentration camp and worked to death, while those not were gassed upon arrival). But the existence of concentration camps was well-known, and was used as a threat towards malcontents, etc.

          • Trofim_Lysenko says:

            Hmm, thanks for the information. I was under the impression that the Nazis didn’t really do much if anything to attempt to conceal their activities (at least from posterity/the world) until mid-to-late ’43.

          • Aapje says:

            The concealment efforts happened at all stages of the Holocaust. A major reason why the death camps were far away in Eastern Europe (and even there in wooded rural areas, away from population centers) was to keep it secret from Westerners/Germans.

            The programs themselves were described with lies that hid what actually happened. Aktion T4 (involuntary euthanasia) was described as “mercy death” for patients “deemed incurably sick, after most critical medical examination.” Parents were told that their children were being sent to special sections for children, where they would receive improved treatment. The children sent to these centres were kept for “assessment” for a few weeks and then killed by injection of toxic chemicals, typically phenol; their deaths were recorded as pneumonia.

            The organization that ran Aktion T4 was called the “Charitable Foundation for Cure and Institutional Care.” Of course, they didn’t give any care or cure anyone.

            The extermination of Jews was referred to as the ‘final solution.’

            When reports about the death camps began to emerge at the end of 1943, the Nazis decided to present Theresienstadt to an investigative commission of the International Red Cross. In preparation for the commission’s visit more deportations to Auschwitz were carried out in order to reduce the overcrowding in the ghetto. Fake stores, a coffee house, bank, school, kindergartens and the like were opened and flower gardens were planted throughout the ghetto. The commission arrived in the ghetto on June 23, 1944. Their meetings with prisoners were meticulously planned beforehand. After the visit the Nazis produced a propaganda film about the new life of the Jews under the auspices of the Third Reich. After finishing filming, most of the actors in the film, including almost all of the independent leadership and most of the children in the ghetto, were sent to the gas chambers of Auschwitz-Birkenau.

    • J Mann says:

      When I talk to people who defend communism relative to fascism, they seem to be particularly motivated by:

      1) the belief that communists typically mean well – that reducing inequality and oppression is a worthwhile goal, even if it often ends in horror and misery.

      2) the belief that horror and misery are not either a (a) necessary or (b) inevitable feature of communism. The idea here is that fascism’s essential aspects (which are poorly defined, but probably cluster around nationalism, militarism, and a cult of personality) are themselves bad, but communism’s essential aspects (sharing, albeit forced sharing) are arguably good.

      I’d also add that fascism since almost no one in the US identifies as fascist, so there’s no one in my circle inclined to defend it. Fascism is associated with mid-20th century German, Italian and possibly Soviet nationalism, so if you’re an American nationalist, you probably just say “I’m not fascist” rather than “Fascism has some good things, although those specific fascist states were indefensible,” the way a communist apologist might.

      • the belief that communists typically mean well – that reducing inequality and oppression is a worthwhile goal, even if it often ends in horror and misery.

        At a considerable tangent …

        Various people, discussing the question of racial differences in the distribution of characteristics, argue that the reason not to look at them is that belief in such things has done enormous damage in the past. Nobody seems to argue that belief in the desirability of policies to eliminate inequality should be shunned because it has done enormous damage in the past. But the connection of such beliefs to deaths from the French Revolution on, most obviously under 20th century Communism, looks a lot closer than the connection of beliefs in racial differences to the history of slavery.

        • nimim.k.m. says:

          Well, it actually boils down to looking at the actual details of the proposals, what the policies are and how they are supposed to be implemented.

          First thing you notice, that after the collapse of Soviet block, the traditional branch of leftism that called itself communism and advocated for introducing a Soviet-style system with Soviet-style “revolutions” guided by the vanguard party … for decades it has been practically in the graveyard mostly about everywhere and especially in the West.

          The parties that called themselves “Communist Party of Syldavia” (or local equivalent) have either dissolved or changed their brand, name, and much of the stated ideology on how the socialist utopia should pursued (often the new name is some vague-sounding variation of “The Left”). Usually in a way that you never ever hear anyone using the word “communism”. At least here, the younger generations of my age generally simply have no idea that certain older politicians would have been card-carrying communists / pro-East Germany in past. Even social democrats don’t talk that much about socialism.

          Heck, outside the certain area of Republican Spain anarcho-socialists until the 1930s, anarchosocialists used to be fairly obscure, and they probably are fairly obscure today (because nobody who does not involve themselves in extreme leftist activism, i.e. majority of mainstream, actually cares about the colors of the flags and which way the red and black areas are divided the anarchists might wave if they were not throwing rocks), and still my guess is that they are more numerous that the Communist Party of Syldavia style communists today.

          (Venezuela might be your weird example of special circumstances, but I believe even their stated ideology is a weird amalgamation of “Chavez likes this but not that”. Other small countries with civil wars and such might apply.)

          [Additional point. Back in Europe: if one looks at how the various leftists complain about globalization and EU’s free trade and free movement and what those things will do to workers’ jobs …. or race to bottom regarding various taxes and controls of finance markets are going to do how the global capitalism manifests on national scale, and how they complained when those changes happened ~25 – 15 years ago despite their protests … it’s fairly clear that they also have not been successful at making anybody to listen to their ideas. And then for some very weird reasons you have populist right becoming popular making those exact same points. ]

          Then we look at the policy proposals intertwined with any kind of “taking genetic racial differences into account”. The milder form I’d imagine is “arguing for less of affirmative action and deducing there’s structural racism every time academic department does not have enough PoC people”. But that can be often said without actually making too much fuss, if you are very careful about how and what you say. And then someone will com to point how IQ has been highly volatile over the last century, and the discussion might have change playing like it does here on SSC. But if you make your main point sound like it’s of utmost importance to remember how there’s categorical differences between qualities between races. In the sense, that’s the summary and punchline people remember.

          Weeeeeell. That really does not scream “slavery tomorrow”, but that does sound quite a bit like how apartheid and similar systems were justified by their proponents. That all kinds of separation is just natural because the differences are hard and natural, too, and easily inferable from stuff like skin color and shape of eyes.

    • Civilis says:

      As much as I despise Communism, I think the single biggest reason Communism isn’t treated the same as Naziism or even Fascism is that the Communists were the ‘good guys’ in the Second World War, whereas Nazis, Fascists, and even small-f fascists (nationalist Spain) have never been real US allies.

      I think the best example I can give is the following poster: http://www.fortmissoulamuseum.org/WWII/detail.php?id=514 Think about it; we’ll come back to it.

      The more I read about the Second World War, the more respect I can give to the Soviet military of the time. Yes, they did field the most troops and experience the most losses of the war. Depictions of the Soviets throwing men away tend to at least annoy me. On the other hand, World War II in Europe started in part because the Soviets divided up Poland with the Nazis, behaved equally aggressively and monstrously before and after the war, and a lot of the hardships the Soviets experienced were as a direct result of the horrors caused by the Soviet regime before the war; purging competent generals and starving peasants. Almost all of the American and European communists of the pre-war period were at best unwitting dupes of the Soviets.

      Part of the US war effort was the propaganda campaign dedicated to depicting our cause as a noble one, and that meant papering over the bad parts of Soviet Communism because they were on our side. And the Soviets were at least partially willing to play along with that depiction, as long as we were on the same side.

      “This man is your Friend” is, to a degree true, but saying of the ‘Russian’ soldier “He fights for Freedom” is a lie. Ultimately, we’ve never been able to undo the lingering effects of that propaganda campaign and the good publicity the Soviets received for their part in bringing down Hitler. Those legitimatized the early Communists and made stigmatizing them for the later actions of the Communists nearly impossible.

    • dndnrsn says:

      So, this is a reason why fascism is worse than communism that goes a ways beyond what most people think about (I doubt people follow this train of thought; it’s mostly emotional for most people): the bad shit that fascists did is basically part of the system, whereas a decent chunk of communism’s dead is due to a combination of the vanguard party idea being a really bad one, and misfortune/bad handling of misfortune/incompetence.

      The fascists in Italy made their commitment to building the empire they felt Italy had been denied pretty clear, and then tried to do that. However, most of the really bad stuff is the Nazis (the Italians in Africa did not behave much worse than any other colonial power did, and if it hadn’t been for the Nazis starting WWII and Mussolini deciding to get into it, Mussolini would likely never have been more than a standard-issue right-wing authoritarian dictator; compare Mussolini to Franco to see what I mean), and the bad stuff the Nazis did is both unprecedented in intensity (eg, they did most of their murdering and starvation in a very short period of time; the vast majority of their victims died in the less than 4 year period between Barbarossa and the fall of Berlin; Communism killed more people than National Socialism but it did so over a longer time scale) and was basically inevitable from the start.

      One of the central parts of Nazi ideology was to conquer eastwards to claim agricultural land for Germany and reducing the natives to helot status, more or less. The Germans made a pretty major attempt at eliminating the Polish intelligentsia, and their own calculations concluded that doing what they wanted in the east following the predicted fall of the USSR would result in the deaths of about 30 million people. They made close to zero attempt to feed Red Army POWs (of whom millions died) and close to zero attempts to feed civilians in the east not considered important to the war effort. And this is even before you get to the extermination of the Jews: while the most convincing historical explanation is that the original intent had been to deport the Jews to somewhere like Madagascar (made impossible by war conditions) or east of the Urals following the USSR’s defeat (ditto), and it was only after Barbarossa did not result in the defeat of the USSR before the fall that the einsatzgruppen began shooting civilians en masse regardless of sex or age, and death camps were built to eliminate Polish (and then Western European and other Eastern European) Jews – even if the USSR had collapsed as predicted, shipping millions of people east of the Urals would have resulted in a huge number of deaths. Hitler’s eight-figure body count (I believe Snyder puts it at 12 million, but you could probably expand that by a few million) was largely intentional.

      Meanwhile, the death toll of communism (leaving aside the Khmer Rouge, who in my opinion are as bad as the Nazis) is due to repressive actions by a vanguard who were supposed to hand over power to the people at some point rather predictably deciding they’d rather not, and due to a potent combination of misfortune and incompetence. Snyder’s estimates of Stalin’s death toll is something like 9 million (a low estimate, and you could expand that without having to use high estimates) but those 9 million are far more the result of intentional or semi-intentional famines, famines due to incompetence or bad conditions compounded by incompetence, etc. The Great Leap Forward, by sources sympathetic to Mao (as in, promulgated by actual Maoists), caused the death of 15 million people – but was largely due to extremely bad decision-making. (I don’t believe the ~100 million death toll one sees bandied about for communism – it is probably closer to 30 or 40). A far lower % of the communist death toll is by intentional murder.

      I can see a possibility for communism where a leadership elite does not become repressive to hold on to power, and where incompetence does not lead to piles of corpses. I can’t see fascism, let alone national socialism, an ideology built on “let’s go knock over other people and take their stuff and let them starve” and getting worse from there, producing anything good.

      • HeelBearCub says:

        I agree with all this, but I still think it ignores the fact that the Allies militarily defeated the Nazis. The Nazis no longer existed (essentially) to retcon their own history, and the Allies have many incentives to investigate and promulgate a full exploration of their crimes.

        Are there examples of Communist countries that look map well to defeated Germany?

        • Nancy Lebovitz says:

          One heartening thing is that there’s been some fading of fondness for communist governments. I’m too young to remember when the USSR became a great disappointment, but Tiannamen Square caused a lot of leftists to like China much less.

          I look forward to enough changing in Cuba that more records of how bad it was come out. I’m not expecting a huge number of concealed murders, but the amount of political prisoners and poverty might have some effect. There’s more to life than vintage cars, good hurrican preparation, and (disputed) good medical care.

          • multiheaded says:

            You do realize that many of the Tiannamen protesters were far-left and critical of the capitalist reforms, right?

        • dndnrsn says:

          The West during the Cold War had an incentive to make the Soviets and Chinese look as bad as possible, didn’t they?

          • HeelBearCub says:

            @dndnrsn:
            Well, sure, and vice-versa. But that ends up in a two-sided conflict of flung accusations, not a one-sided litany of crimes.

            We also didn’t put an occupying force into their countries while the deaths were on-going. Film of concentration camps shot by allies and immediately available became part of the narrative for why we fought the war and how good and wonderful we were to do so. Even when the Soviet Union broke up, it was primarily defeated internally.

          • dndnrsn says:

            True. Probably the most Nazi-like, both in terms of crimes and how they were uncovered, was the Khmer Rouge? After all, they more or less got run out of town by the Vietnamese.

          • HeelBearCub says:

            Right. The Khmer-Rouge were defeated … by another Communist country!

            Hard to get a narrative of the intrinsic horrors of Communism from that.

            Come to think of it, isn’t Vietnam a good counter-example to “Communist countries are horrible”? (Yes, I don’t think there are really Communist anymore, but that isn’t really the point. Neither is Yemen or Syria or Iraq or …)

          • dndnrsn says:

            They were defeated enough that what they’d done came to light pretty clearly, and even the apologists shut up pretty quickly.

            I have no doubt you’re right and had the Germans won in the East in late 1942 or 1943 (so after they started murdering civilians in large numbers), presumably causing the Western Allies to sue for peace, things would be seen differently.

          • HeelBearCub says:

            @dndnrsn:

            They were defeated enough that what they’d done came to light pretty clearly, and even the apologists shut up pretty quickly.

            Sure, but the narrative from the victorious Vietnamese won’t be “see how awful Communism is?”” That’s all I meant.

          • dndnrsn says:

            Ah, yeah, in that case you’re right. But they did publicize what the Khmer Rouge had done more than anyone has publicized their own atrocities – not to say communism is bad, but to point out that they were the good guys in this situation.

          • HeelBearCub says:

            @dndnrsn:
            Exactly.

            As a thought exercise, had the Khmer Rouge been capitalists (say fascist capitalists), I think Vietnam would have attributed their evil to capitalism (not fascism).

          • Trofim_Lysenko says:

            @HBC

            Hundreds of thousands of civilians worked to death in reeducation camps, systematic extermination of refugee civilians during and immediately after after re-unification, ethnic cleansing….

            …yeah, no, Vietnam is NOT a good counter-example to “Communist countries are horrible”. That they only killed hundreds of thousands (lowball) to 1-2 million (high estimate) is not exactly a ringing endorsement.

      • Gobbobobble says:

        Not to detract from your main point, but re: Poland – The Soviets very much did this too and arguably willfully abetted the Nazis.

        This doesn’t contest that the Nazis had a higher death toll overall, but let’s not forget that the whole Molotov-Ribbentrop scheme and its aftermath was a complete spitroast with both parties deliberately doing absolutely abominable things to the Polish people.

        • Le Maistre Chat says:

          But people DID forget that, because the USSR was never defeated and occupied. The USSR was able to retcon all atrocities against the Polish people as Nazi atrocities until the Union collapsed and Yeltsin’s Russia opened the archives.

        • dndnrsn says:

          The way the Soviets behaved, which was essentially like any imperial power, is not inherent in communism as a theory in the way the way Nazi Germany behaved is inherent in national socialist “theory”, insofar as it has a theory (which, compared to communism, it doesn’t).

          • Gobbobobble says:

            Sure. As I said, not arguing your points. Just sayin’ it’s rather remiss to not at least include a “(the Soviets did this too, but it’s not the same [for all those reasons you just mentioned])” when discussing “pretty major attempt[s] at eliminating the Polish intelligentsia”. Like LMC said, the Soviets were able to hide their crimes, deliberate crimes, due to being on the winning team. It’s important that that record be corrected.

          • dndnrsn says:

            OK, that’s a legitimate point.

          • cassander says:

            >The way the Soviets behaved, which was essentially like any imperial power, is not inherent in communism as a theory in the way the way Nazi Germany behaved is inherent in national socialist “theory”, insofar as it has a theory (which, compared to communism, it doesn’t).

            Nazi theory required the murder of tens of millions of slav and jews in order to build a glorious german utopia

            Communist theory required the murder of tens of millions of bourgeoisie and capitalists in order to build a glorious communist utopia.

            Murder is equally inherent to both systems, they just chose different targets.

            As for the assertion that all great powers murder, this is true. But some murder more than others, communist run states are invariably on the high end of the scale. Degree matters.

          • dndnrsn says:

            @cassander:

            Communist theory required the murder of tens of millions of bourgeoisie and capitalists in order to build a glorious communist utopia.

            Source? Most modern communists say this is not necessary, and few communists were saying it prior to taking power. That’s how it ended up (perhaps murder of millions instead of tens of millions, though) but practice is not theory.

          • dndnrsn says:

            @cassander:

            Does that establish tens of millions, or even millions of deaths? The most reasonable precedent Marx and Engels would have been thinking of was the Reign of Terror in revolutionary France, which killed considerably fewer than a million people. What the Soviets did was a novelty.

          • cassander says:

            @dndnrsn

            They don’t provide a list, but the terror, by definition, is mass killing on a large scale of suspect groups and sympathizers. As you say, they are explicitly invoking the french revolution. The target, as near as can be ascertained, is “the bourgeoisie” and their supporters, which is most definitely tens of millions. The proportion of the population that was an aristocrat in 1789 was far smaller both in absolute terms and as a share of the total population. They aren’t calling for just another french revolution, but a french revolution turned up to 11.

          • dndnrsn says:

            @cassander

            But even in the USSR, the bourgeois were not the bulk of the dead. Those who died in the millions died because they were seen as a potential fifth column (Ukrainians) or because they were in the way of collective farming and so forth (kulaks). The chances of a white-collar person with an education living out Stalin’s rule are probably better than some farmer with a bit of land and a couple cows.

          • cassander says:

            @dndnrsn says:

            But even in the USSR, the bourgeois were not the bulk of the dead. Those who died in the millions died because they were seen as a potential fifth column (Ukrainians) or because they were in the way of collective farming and so forth (kulaks). The chances of a white-collar person with an education living out Stalin’s rule are probably better than some farmer with a bit of land and a couple cows.

            No, there weren’t enough of them to make good enemies, so kulaks were attacked instead. But they fit in a similar economic location relative to the society that actually existed in russia at the time. Rich enough to have something to lose if the revolution and to earn the envy of their neighbors.

          • The chances of a white-collar person with an education living out Stalin’s rule are probably better than some farmer with a bit of land and a couple cows.

            I’m not at all sure that is correct.

            I don’t have data, but I have read Conspiracy of Silence, which is an account of the Great Purge by someone who went through it and survived. My impression from that was that the victims were largely educated, middle class sorts. Since such people must have been a fairly small fraction of the population, my guess is that quite a large fraction of them got arrested and either executed or sent to prison camps.

          • dndnrsn says:

            @cassander

            The kulaks as enemies wouldn’t be predicted by Marxist theory, though, would it? They were deemed enemies entirely because they were inconvenient. I’m not sure if “rich enough to inspire envy” is a good definition of the bourgeois’ role. Plus, the kulaks didn’t lose from the revolution – they lost from the push to collective farming, which didn’t happen until some time after the revolution.

            @DavidFriedman

            I don’t have data, but I have read Conspiracy of Silence, which is an account of the Great Purge by someone who went through it and survived. My impression from that was that the victims were largely educated, middle class sorts. Since such people must have been a fairly small fraction of the population, my guess is that quite a large fraction of them got arrested and either executed or sent to prison camps.

            By the 1930s, the parts of the USSR where most of the dying happened were more educated and industrialized than is sometimes thought. The majority of the dead were in the countryside (or Ukraine) due to (sometimes intentional) starvation.

      • Civilis says:

        Franco’s Spain, basically regarded as the third ‘Fascist’ power, didn’t have an ideology built on “let’s go knock over other people and take their stuff and let them starve”. Gobbobobble already pointed out the example of the Soviet seizure of Eastern Poland; let’s not forget the Baltic States or Finland (although that one didn’t go quite to plan).

        The fact that basically every communist revolution ends in the original vanguard holding on to power rather than handing it over to the people suggests that there’s something in the whole approach that strongly favors that end. I think it’s an element of the sunk cost fallacy; when you had to have a violent revolution to seize power to bring about your utopia and your utopia doesn’t appear, a few more years of tyranny won’t make you that much worse. Where as if you give up, all those deaths were for nothing.

        • dndnrsn says:

          Franco was a garden-variety right-wing authoritarian dictator. It’s debatable whether he was properly fascist.

          As for the vanguard party, it’s proven to be an idea that doesn’t work.

          • Civilis says:

            Ok, the dictionary definition of fascism is “an authoritarian and nationalistic right-wing system of government and social organization“. I’m willing to accept an alternate definition, but I certainly consider it highly reasonable that the Antifa referred to in the OP for this thread would consider Franco fascist. If your only example of Naziism is Nazi Germany and your only example of proper Fascism is Fascist Italy, it’s hard to talk about generic fascism, especially in the context of today’s politics.

          • cassander says:

            It’s the reverse. Hitler’s murder fetish is the outlier among fascists. Franco’s fascism is much more in keeping with doctrinal fascism than Hitlerism. And I say hitlerism not Nazism because there were more doctrinaire fascist nazis, they just got killed.

          • Nornagest says:

            Conversations like this never go anywhere, because you can basically define the word as whatever you want. The whole world was riding a wave of authoritarian nationalism in the Thirties, and the trappings of fascism were widely adopted; Mussolini’s Italy just gets a lot of academic attention because (a) it came first, and (b) it probably developed the “theory of fascism” more than anywhere else (though that isn’t saying much).

            Nazism might be a little closer to Italian fascism than to Imperial Japan or Falangist Spain, but all those systems are really united more by aesthetics than by structure. And even the US under Roosevelt took on a lot of those aesthetics.

      • cassander says:

        a combination of the vanguard party idea being a really bad one,

        A vanguard party is inherent to communist theory. And yes, I know that Lenin invented the term, but marx spends his entire life trying to build up a a vanguard party, he just never calls it that.

        and misfortune/bad handling of misfortune/incompetence.

        Communism was universally terrible to an extreme degree. if it is just incompetence, then communism as an ideology attracts only the incompetent, in which case it’s just as bad as if it was malevolent.

        Communism killed more people than National Socialism but it did so over a longer time scale)

        Stalin’s death toll from 28-40 definitely exceeds that of Hitler’s from 33-45. And if you want to complain that most of Hitler’s killing was from 42-45, that’s true, but he couldn’t have gotten there without the previous 10 years, just like Stalin needed a few years for his killing to get going.

        and their own calculations concluded that doing what they wanted in the east following the predicted fall of the USSR would result in the deaths of about 30 million people. They made close to zero attempt to feed Red Army POWs (of whom millions died) and close to zero attempts to feed civilians in the east not considered important to the war effort.

        Their calculations changed repeatedly over the course of the war, it’s hardly fair to judge them by the worst of them and the soviets by the best.

        Meanwhile, the death toll of communism (leaving aside the Khmer Rouge, who in my opinion are as bad as the Nazis) is due to repressive actions by a vanguard who were supposed to hand over power to the people at some point rather predictably deciding they’d rather not, and due to a potent combination of misfortune and incompetence. Snyder’s estimates of Stalin’s death toll is something like 9 million (a low estimate, and you could expand that without having to use high estimates) ,

        9 million is insultingly low. 9 million is only a slightly high figure for the just holdomor, to say nothing of stalin’s other crimes.

        >but those 9 million are far more the result of intentional or semi-intentional famines

        How is Stalin starving people to death less morally problematic than Hitler starving people to death? In both cases, exactly the same calculation was made. People that could have been fed were denied food in pursuit of other goals.

        I don’t believe the ~100 million death toll one sees bandied about for communism – it is probably closer to 30 or 40).

        Again, this is far, far too low.

        I can’t see fascism, let alone national socialism, an ideology built on “let’s go knock over other people and take their stuff and let them starve” and getting worse from there, producing anything good.

        One, there are plenty of fascisms that did exactly that. Thailand, Spain, Portugal, Argentina. There are zero communists that didn’t commit mass murder against “counter revolutionaries”. Please consider the possibility that this “seeing” is not a result of empirical evidence, but ideological sympathy.

        • One issue here is what defines fascism. “an ideology built on “let’s go knock over other people and take their stuff and let them starve” doesn’t seem to me to describe any fascist regime, even Hitler’s–that was part of the program, but it wasn’t what the ideology was built on.

          Someone else offered “an authoritarian and nationalistic right-wing system of government and social organization,“ which leaves open what “right wing” means and doesn’t require invading other countries.

          I think of the standard economic definition as “a system in which the means of production are nominally owned privately but are controlled by the state.” I don’t know if any fascist regimes fully fit that, but my impression is that most of them involved shifts in that direction–as did, of course, the first New Deal.

          Anyone have a good definition that explains what regimes are viewed as fascist, beyond “any regime that said it was”?

          • cassander says:

            One issue here is what defines fascism. “an ideology built on “let’s go knock over other people and take their stuff and let them starve” doesn’t seem to me to describe any fascist regime, even Hitler’s–that was part of the program, but it wasn’t what the ideology was built on.

            Eh, for hitler I think you can make the case that knocking over the slavs and taking the better parts of russia was pretty core to the ideology. Not so much any of the other fascists though.

            I think of the standard economic definition as “a system in which the means of production are nominally owned privately but are controlled by the state.” I don’t know if any fascist regimes fully fit that, but my impression is that most of them involved shifts in that direction–as did, of course, the first New Deal.

            The trouble with that definition is that it arguably includes every developed country government, which, even if accurate, renders the term largely meaningless.

          • hlynkacg says:

            Well if you ask the fascists themselves, namely Mussolini, Franco, and to a lesser extent and Peron. The defining trait of fascism is the conceptualization of society/culture itself as a sort of “super-organism”. Individual lives don’t matter so long as the society prospers. Hitler may have “strayed from the pack” by taking this concept in an explicitly racial direction but the Nazis as a whole certainly embraced this model.

          • HeelBearCub says:

            I’m with hlynkacg here, I always thought that fascism nominally involves a militaristic view of the states where all individuals are ideally required to sacrifice for the good of the state and its glory. The individual should be subsumed in service to the state.

            Sure, in practice all authoritarian autocracies tend to look alike, but the difference between fascist ideology and Communist ideology is that the individual is theoretically supposed to be served by the Communist state.

          • I think of the standard economic definition as “a system in which the means of production are nominally owned privately but are controlled by the state.” I don’t know if any fascist regimes fully fit that, but my impression is that most of them involved shifts in that direction–as did, of course, the first New Deal.

            The trouble with that definition is that it arguably includes every developed country government, which, even if accurate, renders the term largely meaningless.

            That’s a result of trying to use terms such as “fascist” or “socialist” as binary categories, which they aren’t. The American public school system is a socialist institution–government ownership and control of the means of production. The U.S. is more socialist than it would be if everything else was the same but schooling was private, less socialist than it would be if agriculture was government run as well.

            Similarly, the first New Deal was a program that made the U.S. more fascist, in the economic sense.

            Both terms describe continuous variables, not things that a country either is or isn’t.

          • Paul Brinkley says:

            The way I’d always heard it was that a fascist state promoted the vision of the state as a single body, with the people as cells; the working class and military, its limbs; the trades, its organs; the government, its brain. Everyone working together as an entity of one mind. The premise being that the government, of course, had the body’s best interests as its own, though it recognized the need to slough off a few dead cells or sacrifice parts that the rest might live. This was necessary in a harsh world where resources were limited, enemies were many, and absolute cooperation was vital.

            Some communisms have behaved this way, but the essential bit was that philosophy of a single whole eclipsing whatever individual needs might be said to exist. What was good for the state was good for its individuals; it was more proper to stop thinking of individuals at all. Worrying about individuals was a waste of time, like thinking about every one of your cells.

        • dndnrsn says:

          @cassander

          A vanguard party is inherent to communist theory. And yes, I know that Lenin invented the term, but marx spends his entire life trying to build up a a vanguard party, he just never calls it that.

          There are communists who don’t theorize a vanguard party. They’re unrealistic, probably. You’re unlikely to get communism without a vanguard, and you’re unlikely to ever see the vanguard hand over power to the people. So, I’m not a communist.

          Communism was universally terrible to an extreme degree. if it is just incompetence, then communism as an ideology attracts only the incompetent, in which case it’s just as bad as if it was malevolent.

          You’ll notice I’m not arguing in favour of communism. It’s a system that’s resulted in a great deal of human suffering.

          Stalin’s death toll from 28-40 definitely exceeds that of Hitler’s from 33-45. And if you want to complain that most of Hitler’s killing was from 42-45, that’s true, but he couldn’t have gotten there without the previous 10 years, just like Stalin needed a few years for his killing to get going.

          Sources? I’m going by Snyder’s figures, and he’s hardly pro-Stalin. I believe the official commie line these days is that Snyder is the catspaw of Ukrainian nazis or something.

          Their calculations changed repeatedly over the course of the war, it’s hardly fair to judge them by the worst of them and the soviets by the best.

          OK, so, what’s the low estimate the Germans had for GPO? 20 million, something like that?

          9 million is insultingly low. 9 million is only a slightly high figure for the just holdomor, to say nothing of stalin’s other crimes.

          It’s Snyder’s figures. Feel free to provide your own. 12 million is low for Hitler, too, because you can blame the European war dead on him to a significant extent. And double the Ukrainians’ own numbers is not “slightly high”.

          How is Stalin starving people to death less morally problematic than Hitler starving people to death? In both cases, exactly the same calculation was made. People that could have been fed were denied food in pursuit of other goals.

          What % of the Soviet dead were starved to death, vs what % of the Nazi dead?

          One, there are plenty of fascisms that did exactly that. Thailand, Spain, Portugal, Argentina. There are zero communists that didn’t commit mass murder against “counter revolutionaries”. Please consider the possibility that this “seeing” is not a result of empirical evidence, but ideological sympathy.

          That’s a pretty wide net for “fascism”. Take Franco, for instance: it’s debatable whether he was a fascist, or an authoritarian conservative who found fascist support useful. I also find it an odd suggestion that I’m ideologically sympathetic to communism.

          • cassander says:

            Sources? I’m going by Snyder’s figures, and he’s hardly pro-Stalin. I believe the official commie line these days is that Snyder is the catspaw of Ukrainian nazis or something.

            Snyder’s figure is geographically and temporally constrained. It does not include the totality of Stalin’s crimes, to say nothing of the USSR as whole.

            OK, so, what’s the low estimate the Germans had for GPO? 20 million, something like that?

            It’s Snyder’s figures. Feel free to provide your own. 12 million is low for Hitler, too, because you can blame the European war dead on him to a significant extent. And double the Ukrainians’ own numbers is not “slightly high”.

            WW2 started when Stalin and Hitler decided to invade Poland mostly simultaneously. If you’re going to start counting war dead, it doesn’t all go on Hitler.

            What % of the Soviet dead were starved to death, vs what % of the Nazi dead?

            The Soviet figure was probably higher in percentage terms, but I fail to see how that matters. Dead is dead.

            That’s a pretty wide net for “fascism”. Take Franco, for instance: it’s debatable whether he was a fascist, or an authoritarian conservative who found fascist support useful. I also find it an odd suggestion that I’m ideologically sympathetic to communism.

            He called himself a fascist, he used explicitly fascist rhetoric, imagery,
            and political arguments
            , was called a fascists both by his ideological enemies (means relatively little) and other fascists (means a lot more). Ditto Peron, who said “Mussolini was the greatest man of our century, but he committed certain disastrous errors. I, who have the advantage of his precedent before me, shall follow in his footsteps but also avoid his errors.” It’s hard to get any more fascist than that.

          • dndnrsn says:

            @cassander

            Snyder’s figure is geographically and temporally constrained. It does not include the totality of Stalin’s crimes, to say nothing of the USSR as whole.

            I’m not taking the number from Bloodlands, but from Snyder in the NYRB.

            WW2 started when Stalin and Hitler decided to invade Poland mostly simultaneously. If you’re going to start counting war dead, it doesn’t all go on Hitler.

            Granted. It’s impossible to split up the blame accurately. I’m not sure if Snyder counts Red Army POWs, though.

            The Soviet figure was probably higher in percentage terms, but I fail to see how that matters. Dead is dead.

            In a court of law, there are different degrees of murder, negligent manslaughter, etc.

            He called himself a fascist, he used explicitly fascist rhetoric, imagery,
            and political arguments, was called a fascists both by his ideological enemies (means relatively little) and other fascists (means a lot more). Ditto Peron, who said “Mussolini was the greatest man of our century, but he committed certain disastrous errors. I, who have the advantage of his precedent before me, shall follow in his footsteps but also avoid his errors.” It’s hard to get any more fascist than that.

            And yet you have scholars who debate whether he was a fascist. He’s certainly a noncentral example.

          • cassander says:

            @dndnrsn

            In a court of law, there are different degrees of murder, negligent manslaughter, etc.

            There are, but in no court is locking someone up and deliberately not feeding them anything but first degree murder. Stalin took all the food out of the ukraine, then shot anyone who tried to leave. his culpability was identical to that of the German for Russian POWs.

            And yet you have scholars who debate whether he was a fascist.

            Such as?

            He’s certainly a noncentral example.

            That says much more about people’s perceptions than it does the reality or Franco’s ideology.

          • He’s certainly a noncentral example.

            (about Franco)

            Doesn’t that leave you with only two examples of what is supposed to be an ideology, and two very different ones?

          • Brad says:

            Doesn’t that leave you with only two examples of what is supposed to be an ideology, and two very different ones?

            What about people like Marinetti and De Ambris? Why are dictators the only choices?

          • Anon. says:

            Marinetti I’d say is a highly non-central example. While he was a “small-f fascist”, his relationship with the actual Fascist regime was not very warm at all. He was more of a pie in the sky avant garde type, and in terms of real-world policy completely irrelevant.

            In any case, he’s a fun writer.

            The oldest among us are not yet thirty years old: we have therefore at least ten years to accomplish our task. When we are forty let younger and stronger men than we throw us in the waste paper basket like useless manuscripts! They will come against us from afar, leaping on the light cadence of their first poems, clutching the air with their predatory fingers and sniffing at the gates of the academies the good scent of our decaying spirits, already promised to the catacombs of the libraries.

            But we shall not be there. They will find us at last one winter’s night in the depths of the country in a sad hangar echoing with the notes of the monotonous rain, crouched near our trembling aeroplanes, warming our hands at the wretched fire which our books of today will make when they flame gaily beneath the glittering flight of their pictures.

            They will crowd around us, panting with anguish and disappointment, and exasperated by our proud indefatigable courage, will hurl themselves forward to kill us, with all the more hatred as their hearts will be drunk with love and admiration for us. And strong healthy Injustice will shine radiantly from their eyes. For art can only be violence, cruelty, injustice.

            The oldest among us are not yet thirty, and yet we have already wasted treasures, treasures of strength, love, courage and keen will, hastily, deliriously, without thinking, with all our might, till we are out of breath.

            Look at us! We are not out of breath, our hearts are not in the least tired. For they are nourished by fire, hatred and speed! Does this surprise you? it is because you do not even remember being alive! Standing on the world’s summit, we launch once more our challenge to the stars!

          • dndnrsn says:

            @Cassander

            There are, but in no court is locking someone up and deliberately not feeding them anything but first degree murder. Stalin took all the food out of the ukraine, then shot anyone who tried to leave. his culpability was identical to that of the German for Russian POWs.

            Snyder notes that the Ukrainian famine was intentional. But take, say, the Great Leap Forward – Mao wasn’t trying to kill 15 million plus. It was a badly botched attempt to transition from an agrarian to an industrial economy.

            Such as?

            That says much more about people’s perceptions than it does the reality or Franco’s ideology.

            Wikipedia has some references; my knowledge is Nazis>Italy>Spain. Franco was a conservative and a monarchist. He was not a fascist going into the whole thing in the same way that Mussolini was or that Hitler was a national socialist.

            @DavidFriedman

            Doesn’t that leave you with only two examples of what is supposed to be an ideology, and two very different ones?

            Yeah, we have fewer examples of fascism than we do of communism, because they lost the war. Had they managed to knock out the USSR or do whatever to get the Brits/Brits and Americans to sue for peace, all those third-world resistance movements that were communists to get Soviet/Chinese funding and weapons would have been fascist to get German/ITalian funding and weapons.

          • cassander says:

            @dndnrsn says

            Snyder notes that the Ukrainian famine was intentional. But take, say, the Great Leap Forward – Mao wasn’t trying to kill 15 million plus. It was a badly botched attempt to transition from an agrarian to an industrial economy.

            No, but he did take their food, and then didn’t feed them. That he wasn’t trying to kill them is irrelevant to the discussion of intentionality. he took actions that would inevitably lead to them starving. and not just staving, many froze to death after he requisitioned their coats so he could export the cotton.

            Wikipedia has some references; my knowledge is Nazis>Italy>Spain. Franco was a conservative and a monarchist. He was not a fascist going into the whole thing in the same way that Mussolini was or that Hitler was a national socialist.

            that is inaccurate. he had an entire ideological system that was explicitly fascist and corporatist. He was, in many ways, a better fascist than hitler.

          • But take, say, the Great Leap Forward – Mao wasn’t trying to kill 15 million plus.

            I think that’s correct. But he continued with the policies after it became clear that people were dying, and, if I remember correctly, he purged members of the leadership who reported the famine.

            At a slight tangent … . One interesting difference between Stalin and Mao was that when Mao purged people, he didn’t kill them, he sent them off to be peasants or factory workers–with the result that they were available to be unpurged later if he needed them. I think Deng got purged twice, and survived to become the central figure of the post-Mao party.

          • dndnrsn says:

            @cassander

            No, but he did take their food, and then didn’t feed them. That he wasn’t trying to kill them is irrelevant to the discussion of intentionality. he took actions that would inevitably lead to them starving. and not just staving, many froze to death after he requisitioned their coats so he could export the cotton.

            I suppose the question is whether he had fooled himself into thinking it would work.

            that is inaccurate. he had an entire ideological system that was explicitly fascist and corporatist. He was, in many ways, a better fascist than hitler.

            The article in the first paragraph states “Under the leadership of Francisco Franco, many of the radical elements of Falangism considered to be fascist were diluted and it largely became an authoritarian, conservative ideology connected with Francoist Spain.”

            @DavidFriedman

            At a slight tangent … . One interesting difference between Stalin and Mao was that when Mao purged people, he didn’t kill them, he sent them off to be peasants or factory workers–with the result that they were available to be unpurged later if he needed them. I think Deng got purged twice, and survived to become the central figure of the post-Mao party.

            Not strictly true – there were figures who got purged under Stalin who weren’t killed. Rokossovsky is a good example – he was tortured and imprisoned, then pulled out of prison and put back in command, rising quite high afterwards.

          • cassander says:

            @dndnrsn says:

            I suppose the question is whether he had fooled himself into thinking it would work.

            Again, he didn’t ask them to give them the food, and then they did, and then everyone died. HE forcibly took all the food then hoped by magic they’d grow more. “Yes officer, I shot him several times, in cold blood. But I figured he’d be able to crawl to a hospital, so it wasn’t murder” is not a defense.

          • My understanding of the famine in the Great Leap Forward is that Mao had convinced himself that, by trying very hard, peasants could greatly increase yields. People lower down didn’t want to tell him they couldn’t do it so they lied about crop yield, with the result that the figures going to him showed about twice the actual harvest. So from his standpoint, he wasn’t taking all of their food away, just some of it.

            It’s only at the point when starvation is actually happening that the situation becomes clear, various people at the level below him try to persuade him to reverse course, and for a fair while he doesn’t.

          • onyomi says:

            The problem with intentionality, or lack thereof, as a mitigating factor, is that it can, at best, only partially exonerate individuals, not ideologies.

            For example, more people died as a result of the Great Leap Forward than the Holocaust. But Hitler ordered people to be killed in the case of the Holocaust, whereas Mao merely enacted policies which he should have realized, but didn’t, would result in mass starvation. This may be a good reason to say that Hitler, the individual, is more morally evil than Mao, the individual, but it says nothing about the good or evil of promoting fascism or communism in the future, once the reality of these tragedies has come out.

            The question then becomes, which of these two actions was more “central” to the respective ideologies? Are concentration camps and gas chambers a central feature of fascism? I don’t think Mussolini would have said so? It certainly seems possible to imagine and/or point to historical examples of fascism without genocide. Is collectivized farming a central feature of communism? Absolutely. Arguably the central feature.

            Thus, we may simultaneously say that Hitler was a more evil person than Mao, but communism is a more evil ideology than fascism, especially now that the record of attempting to implement it is well known.

          • Loquat says:

            To expand on David Friedman’s comment, Mao’s China had the problem that an idea’s political correctness was often more important than its actual correctness. People came up with stupid ideas, like that planting crops way more densely than usual would improve yields, and they’d get a few influential backers, and suddenly this is the new dogma and if you don’t believe it your loyalty is suspect. I recall reading an account of that time that claimed dishonest party hacks looking for promotions had basically assembled a few Potemkin fields to “demonstrate” super-dense planting in action by transplanting a bunch of healthy rice plants into the field shortly before the inspection, knowing they’d die but not until after the higher-ups had left.

          • Aapje says:

            @Loquat

            The paranoia underlying the communist ideology and resulting bad faith assumptions about those who fail at implementing communism surely plays a big role in pushing that kind of lying. I’ve seen pretty deceptive practices in business where the penalty for failure is substantially less than being sent to the Gulags.

    • sandoratthezoo says:

      I think that most of your responses are radically overthinking it. Communism doesn’t have the same bad name that Nazism does because Communism had a large number of domestic adherents that worked to raise its status, in a way that Nazism didn’t post-WWII.

      • JayT says:

        I think this is the closest to the truth. Also, it’s not just a matter of after WWII, even before and during the war there were plenty of high profile communist sympathizers, but very few fascists in America.

    • Nornagest says:

      It might be a little pat, but we kicked the Nazis’ asses. And as part of kicking their asses, we sent people to places like Dachau and Auschwitz and Belsen, who came back and told us what they’d seen. I can hardly think of a better way to prove to ourselves that we were in the right and they were in the wrong.

      We didn’t kick the Communists’ asses; we just fought a few inconclusive brushfire wars and then the ideology choked to death on its contradictions. The gulags were never liberated; they did what they were meant to do, and by the time Russian communism fell, the memories of their worst years were fifty years old and mostly belonged to former guards. (Aleksandr Solzhenitsyn is almost solely responsible for what reputation they do have.) The Chinese situation is about the same. Sure, the mountains of skulls were even bigger, but there’s really only one genocide in the Communist bloc that got the kind of quick and thorough exposure the Holocaust did: the Cambodian. And that was exposed mostly through other communist states.

    • cassander says:

      Also ironic, it seems like a lot of the opposition to Vietnam resulted from our attempt to fight the war in a manner that would be broadly more acceptable to the left (defensively), whereas we fought WWII in a way which seems more broadly right-wing (offensively).

      Vietnam was fought the way it was to (A) prevent another korean style chinese intervention, (B) prevent the russians, who only acquired a serious nuclear arsenal at about when vietnam was starting, from getting spooked, and (C) to avoid the expense of a Korea style mobilization out of fear it would be politically difficult/make LBJ’s great society impossible to implement.

      >At the end of the day, is it possible that the left hated Vietnam but loved WWII really just boils down to WWII being a war against what was perceived as a right-wing threat, while Vietnam was a war against a left-wing threat?

      there is something to this, but there’s more to it that just this. When nixon comes into office, vietnam had gone on for about as as long and had about as many casualties as the Korean war had when Eisenhower got into office, but the US domestic political response could hardly have been more different.

      • onyomi says:

        the US domestic political response could hardly have been more different

        Yes, but why is this? Just flower power, etc?

      • Nornagest says:

        who only acquired a serious nuclear arsenal at about when vietnam was starting

        This is misleading, at best. The Vietnam War existed in one form or another from about 1955, which is about when the Soviet nuclear arsenal was finding its legs, and could reasonably be seen as an extension of the First Indochina War that started in 1946.

        But the story of American involvement really only gets going in August 1964, with the Gulf of Tonkin Resolution. That’s two years after the Cuban Missile Crisis and seven years after the so-called “missile gap”.

        • cassander says:

          But the story of American involvement really only gets going in August 1964, with the Gulf of Tonkin Resolution. That’s two years after the Cuban Missile Crisis and seven years after the so-called “missile gap”.

          the missile gap was a lie. It wasn’t real, and kennedy knew that, but it was good politics. And the cuban missile crisis was resolved the way it was because the USSR did NOT have a credible nuclear arsenal capable of striking the US, and wouldn’t for a couple more years.

          • Nornagest says:

            Yeah, but that wasn’t the question. The question was whether the Soviets had a credible nuclear arsenal then, and if the missile gap wasn’t laughed off the stage in 1957, it was clearly credible at that point. There was no gap, the reports were wrong, but you didn’t ask for a superior nuclear arsenal.

            Missile range is beside the point; everyone’s missiles sucked then, ICBMs existed (R-7, R-16, Atlas) but they were a pain in the ass and not widely deployed, and the air component of the nuclear triad was by far the most important one. As you note, that would change over the next few years.

          • cassander says:

            @Nornagest says:

            Yeah, but that wasn’t the question. The question was whether the Soviets had a credible nuclear arsenal then, and if the missile gap wasn’t laughed off the stage in 1957, it was clearly credible at that point. There was no gap, the reports were wrong, but you didn’t ask for a superior nuclear arsenal.

            I think you’re confusing what policy makers thought was credible and what the US population thought was credible.

            Policy makers on both sides knew that that, as you say, the missiles the Russians had at the time were largely useless. There’s even a typically bombastic quote from Khrushchev about the missiles in Cuba to the effect of “Now those damned Yankees will know what it’s like to be constantly threatened!” The soviets couldn’t directly threaten the US except with whatever was on Cuba, and both sides knew this. They also knew that US had enough bombers to threaten the Russians. Opinions differed about how many might get through, but they were a real threat. Because of this asymmetry, US policy makers could make an incredibly aggressive and uncompromising stance in order to force them to back down, and they did.

            The US population, however, did not know the missile gap was a lie. And they definitely didn’t know that missiles in Cuba would be strategically irrelevant in a couple years when the first real ICBMs started coming online in meaningful numbers.

        • bean says:

          It may have been 7 years after the reports that the missile gap was based on, but those reports were simply wrong, and the way the intelligence assessments were done sort of hid that. By 1960, we knew they were wrong, but Kennedy refused to drop it as a campaign issue. The Soviets didn’t overtake us in nuclear weapons until the late 60s/early 70s, after nearly a decade of McNamara doing everything he could to wreck the US military.

  18. dodrian says:

    Do you use a password manager?

    I realise that I reuse passwords much too often, and with more and more companies targeted (and demonstrating bad security) I’m starting to think that I need to take identity/account theft more seriously.

    I do most of my internet surfing on my home desktop (Linux Mint) or Android phone, but do occasionally find myself needing to log in to a web service on another machine. I like the idea of having it in ‘the cloud’, but I’m not sure what companies I’d trust to do that. I’d be willing to pay for a service if I was convinced it was security and privacy conscious.

    If you use a password manager, would you recommend it? What works and what’s annoying about it?

    • dodrian says:

      Immediately after posting I run across this article, demonstrating one of the problems with using an online service (though that’s more of an inconvenience rather than a serious problem).

    • Alex says:

      KeePass with strong passphrase on USB attached to my (physical) keyring. Never would trust the cloud on this. Make backups in case you loose your keys.

      • random832 says:

        The file is encrypted – that’s what the passphrase is for – so you don’t have to trust the cloud or worry about losing your keys from a security perspective.

        Unless by “trust the cloud” you meant using an online application for the password management itself.

    • J Mann says:

      I use KeePass with the data file in Dropbox, mostly so I can update and use it from either my PC or my phone, and so that I don’t need to have a backup strategy.

      The most annoying thing is that I lost my entire password database twice when I couldn’t remember my password. I think I’ve now got one that I can remember without writing anything down but that couldn’t be reverse engineered.

      Otherwise, I love it.

      • Alex says:

        I don’t use dropbox but if you do anyway, this is a good solution.

        • dodrian says:

          This does seem to tick the boxes for most of what I need. Going by this I would need to memorize unique passwords to log into my home desktop, KeePass, and Dropbox.

          Going up a notch technically – does it work with ssh keys? My ssh knowledge is pretty bad – but am I right in thinking that a private key for logging in isn’t directly tied to a specific computer?

          • Alex says:

            Not sure if I understand the question. To open/decrypt a KeePass file you can use a passphrase, a password file or tie it to a windows account or any combination thereof. Maybe Versions for other OSes allow using that OS’s account scheme. I only use a passphrase which is independent of being on a specific machine – like I said, I carry KeePass around with me on USB – and if you use one of the other options I stongly suggest to also use a passphrase in addition.

            Encryption of the password database AFAIK is symmetrical, I vaguely remember that it is AES. The biggest risk is that KeePass screwed the implementation. It is open source for whatever that is worth. I have not personally checked the sourcecode.

            I do not see an obvious connection to ssh. The program simply stores an encrypted file with your passwords in it, nothing more. But maybe I answered your question anyway.

          • random832 says:

            I think the question is if there is a password manager that can act as a ssh-agent (storing private keys in addition to passwords) and therefore you can only log in using your ssh key while your password database is unlocked. Apparently there is a plugin called KeeAgent that does exactly this.

          • J Mann says:

            The weak points are that I lose if:

            (1) There’s an exploitable flaw in KeePass;

            (2) Someone installs a keystroke logger in my phone or PC and captures my password or gets it by surveillance;

            (3) Someone captures me and tortures me for my password.

            #1 is the only systematic one, where a bot-net could capture my password as part of a systematic hack.

          • random832 says:

            A password manager also mitigates the risk of keyloggers – a hardware keylogger can’t capture your actual passwords (only the password manager passphrase. It is different, right?), and can’t capture the password database file.

            (Of course, if they can install a hardware keylogger they have physical access and can capture your unlock password to return and copy the file along with installing any software they want to.)

    • Anonymous says:

      I use two-factor auth for everything critical; since there aren’t many accounts, I can actually remember the passwords for these. For most everything else of any import, I’ve got a physical notebook – it makes the interface more human-shaped. If I lose the notebook, I know I must change the passwords, just like when I lose the key to my home.

    • J Mann says:

      How often does KeePass ask users to buy the developer a beer? I’ve bought two for him already – either time is passing faster than I think or it’s more than once a year.

    • GlenWill says:

      I’ve been using Lastpass for 4 years, and I’ve been happy with it. I started because I was reusing too many passwords, and had a sudden reason to think that one of my common passwords was compromised. Once I setup Lastpass, and started saving my passwords with it, I started having it generate different complex passwords for all my websites.

      It has the ability to one-click change passwords for many common websites, where it will open a new tab, change your password to a randomly generated one, and update your database with the new password. They will also notify you if you have a password on a site that has a known hack occur, and suggest you change your password for that site.

      It encrypts your password store on your computer, and only stores the encrypted version in the cloud. If you have multiple devices, passwords added or changed sync to other devices within a few minutes.

      My wife uses Lastpass too, and for accounts we share access to, I can create it in Lastpass, and then share it with her. She gets a copy she can accept into her store. If one of us changes the password, the other gets notified and updated.

      They actively pay a bounty for security bugs found, so there is an active community finding bugs and reporting them. They are very quick with fixing them as well.

      That link for the UK outage doesn’t overly worry me, as your encrypted database is cached locally, and as they indicated you can switch to offline mode and keep going. You just can’t share updates across devices while offline.

      The Android app is decent. Sometimes it doesn’t detect a field to autofill password, and I have to go to the Lasspass app and copy the password, then switch back to the app I am trying to login in to.

    • Nornagest says:

      I use PasswordSafe, which is less a password manager in the LastPass sense and more a password generator and repository. You need to unlock your repo to get passwords out of it (though it stays unlocked on your device for a configurable amount of time), but the data’s stored encrypted, and if you host it on a cloud service (I do, through iCloud), that service doesn’t have the key. This makes it a lot more inherently secure: you don’t need to trust the storage service, you just need to trust the local encryption.

    • Reasoner says:

      I realise that I reuse passwords much too often, and with more and more companies targeted (and demonstrating bad security) I’m starting to think that I need to take identity/account theft more seriously.

      If the company’s app was designed by someone who knows what they’re doing, your password will be hashed using bcrypt or something like that so that even if the site gets hacked, your password won’t have been compromised.

      • Nornagest says:

        You would be astonished how many people don’t know what they’re doing when it comes to security, even those that should know better. Especially those that should know better.

  19. Thegnskald says:

    Without delving into contentious cases, is random chance opposed in some fundamental way to justice?

    Is it unjust for somebody to have to pay to repair their car that, say, strong winds blow a tree branch into while driving down the road?

    Separate from this, should society seek to rectify the harms of random chance? Should insurance against this sort of random harm be mandatory or provided by government/taxation? (Or, assuming you are libertarian, would it be ideal for everybody to buy into such insurance, provided an ideal universe where risk is correctly allocated?)

    • Aapje says:

      I would argue that society should ensure that everyone can insure themselves for certain kinds of risks, primarily those related to basic human rights/needs.

      I would argue that it might be beneficial if the government is responsible for compensating people for risks where the government is the primary party that can reduce (or increase) that risk. For example, rebuilding after floodings or making good on deposit guarantees on savings. Although arguably, the government should then insure, rather than merely be on the hook once problems happen, to prevent them sticking their hand in the sand. If they are forced to insure, a well functioning market would push the government to making investments in risk reduction that are worthwhile. I also forces the government to choose to what level to insure, so you have precommitment to a certain level of compensation, so citizens know what to expect.

      I would argue that respect for agency dictates that people ought to be allowed to risk disparate impacts and face the consequences. If Bob and Alice play the lottery and Alice wins, Bob doesn’t need to be compensated for not winning. If we make the government too responsible, this creates huge incentives for society to ban things that are risky, creating bubble-wrap society.

      • I would argue that it might be beneficial if the government is responsible for compensating people for risks where the government is the primary party that can reduce (or increase) that risk. For example, rebuilding after floodings

        Bad example. One way, arguably the most practical, of avoiding damage due to floods is not to build expensive things in flood zones. The government provides subsidized insurance, which reduces people’s incentive to do so.

        • CatCube says:

          One way, arguably the most practical, of avoiding damage due to floods is not to build expensive things in flood zones.

          I dunno. Water transport is the cheapest and most efficient transport mode, and that’s tough to use if you don’t build on the water.

          • That’s a reason to build ports on the coast. It’s not a reason to build vacation housing there.

            I have no objection to people who like the seaside building there, but they shouldn’t be subsidized by having someone else pay for one of the costs of their choice.

        • Aapje says:

          @DavidFriedman

          One way, arguably the most practical, of avoiding damage due to floods is not to build expensive things in flood zones.

          Much of my country is a flood zone. Not going to happen.

          Although I agree that people who build in particularly bad spots can be held responsible for that.

    • Brad says:

      Random chance is rarely truly random. Who owns the land that tree was sitting on? Was it totally unforeseeable that a branch would blow off of it onto a car?

      The law and economics spin on the common law rule is that the least cost avoider should pay. If it would be cheaper for the trees to be trimmed than the landholder has to trim the trees, if he doesn’t he owes the car owners new windshields. But if it is cheaper to just repair the windshields when one is hit then the car owners have to pay.

      • random832 says:

        It’s almost certainly cheaper to trim one tree than to replace a windshield. Should this scale to the number of trees that they own? Only to the extent that the risk scales? Trees closer to the road probably present a greater risk to cars, but should that mean that the car owner has to pay if by some happenstance it happens with a tree more than X feet from the road, on the grounds that it’s inefficient to expect the land owner to trim all his trees?

        Should whoever built the road have been responsible for paying, on an eternal basis, the adjacent land owners to trim or remove the trees within X feet of the road, or for acquiring the right (and responsibility) to do it themselves?

        What if no-one owns the land?

        What if someone can be blamed for the wind? (nearby skyscrapers changing wind patterns)

        But if it is cheaper to just repair the windshields when one is hit then the car owners have to pay.

        Repairing the windshields doesn’t avoid (and trimming the trees doesn’t avoid the cost of trimming the trees). What if it would have cost the car owner more to drive a different way that doesn’t pass by the tree than for the landowner to trim the trees? What if it’s more expensive to repair the windshield than to trim the trees, but less expensive to avoid driving by them – does the driver have a responsibility to do so? What if there are trees along the other road too? Then the landowner is victim of the random chance that this driver chose to drive by their land instead of someone else’s.

        What if it would have cost less for the owner of a nearby toll road that doesn’t go past these trees to have lowered their rates such that the driver would have chosen to drive that way instead?

        —-

        If someone parks across the line of a parking space and I key their car, should they have to pay for repainting their car because they could have avoided it cheaply by parking better or choosing not to care about how their car looks? What of the cost to the parking lot owner of an arrangement of parking spaces that could have made it easier to park correctly? What about the other people in the market who care about how cars look and reduce the resale value of a car whose paint has been scratched?

        What if their car has already been scratched many times, and therefore there is no measurable impact on the car’s resale value – and the car owner clearly doesn’t actually care (or maybe simply can’t afford the repair) until they found someone they can extract payment from – but I’m the only one who gets caught in the act?

        Does it make a difference if it’s an accidental (or ‘accidental’ – suppose I parked the best I could, but couldn’t quite fit through the door as far as I could open it) door ding rather than a deliberate key-scratch? What about the damage to my car in that case?

        I started this scenario with having deliberately keyed someone’s car because it seems obvious to blame the person who deliberately acted… but they deliberately left their car improperly parked.

        • beleester says:

          I think you could even more easily avoid the damage by not keying their car, since you can avoid it by simply doing nothing.

          IANAL, but I’m pretty sure there’s no “He started it!” legal doctrine. If you’ve been damaged by the guy parking in your parking spot, then you can seek to get those damages restored (such as getting the guy’s car towed, so the spot is open again). You don’t get to inflict additional damages of your own.

        • Brad says:

          I know it is lazy and I’m sorry but I’m going to point you to Calabresi – The Cost of Accidents and Posner – Economic Analysis of Law.

          • Friedman, Law’s Order, is more readable than Posner. And shorter. And can be read for free online. Either with or without the layout of the print copy and the virtual footnoes.

          • Brad says:

            Or that ^^

          • Whatever Happened To Anonymous says:

            I don’t doubt all this is true, but I can’t help but feel it’s kind of bad form to recommend your own book.

          • Why?

            I’m not proposing that people buy it–the full text is webbed, and I linked to it.

          • dndnrsn says:

            It not being “I’ve just updated the course text I wrote so the page numbers in the syllabus are different from the past version PS it’s 200 bucks” is already better than 95% of profs assigning their own stuff.

    • J Mann says:

      In theory, mostly yes.

      1) It’s possible that people are risk loving in some areas. It might be that people prefer a system that produces a few lottery winning millionaires to a system that produces no millionaires.

      2) It’s difficult to separate out the random chance from the effects of decisions. From an efficiency standpoint, the common law typically placed the loss on the person best able to avoid the risk. In your example, if we pay for falling tree damage to cars, we may reduce the incentive to examine a parking spot for safety.

      3) There’s a tremendous amount of random chance, such that we can’t ameliorate all of it.

      4) The cost of sorting out the random chance from effects of decision-making, then compensating for it, would be tremendous.

      5) By definition, random chance is random. For affordable losses where it’s hard to separate out chance from effects of decisions, we generally hope that things break fairly evenly. Hopefully, everyone will have reasonably close to the same amount of random minor effects over the course of a lifetime thanks to the law of large numbers.

      Another interesting question is whether we should capture beneficial random chance events for the public good. Greg Mankiw had a pretty good proposal for a height tax – we know people who are taller make more money; generally tall people don’t have much influence on how tall they are; and it wouldn’t discourage much production, assuming tall people aren’t actually more productive economically.

    • Jiro says:

      One problem is that pretty much anything can be argued to be “random chance”. It’s random chance that one person is smarter than another (a previous thread already seriously argued that we should redistribute gains attributable to high IQ, because it’s random chance). It’s random chance that one person is prettier than another. It’s random chance that one person is more conscientious than another. It may be random chance that someone is a bank robber, if a person with that brain structure, that upbringing, and those circumstances would inevitably end up robbing a bank.

  20. Thegnskald says:

    On a related note to the previous comment, but along a different line, given that government is responsible for preventing interpersonal crime, is any interpersonal crime that occurs the result of government negligence? Should government be responsible for paying out claims against, for example, theft?

    • Anonymous says:

      That seems rather patently abusable to obtain reimbursement from the public coffers (meaning: other people’s pockets, indirectly) fraudulently.

      Just imagine how much the copyright enforcement crowd would demand from the government, and how much the government would demand from us in turn.

    • Loquat says:

      If government is to prevent interpersonal crime, government must have the ability to reliably detect and intercept interpersonal crimes in progress. This basically means universal surveillance (even in private homes, because we need to prevent domestic abuse too) and cops everywhere so they can always get to you quickly. I don’t know about you, but I really wouldn’t want to live in a society like that.

      • Urstoff says:

        Tangentially, this makes me wonder if retributive punishment, while morally indefensible on the level of the individual punishment, is still important for maintaining bonds of trust among individuals in society and between those individuals and the state.

    • JayT says:

      I don’t think any random crime would be negligence. They can’t be expected to give you a 24 hour security detail. However, I wouldn’t be surprised that, if the government completely ignored a community, there would be the possibility to sue them for negligence if there was a history of ignoring an issue.

  21. random832 says:

    I happened to read Book Review: Singer on Marx recently and some of the comments stuck out at me.

    In one in particular, @memeticengineer wrote:

    But it doesn’t seem that likely. On the level of the whole economy, owners also decide what firms exist (through entrepreneurship, mergers and acquisitions, liquidation, etc).

    And this is following a general theme of pointing out that the valuable function that ‘capitalists’* provide is coordination. But they are, almost by definition, not really coordinated at all. What coordination they provide is, more or less, the ability of the system to efficiently channel Moloch. Would a centrally planned economy of any kind have built Vegas? Should the things listed in that article be considered part of capitalism’s ‘pile of skulls’?

    P.S. I really wanted to work in something about revealed preferences being the only kind of preference Moloch has, but didn’t have anywhere good to put it. I don’t want to not say it, so I’ll just leave it here.

    * It is unfortunate that “capitalist” both refers to “someone who holds capitalism as an ideology” and “someone who owns capital”.

  22. US says:

    I was debating for a while whether to share this link in a post like this one and to some extent I’m still debating it, but I put some time and effort into it, some people might find it interesting, and aside from the link drawing attention to my blog I could think of few reasons not to share it. The post is a very long post about health economics which took me most of a day to write (…I guess? I wrote it over multiple days, so it’s hard to estimate, but it took quite a bit of time to write). Topics covered in the post include various issues pertaining to the topic of preventable diseases, the level of complexity that is present in the health care sector, and stuff related to health care cost growth.

  23. ADifferentAnonymous says:

    What’s the relationship between the blue/grey tribe split and the Cinton-democrat vs. Sanders-socialist split? They don’t seem to map 1:1.

    • HeelBearCub says:

      To the extent that gray tribe members are Sanders supporters, they would seem to be really confused?

      • Urstoff says:

        I’m still not clear on what the “gray tribe” is. My stereotype is libertarianish tech bros, but that doesn’t seem like a large enough population to be granted tribe status.

        • John Schilling says:

          It’s larger than most entities described as “tribes”, it wields disproportionate influence in contemporary US culture and economy, and is nominally represented in politics by a party with permanent ballot access in most of the United States. It’s certainly not the equal of Blue and Red, but it is I think worthy of specific recognition.

          The fault with Scott’s chromatic tribal classification is not that it includes the Grays, but that it doesn’t include the Browns, the Blacks, the Greens, and probably several others of similar stature.

        • random832 says:

          It sounds to me like a big-tent “internet people” thing, which includes libertarians, bernie supporters (and ron paul supporters from the last election, and howard dean supporters from the one before that), people who support trump ‘ironically’, atheists, ants, death eaters, and all sorts of other miscellanea.

          Smart people and people who think they’re smart.

          • rlms says:

            I’ve always interpreted it as being a very specific reference to Scott’s rationalist ingroup (IIRC, when he introduced the terms with examples of things each tribe likes, MealSquares were mentioned in connection with the Greys).

          • John Schilling says:

            “There is a partly-formed attempt to spin off a Grey Tribe typified by libertarian political beliefs, Dawkins-style atheism, vague annoyance that the question of gay rights even comes up, eating paleo, drinking Soylent, calling in rides on Uber, reading lots of blogs, calling American football “sportsball”, getting conspicuously upset about the War on Drugs and the NSA, and listening to filk”
            – Scott Alexander, I Can Tolerate Anything Except the Outgroup

            No mention of Mealsquares, but presumably Soylent is close enough. All three definitions, taken literally, involve too many brand names or other specifics to be worth considering literally; they are templates for typical Red/Blue/Grey tribe members, nothing more.

            So, people who don’t travel by SUV, Prius, or Uber can still be members of one of the canonical tribes, characterized by whether their preferred mode of transportation is more like an SUV, Prius, or Uber. And then averaged over all the other qualifiers, so a libertarian atheist paleo-Soylent eating, blogreading, filking, non-sports-fan who drives a pickup truck is still Gray Tribe even if a pickup truck is more like a Red-tribe SUV.

            Which is a longwinded way of saying, no, Gray Tribe isn’t just Scott’s friends in Bay Area rationalism.

          • random832 says:

            Paleo and Soylent seem almost directly opposed, so it doesn’t seem to be a particularly literal thing, more that both of these things are examples of people who think that they’re more clever about what they eat than everyone else.

          • Brad says:

            So, people who don’t travel by SUV, Prius, or Uber can still be members of one of the canonical tribes, characterized by whether their preferred mode of transportation is more like an SUV, Prius, or Uber. And then averaged over all the other qualifiers, so a libertarian atheist paleo-Soylent eating, blogreading, filking, non-sports-fan who drives a pickup truck is still Gray Tribe even if a pickup truck is more like a Red-tribe SUV.

            But on the other hand, the simplification that says Libertarian -> Gray Tribe or even worse ‘I’m a Republican but I like to call myself a libertarian even though my positions are barely distinguishable from huge swaths of the Republican party’ -> Gray Tribe is indefensible.

        • The Nybbler says:

          The original description of Gray Tribe is

          (There is a partly-formed attempt to spin off a Grey Tribe typified by libertarian political beliefs, Dawkins-style atheism, vague annoyance that the question of gay rights even comes up, eating paleo, drinking Soylent, calling in rides on Uber, reading lots of blogs, calling American football “sportsball”, getting conspicuously upset about the War on Drugs and the NSA, and listening to filk – but for our current purposes this is a distraction and they can safely be considered part of the Blue Tribe most of the time)

          But what’s happened in the meantime is the Blue Tribe proper has been taken over by the Social Justice group. This has lead to a split in Blue Tribe at a completely different fission point, and people sometimes talk about Gray Tribe as being everything on the non-SJ side of _that_ split. The original Gray Tribe remains fairly insignificant, but they can no longer reasonably be considered to be part of Blue Tribe.

          (note both Scott and I have been inconsistent about “Grey” vs “Gray”)

          • John Schilling says:

            But what’s happened in the meantime is the Blue Tribe proper has been taken over by the Social Justice group.

            A small fraction of blue tribe. It happens to be the fraction that Scott, and many of us, are closest to, but it’s still only a part of the mostly-not-SJ-dominated whole. Just like the Bay Area techies aren’t the whole of the (fairly well established by 2014) Gray Tribe, but they are the part that the SSC/Rationalist community sees most often.

          • The Nybbler says:

            A small fraction of blue tribe. It happens to be the fraction that Scott, and many of us, are closest to, but it’s still only a part of the mostly-not-SJ-dominated whole.

            I disagree. Hollywood, (non-right-wing) journalism, much of academia, all SJ-dominated. They may not be the bulk of the old Blue Tribe, but they are its representatives and leaders. Aside from the (very large) issues of politics, this can be seen by the “cultural appropriation” meme being current.

          • Urstoff says:

            But what’s happened in the meantime is the Blue Tribe proper has been taken over by the Social Justice group.

            Again, culture wars seem to exert an outsize influence on SSC commenters. The Blue Tribe in no way has been taken over by the Social Justice movement. If that were true, Clinton and Sanders would not have been the two major candidates for the Democratic nomination. Neither are remotely SJ types; Clinton occasionally pays lip service to it while Sanders seem to think it’s nonsense distracting from class issues. The policy of Democratic politicians doesn’t seem to reflect a SJ core, either. If the Blue Tribe were dominated by SJ types, shouldn’t that be reflected in who actually holds power at the behest of Blue Tribe voters?

          • The Nybbler says:

            If the Blue Tribe were dominated by SJ types, shouldn’t that be reflected in who actually holds power at the behest of Blue Tribe voters?

            That’s a lagging indicator. And note when I say “dominated”, I do not mean “predominates”.

          • Aapje says:

            @Urstoff

            I think that what Nybbler means is that you may not be against SJ in the blue tribe. Clinton and Sanders clearly didn’t dare to speak out against SJ.

          • random832 says:

            @The Nybbler

            I disagree. Hollywood, (non-right-wing) journalism, much of academia, all SJ-dominated. They may not be the bulk of the old Blue Tribe, but they are its representatives and leaders. Aside from the (very large) issues of politics, this can be seen by the “cultural appropriation” meme being current.

            Current, but not very effective. The making of movies that can be accused of cultural appropriation, whitewashing, or otherwise ‘problematic’ hasn’t exactly stopped.

          • John Schilling says:

            The part of Hollywood that e.g. runs the Academy Awards, may have been captured by the Dread and Terrible Forces of Social Justice(tm). The part of Hollywood that e.g. produces, distributes, and markets movies, has not. If I’m the supreme commander of the Forces of Good(tm) in the Culture Wars, I’ll count that as a win, and a win in solidly Blue territory.

          • Glen Raphael says:

            @random832:

            Current, but not very effective. The making of movies that can be accused of cultural appropriation, whitewashing, or otherwise ‘problematic’ hasn’t exactly stopped.

            Is it possible for the making of such movies to stop? Your phrasing suggests that whether movies can be accused of such things is limited by the nature of the movies, whereas I suspect it’s only limited by the creativity of the accusers. Is there any reason to think the set of issues deemed ‘problematic’ won’t keep arbitrarily expanding?

        • Reasoner says:

          For what it’s worth, I don’t think “tech bros” are really a thing except in journalist’s fantasies. Having worked in technology and also seen something of other industries, people working in technology strike me as less “bro-ish” if anything.

          • Aapje says:

            I think that there is a misconception that lack of social skills = bro.

          • dndnrsn says:

            The common stereotype of “bro” doesn’t seem to have much to do with lack of social skills. Lack of social graces, maybe.

          • Paul Brinkley says:

            My impression of “tech bros” is basically Aly and Jason from Silicon Valley. Males, in their element, to the point of feeling entitled to swagger about it without being physically violent or bullying.

      • ADifferentAnonymous says:

        So would you say the model is that Clinton/Sanders was an intra-blue split?

        • HeelBearCub says:

          Yes, plus an attempt by Green’s and other far-leftists to co-opt the Blues, depending on how large or small you want to expand Blue Tribe.

          I’m just saying that Sanders is not exactly a fan of the free market or other libertarian thought, which seems like a marker of Gray Tribe (assuming Gray tribe even exists).

  24. Nancy Lebovitz says:

    Type 2 diabetes is correlated with exposure to serious discrimination for being fat

    I don’t have access to the whole paper, but at least it’s a large study, isn’t about people hypothesizing how they’d react, and is on humans rather than mice.

    “Everyday discrimination was not associated with incident diabetes. In conclusion, major experiences of discrimination were associated with increased risk of incident diabetes, independent of obesity or behavioral and psychosocial factors.”

    • Jiro says:

      If everyday discrimination was associated with diabetes and major experiences were not, or if both were, I’d expect to see similar articles. And they specifically said that the study was meant to test the hypothesis that both forms are associated with diabetes. They also evaluated discrimination based on ethnicity separately. In other words, this sounds like obvious P-hacking.

  25. Nancy Lebovitz says:

    In case you want more fiction with rich wordplay, check out Moonwise by Greer Gilman. Note: the hardcover has an introduction by Michael Swanwick. The paperback doesn’t.

    Moonwise goes direct to the overlapping meanings of words (including some obscure ones) rather than routing through kabbalah. Did you know that cloud and law both mean hill?

    A Cloudish Word-Hoard, a glossary by Michael Swanwick

    I did a close reading of the first few chapters.

  26. Nancy Lebovitz says:

    I’m currently reading Unsong in the hopes of finishing this Sunday with the last chapter.

    This means basically not reading the comments, but I’ve noticed that a very high proportion (more than half) of the chapters so far (I’m at chapter 46) seem to have about 150 comments. Any thoughts about why there’s a strong trend?

    Since I’m just eyeballing it, it’s possible that the trend isn’t as strong as I think.

    • beleester says:

      If there’s a dedicated group of fans following a story, and it hasn’t changed much in size since the story started, then you’d expect all the chapters to have a similar number of comments, right?

      • Nancy Lebovitz says:

        I would expect more variation in how much people think is worth saying about different chapters, though perhaps having chapters come out on a regular schedule means that the conversation moves on. There are a few(?) chapters with 300 comments.

        There may be another high point at around 75 comments.

  27. Le Maistre Chat says:

    An odd thought: does mass media PC apply to white ethnic groups?
    The question occurred to me after watching the Rifftrax (Mystery Science Theater 3000 without the puppet skits) of the 1967 “comedy” Hillbillys in a Haunted House, with its anemic premise that it would be funny if Basil Rathbone, Lon Chaney Jr.. and John Carradine made a haunted house as a Scooby Doo-esque front and got foiled by bumbling hillbillies. This sort of ethnic humor seems to have flourished until, I dunno, the cancellation of The Dukes of Hazzard.
    On the other hand, there’s stand-up comedian Jeff Foxworthy, but surely that’s akin to black comics being allowed to make fun of African-American culture.
    Likewise, it used to be acceptable for the mass media to use being Italian, etc. as a source of humor.

    Thoughts?

    • Urstoff says:

      it depends on what you mean by “mass media PC”. It still seems okay to make ethnic-based comedy (for example) as long as it’s done well and done from a place of affection. Sitcoms like “Blackish” and “Fresh off the Boat”, while PC in various ways, are also non-PC in various ways as well. Similar sitcoms about rural life or the South crop up now and again.

      The shift I’ve seen in recent years has been from viewing “rednecks” as a target of scorn by white progressives to viewing them in a more anthropological light (see the pilgrimages journalists have made over the last year to Trump country to see “what went wrong”; some journalists are better at this than others). I’m not sure if this is an improvement, but it’s more pleasant than the constant ridicule of poor whites from places like Peopleofwalmart.

    • HeelBearCub says:

      Town mouse/country mouse is a very old trope.

      But, I think some of the flourishing of that particular brand of humor that your describing (Beverly Hillbillies, Green Acres, etc.) might come out of the the advancement of the highway system. You had the ability of people get exposure to the difference between the urban and country cultures, but not enough of an ability to normalize the cultures. I’m sort-of pulling that take from my nether regions, but that kind of comedy depends on being able to reduce certain aspects of a person or culture to a caricature.

      Whereas the proliferation of media availability makes exposure to all sort of people, from all sorts of walks of life much more available. Think of “Dirty Jobs”. It’s not treated as exposure to the exotic, but rather the everyday.

      Of course you still have all sorts of media where people are reduced to caricatures, but it’s less prevalent and more easy to overcome. For an example of the latter, she of “Cash Me Outside” fame.

    • The Nybbler says:

      Not a lot of comedy about white ethnic groups nowadays, but some seriously harsh drama. I’m not sure you could make something like “Winter’s Bone” or “Breaking Bad” about black criminal subcultures. But drama is still freer in general; “Breaking Bad” was not particularly kind to Hispanic drug subcultures either. Is there anything which treats Islamic terrorist subcultures that unsympathetically?

      • Trofim_Lysenko says:

        Not in depth and at length, but I can think of plenty of movies where they are unambiguously the bad guys, though the first two that sprang to mind (The Kingdom and Act Of Valor) both received criticism on that ground (among various other criticisms, especially Act of Valor which isn’t that deep and has the vices of its virtue of using actual SF personnel as its main stars) and neither were well received by critics.

        • keranih says:

          I think The Kingdom deserves a second look by anyone who gave it a pass on grounds of negative portrayal of Muslims/terrorists/whatever the first time around. I found it to be far more nuanced than the bad press implied.

          (Think, oh, Red Dawn being sold as a movie for jingoistic rednecks who just want to kill commies.)

          (Yes, the first one. Not the second one where they completely wasted Jeffery Dean Morgan’s character.)

      • Fahundo says:

        I’m not sure you could make something like “Winter’s Bone” or “Breaking Bad” about black criminal subcultures.

        Is the Wire too old to count or too different from those examples?

        • The Nybbler says:

          I didn’t watch _The Wire_ (though I did watch the older _Homicide: Life on the Street_). I thought it was told from the perspective of the law enforcement; if that’s so it’s too different. If it isn’t so, it may be a counterexample.

          • sandoratthezoo says:

            The Wire has viewpoint protagonists who are cops and viewpoint protagonists who are drug dealers.

    • Conrad Honcho says:

      Depends on whether they’re the butt of the joke or not. Them Duke boys were modern folk heroes beloved by southerners.

  28. Mark says:

    Historical question.

    Was there any major difference in the lives and rights of peasants (or the equivalent) across different cultures? Were there any countries in which peasants had a better time of it? Or is it pretty much terrible oppression and horrible conditions across the board, until the 17th/18th century?

    I started thinking about this after watching ‘Silence’ the other day, which I actually thought was a really good movie to watch on the computer, taking breaks. I don’t think I’d have enjoyed watching it at the cinema, though.
    I mean, there was the Shimabara rebellion, but it seems particularly awful that these people would have, in general, been forced to submit to the violent treatment of their friends and family with no response.
    I suppose submission has probably been the norm for most of history.

    • keranih says:

      Yes, there were a great deal of differences across cultures. The following are my impressions from what I remember of a comparative seminar in world history decades back.

      The Scandinavians had a very loose system of ‘lordship’ – as much as because there wasn’t much wealth to get from peasant farmers. Their attitude was shared with the early Russians, where the nobles were more armed warriors keeping the villages safe. This changed after the Mongols came through – the 1800’s Russians were among the most miserable, esp compared to the West, which was less centralized in power.

      Japanese culture from the Endo period on was far more stratified, but the people of a particular region seemed to identify quite a bit with their own nobility. However, they were probably more physically abused, and Japan was always a bit poorer, than mainland Asia. Chinese culture became (at least in theory) a meritocracy faster than many, and the constant churn of peace-famine-warlords-revolt-newemperor-peace cycle kept things from getting out of hand.

      India was a bloody mess, but the stratification of the castes also meant that unless you were a gravedigger/untouchable, there was always someone else to look down on.

      I don’t know enough of the pre-Columbian Aztecs to say, except that everyone and their freaking brother was so damn happy when Cortez, et al, showed up and drug down the Aztecs that they must have been pretty miserable bosses.

      (Again, take with a grain of salt.)

      • Hircum Saeculorum says:

        I can’t say much about Scandinavia, India, etc., but I do think the point about the Aztecs should be clarified – people often claim that Aztec society was uniquely oppressive, violent, and brutal, and isn’t particularly true.

        It is true that, when the Spanish showed up, they found plentiful allies – but that isn’t because a world-historically unique level of oppression was intrinsic to Aztec culture, it was because they were an empire, and they had subjects and rivals with their own political goals and aspirations. Remember to that Europe didn’t exactly want for rebellions in the sixteenth-century.

        It is true that Aztec culture was heavily stratified and legalistic; human sacrifice certainly did happen; they were absolutely warmongers. But you could say the same about numerous other cultures, if you swap out “human sacrifice” for “brutal, state-sanctioned death”. Is heart-removal really worse than crucifixion? Hanging, drawing, and quartering? Burning at the stake?

        And it would also be remiss to neglect the positive achievements of Aztec culture. Leaving out “high culture” stuff (art, botanical gardens, etc.), and focusing on innovations and systems that impacted the life of the average peasant or city-dweller, there’s a lot to praise – good sanitation, a rigorous (if harsh) legal system, efficient agriculture.

        As far as extraction goes, I can’t find anything to suggest that it was more miserable than the Roman system, or those of Early Modern Europe. Service to the state seems to have been primarily in the form of military service (fairly routine Europe at the time), corvee, and land and commerce taxes. Fairly ordinary stuff. There’s a good paper on it here – the conclusion is fairly succinct.

        Outlying regions would have sent goods to the core and supplied the army on the frontiers; again, nothing usually cruel, though rebellion would be treated harshly (as it would in the Old World – see anything from the Pilgrimage of Grace, to the Muster Rebellion, to the Reign of Terror).

        Human sacrifice was generally performed on prisoners of war, albeit on an unusual scale relative to other Mexican cultures.

        Basically, the Aztecs weren’t all that exceptional in their cruelty, I don’t think it’s clear that your average Joe Schmo had a particularly terrible life relative to any peasant in Europe or China; if they did, it certainly wasn’t that that motivated the uprisings during the empire’s dying days.

        Edit: in terms of general social structure, I found a fairly good “money quote” in a paper called “Demystification, Enriddlement, and Aztec Cannibalism: A Materialist Rejoinder to Harner” (I know it sounds a bit wooley, and there is some fluff at the beginning, but there’s quite a lot on crunch from page 2 on):

        Social stratification in sixteenth century Central Mexico comprised an upper class, which consisted of a local or supralocal ruling house, and a landed nobility, which formed an officer corps for the army and probably contributed to the upper echelons of the (celibate) priesthood. An incipient “middle class” of professional traders and craftsmen in sumptuary goods may have begun to put some pressure on the existing order, perhaps through increasing capital formation in this sector, perhaps for other reasons. Such strains would have been more important in some polities, where these groups were most heavily represented, than in others. At the bottom of the social pyramid, a large rural peasantry constituted primary producers. Of these, some were “free,” holding land as members of calpulli groups (themselves internally stratified); others serflike, bound to the landed estates of the nobility, of civil offices and the ecclesiastical establishment, and of the
        royal houses. While the nobility was hereditary, an institution comparable to the life peerage also existed, wherein the king could ennoble ommoners who had achieved (usually military) distinct

        Sounds quite analogous to Europe – nobility, a variety of landholding arrangements for peasants, and an artisan class.

        • Is heart-removal really worse than crucifixion? Hanging, drawing, and quartering? Burning at the stake?

          Not obviously.

          Those other things were all punishments for crime. Human sacrifice isn’t.

          • Hircum Saeculorum says:

            Those other things were all punishments for crime. Human sacrifice isn’t.

            Sometimes, nominally. Remember, though, that the crimes to which those punishments were applied were often crimes of ideology – religious or political dissent, real or imagined – or for relatively petty crimes which we wouldn’t consider worthy of execution.

            They could also be visited upon civilian populations – see the crucifixion of the inhabitants of Tyre by Alexander the Great, or the horrors that accompanied the capture of any pre-modern city. They could be performed to keep servile populations in line – see the execution of 400 slaves for the murder of a single Roman citizen during the reign of Nero.

            Also, the victims of Aztec sacrifice were most often high-status prisoners of war – not that that makes it morally better, but it does mean that the fear of being chosen for sacrifice didn’t necessarily pertain to the average Joe most of the time.

            That isn’t to say that human sacrifice is morally OK – that would be insane – but that it was another exhibit in the menagerie of human cruelty, not a totally exceptional phenomenon.

        • Deiseach says:

          It is true that Aztec culture was heavily stratified and legalistic; human sacrifice certainly did happen; they were absolutely warmongers. But you could say the same about numerous other cultures, if you swap out “human sacrifice” for “brutal, state-sanctioned death”. Is heart-removal really worse than crucifixion? Hanging, drawing, and quartering? Burning at the stake?

          They were indeed a developed civilisation and culture. At the same time, flower wars are not necessarily much better than the ordinary wars of conquest that the Aztecs also engaged in, and it is undeniable that (a) they did engage in this semi-ritualistic warfare frequently and at least partly to keep the gods satisfied by the deaths in battle (b) for several reasons, when the Conquistadores showed up, local tribes were very glad to see the Aztecs getting pounded. Maybe the constant risk of having your elite whittled down by the Aztecs in a ‘flower war’ so that you were always a weakened opponent that could never throw off their yoke or have a realistic chance of winning such battles wore their neighbours/client cities down to such a state, seeing the Aztecs overthrown looked like a good deal to them.

    • Eric Rall says:

      It varied a lot by country, by culture, and by time period. Two big trends I’m familiar with:

      1. A relatively balanced rivalry between local nobles and the central government (usually a King in the time period we’re talking about) tended to be good for commoners, since the nobles and the King would compete for their loyalty of the commoners and each could serve as a check on arbitrary excesses by the other.

      2. Peasants in particular (a large subset of commoners, but not the whole set by any means) tended to be better off when the population was low relative to the carrying capacity of the land. This is due to declining marginal returns on labor: more peasants on the same land gets you more food total (through more labor-intensive farming techniques), but less per peasant. When the peasants-to-farmland ratio was unusually low (after a war or plague wipes out a big chunk of the population, or in newly-settled frontier country, or after a technological shock like the introduction of potatoes and other New World crops around 1500 AD radically increases the amount of food a certain combination of labor and land can produce), peasant labor was relatively more valuable and land was relatively less valuable, so peasants tended to get a better deal from their landlords, both in terms of material conditions and in terms of treatment.

    • Rob K says:

      I keep meaning to recommend this book here; “The Brenner Debate”, a historical discussion over the relative power of the peasantry in various European cultures in the early modern era and how it influenced the rise (or not) of capitalism is really interesting reading.

      (Short summary – the more powerful the peasants, the less workforce there was available for early industrial enterprises, since people preferred to remain on the land. Traditions around property rights also mattered. England ended up on the extreme of those two measures.)

  29. Vermillion says:

    Hey Atlanta SSCers we’ll be having another meetup this Sunday, over by GT this time.
    We’d love to see you there, especially if I couldn’t read your email address from the first meetup.

    Sunday 5/14 7-10PM
    Octane
    1009 Marietta St NW, Atlanta, GA 30318

    Topic of conversation will probably include Unsong, but I still haven’t read it and probably won’t before Sunday so as an alternate I’m gonna take a page from one of my favorite podcasts No Such Thing As A Fish, and ask everyone to bring your favorite fact of the week and we’ll discuss em.

    My fact will be about comparative neuroanatomy, everyone loves facts about that right?

  30. dodrian says:

    Saw this project on Kickstarter today: “Solar Gaps.”

    Essentially it’s solar panels in the form factor of window blinds. But reading through the campaign and it seems highly suspect.

    The video implies that you plug it into a power outlet and it generates power for you. Reading carefully through the campaign and FAQ it looks like they prefer you to connect it to a home-battery (eg. Tesla Powerwall) or hook it up to a green energy thingy given by your power company (that lets you sell power back to the grid).

    I wouldn’t buy it in any case (cost, lack of suitable windows, etc etc), but I’m curious as to what y’all think. Viable product? Vaporware that gives exciting updates but disappears after a year or two? Would plugging a solar panel directly into a regular power outlet even work?

    • sandoratthezoo says:

      There’s no way that you get cost-efficiency out of vertically-oriented solar panels that also have to function as blinds (ie, be collapsible or whatever).

    • John Schilling says:

      When I open my solar blinds, do I at least get to look out on the solar road in front of my house?

      The proper place for solar panels is on the roof. Even the most inconveniently-oriented roof is likely to be better positioned than a window for this purpose. And bigger. And fixed in place, the better to attach wiring that won’t work its way loose over time.

      For solar power on a scale or in a place where the roof isn’t big enough, the proper place is a big vacant plot of land, of which there are many within reasonable power-transmission distance of most any market.

      So long as most roofs and most vacant plots aren’t covered in solar panels, anyone selling a scheme to put solar panels someplace that sounds really cool, is selling hype.

      Oh you can plug alternate power sources (converted to 110VAC or whatever and no more than the rated amperage of the circuit) into your power outlet and it will work just fine in most circumstances. But if there is a power outage because e.g. a tree fell on the line connecting you to the grid, the lineman sent to repair it may turn off the circuit at the transformer and then get himself electrocuted because the line is still getting full voltage from the other side. So that sort of thing is I believe illegal in most jurisdictions; you have to have a licensed electrician do the installation and verify that there are properly-functioning automatic cutoff switches.

      • Aapje says:

        My solar inverters are grid-tie inverters, which match phase with the utility-supplied sine wave. They shut down automatically upon loss of utility supply, for safety reasons.

        The downside is that I don’t have any power during a power outage, but those are very rare in my neck of the woods.

        • random832 says:

          My solar inverters are grid-tie inverters, which match phase with the utility-supplied sine wave. They shut down automatically upon loss of utility supply, for safety reasons.

          Is there any reason this is better than disconnecting from the grid (but staying connected to your house)?

          • AnarchyDice says:

            Generally it is best to shut down solar panel systems upon loss of grid power unless your solar system is sized large enough to supply all your power needs. Otherwise, it can be bad for your electrical equipment to suffer through not having enough power, as it will likely drop the voltage below their intended range, go out of phase, or change frequency.

            Alternatively, it is for safety so that you can be certain that there is no power coming in during an outage so that no one gets shocked assuming that power is disconnected during a utility outage. This is the same reason many utilities require alternative power systems to have a clearly marked disconnect switch controlling the power source outside the building for them to shut off should they need to work on your local transformer or wiring.

          • Aapje says:

            @random832

            A power outage is called a blackout, but a partial loss of power is called a brownout. As AnarchyDice argued, this can do bad stuff to electrical equipment.

            I would only want to use my solar power during a power grid outage if I had a big battery. Something like Tesla’s Powerwall would do, but it would turn my solar installation from break-even in 7 years to making a permanent loss. My solar installation is an actual investment where I expect more ROI than putting my money in a bank/stocks, so that would be bad.

            The average power outage per house per year in my country is about 21 minutes. So the quality of life improvement for spending 7000 euro on a Powerwall is 21 minutes of Internet/TV, if I’m not at work (in which case, a power outage means I get to sleep behind my desk).

            If we are talking about dealing with disasters, then there are many, many more cost-effective investments I could do to increase my chance of survival.

  31. Le Maistre Chat says:

    What does everyone make of the story that went viral the other day, about Czech Girl Scout Lucie Myslikova arguing with a skinhead who says “We have to stop immigration or they’ill rape you” and she says “If they rape me, I’ll get over it”?

    • Mark says:

      It’s hard to think of an ideology to which her statement isn’t a betrayal.

      I suppose what she is saying is, really, that a rape would be an unfortunate and unlikely occurrence, not really related to immigration. So, STFU.

      But it’s a really unfortunate way of expressing that sentiment. If she’s 16 years old, the fault isn’t with her, but with whoever is lionising the statement.

      I mean, I’m quite stupid now, but when I was 16, I was far worse. Such is life.

      [Having done a quick search, I couldn’t find any reference to the “I’ll get over it”, so I guess it’s the Guardian taking advantage of the photo opportunity, and then the Nazis responding by saying – “hey, look, the 16 year old said something dumb”
      Liberal women don’t want to get raped by Muslims, they just don’t think that immigration is going to increase rape.
      ]

      • sandoratthezoo says:

        It’s not a betrayal to anyone to get over horrible experiences. It may be an unrealistic self-assessment, and I get that people can take it to mean “people who can’t get over a horrible experience are bad, weak people.”

        But we should also be careful not to suggest, imply, or speak-carelessly-making-it-easy-to-infer that:

        a. Nobody can ever get over rape.
        b. Women who do get over rape are betraying women in general or other rape survivors in particular.

        • Gobbobobble says:

          I think the interpreted implication of “I’ll get over it” here is “No big deal”

          • sandoratthezoo says:

            I think that’s an uncharitable interpretation, but sure. And, again, I see how that can easily be turned into an insult to people who have been raped and to whom it is of course a big deal.

        • Mark says:

          Come on.

          This is one of those things, isn’t it. Where what you’re saying is completely true, and completely right, but you’re making a truthful statement in application to a completely unrelated discussion.

          OK. example.

          “There is a group of people going around stabbing people.”
          “I’ll heal.”

          Person A – That’s an awful way to make that argument

          Person B – Hang on a minute, are you saying that people can’t heal? Come on, dude, that’s medically speaking inaccurate, and probably discourages people from healing. So… I’d just like to make it clear to everyone that people can heal. (Person A, you fiend.)

          [Yes, if someone is saying that rapefugee immigration (as the Nazis understand it) isn’t a problem because women will get over rape, if they truly believe that, (she doesn’t) then I have a hard time understanding their ideology.]

          • sandoratthezoo says:

            So there are these actions that have historically been seen as less-harmful than they objectively are. Rape is one of them. Bullying is a good example of another. Stabbing is not one of them.

            And there have been (good! positive!) recent campaigns to recognize how harmful those actions are.

            But I notice that the negative side of those campaigns is that they make it so that anyone who talks about healing, getting over the trauma, or otherwise not being defined by the crime now starts to get suspicious stares and accusations of trivializing the crime.

            Clearly, Myslikova did not make her argument in the way that was most sensitive to the victims of rape. But in my set, at least (no idea what the situation is in the Czech Republic), I don’t really need to worry that my peers are going to think that rape is no big deal. I do think that I need to worry at least a little that people will feel ashamed to get over their traumas because they feel like they’re betraying others who haven’t (yet?) healed.

            None of that dynamic applies to stabbings.

          • Mark says:

            So, what’s your point?

            Are you saying that she was making a valuable contribution by reminding us that it should’t be shameful to recover from rape?

            Honestly, I know nothing of your culture, but to me the fact that someone might recover from rape seems as normal as the fact that they might recover from a stabbing, and the opposite also.

            So, perhaps this is just a culture difference.

            New statement:
            Given the political circumstances, this probably wasn’t the best opportunity for the school girl to make her entirely true and helpful comments about rape survival.

            [To make my point clear:
            Her statement will be read as “I would rather be raped than restrict immigration”. That statement is an abomination. She should have made her statement in another way, but she is very young so she really can’t be blamed at all, and really, it shouldn’t be an issue.]

          • sandoratthezoo says:

            My point is:

            a. She’s being insensitive to rape victims.

            but

            b. People who react like you are being overly dramatic about the amount of harm she’s causing.

            and

            c. People who react like you are also doing some measurable harm to rape victims.

            and also

            d. In the circles I run in, there is an overwhelmingly large consensus that rape is an incredibly damaging, horribly traumatizing crime. I don’t think that that consensus is going to be heavily challenged by the thoughtless comments of one 16 year old. It behoves people who are thoughtful and trying to be kind to make their (correct) arguments about rape being incredibly damaging, horribly traumatizing without adding to the burden of victims of rape by making them feel guilty about whatever measure of healing and closure they manage to attain (whether it’s minimal or as-complete-as-human-experience-can-manage). In the event that the pendulum starts to swing in the other direction, I’d be more harsh on Myslikova’s comment.

          • Mark says:

            I now feel as if it is a bit rude to say or think anything about this issue.

            And, by “issue”, I mean my own opinion.

            So, if this is in bad taste, please stop reading.

            Any mention of rape will cause harm to some rape survivors, which is why it’s good practice to have trigger warnings, to make sure statements aren’t too severe, etc.

            So, I’ve probably failed in that respect.

            On the other hand, I stand by my opinion. And if that opinion upsets rape survivors more than any other opinion, with the best will in the world, I will still hold that opinion until they give me good reason to think otherwise.

            I mean, I could just not say anything, and 9 times out of 10, when it comes to rape that is exactly what I’d do. But, I guess there has to be somewhere we can voice our opinions about these issues?

            I mean, I really get the feeling that this is more about “tone policing” than anything else.

            Anyway, on a personal level, I don’t think she has caused harm. It’s the whole media-clicky-rage thing that causes the harm. But, if we leave the girl out of it and look at the statement – yes, this is absolutely the kind of statement that will drive Nazis crazy. And, for *years* this will be grist to their mill as they revel in telling us how liberal women want to be raped by immigrants.
            They only need a little encouragement.
            I mean, as long as you stay off the internet, it probably doesn’t matter, but some of us feel a weird compulsion to read that shit.

            So, that’s why I felt the need to comment.

            [But thank you for letting me know how you interpreted my statement, it was kind of out of left field for me, and felt like an attack, but I’ll assume it was in good faith and try to bear that perspective in mind in future.]

          • sandoratthezoo says:

            I only JUST NOW got that your objection to Myslikova’s comment is that you think it will add ammunition to the ability of people to make the case for bad policies and/or harass people, rather than thinking that Myslikova’s comment itself harassed people.

            I don’t think that Myslikova’s comment at all suggests that she wants to be raped. I also don’t think that Myslikova’s comment at all suggests that she stipulates that immigrants are disproportionately likely to be racists.

            I do think that:

            a. A plain reading of her remark somewhat trivializes the distress that rape victims experience.
            b. An uncharitable reading of her remark suggests that rape is no big deal.
            c. A frankly mendacious reading of her remark suggests that she believes that there are going to be lots of rapes of native women by immigrants.
            d. A sufficiently motivated reader could imagine in their dumb, sick brains that the thought process behind her remark is evidence in favor of their worldview that liberal women want to be raped.

            (I do not think that you are someone who believes b, c, or d).

            I guess that I think that to the extent that her remark is going to be weaponized by, whoever, Nazis or harassers or trolls, it’s not going to be particularly effective ammunition, and that given how much of a contortion you have to go through to believe c or d, someone who believes c or d could find dozens of other statements that provide as much ammunition as Myslikova did.

          • Mark says:

            Hmmmm… I don’t know. I think if you look at the context of the remark,
            “Immigrants will rape you”
            “I’ll get over it”

            (c) and (d) are kind of in there, right?

            Aren’t they? I mean, do you really find that reading so obscure?

            Googling “Lucie Myslikova “I’ll get over it” returns

            A girl wearing a Scout uniform willingly acknowledged that she could be raped by an immigrant, something that is happening to women across Europe in record numbers. Then, she essentially reasoned, I’ll get over it, and the Western world lifts her up as a hero. She justified rape in the name of political discourse. A Girl Scout opened herself up to sexual assault for the sake of her political beliefs.

            Honestly, it took (a little) work for me to think of an alternative reading. (Hence comment)

            [I’m against open Med borders – probably why I’m in tune with the Nazis here. I object to her comment because I think it’s a bad and inflammatory way of putting what is probably (even in its best form) an incorrect argument. ]

          • sandoratthezoo says:

            Nope, I don’t see that as a plausible reading. The natural reading to me was something like this:

            “You’re trying to shock me into fear, and you’re trying to define me as a fearful, scared little girl, and further to make my identity one of either ‘rape victim’ or ‘potential rape victim.’ I’m not afraid of your ridiculous counterfactuals, my identity is not ‘potential rape victim,’ and in the unlikely event that I am a victim of rape, my identity will still not be ‘rape victim,’ I’ll just get over it.”

            Now, did she state all that explicitly? Clearly not. Did she create a carefully defensible answer like an expert politician which can survive hostile quoting? Clearly not. Was she perhaps overly blithe about her ability to deal with rape? Yes. But I don’t think it’s at all a natural reading to imagine that she’s conceding that she’s likely to be raped by an immigrant.

          • Mark says:

            OK, i find that quite annoying.

            Is it fair to say that the idea that the Nazi’s statement is a “ridiculous counterfactual” is doing a lot of work here?

            Like if he said,
            “Mow the lawn, or the grass will rape you.”

            Any normal person would read the response “I’ll get over it” as a partially sarcastic response?

            So, if you consider things from the perspective of the Nazi, or if you consider that the counterfactual might not be so ridiculous, would the alternative interpretation make more sense?

          • sandoratthezoo says:

            No.

            Look, I’m not here to get into an argument about whether the scenario really is a ridiculous counterfactual. I mean, you can guess my opinion, but I’ll leave that at agree to disagree.

            What I’m saying is that this exchange, as reported, strikes me as one in which the motivations and subtext of what Myslikova said are fairly clear, and that if she considered the whole scenario of “Immigrants will come and rape you” to be a non-ridiculous counterfactual, she would have responded differently. The reading of her regarding it as a ridiculous counterfactual appeal to emotion is a very natural one. The reading of her regarding it as a totally plausible scenario but that’s cool because she secretly longs to be raped by the demonized other is bonkers crazy.

          • Mark says:

            Yeah, you’re probably right.
            I guess the real issue (for sane people) would be that people object to her regarding it as an absurd counterfactual.

            I’ve probably been reading too much insane stuff recently.

          • lvlln says:

            FWIW, Mark, I think you’re right specifically about the point that the girl’s reaction has a built-in concession to the skinhead that they do have a point about immigrants raping her. I personally find absurd the idea that immigrants are likely to rape her or that the likelihood is sufficiently high compared to the base rate that this is something she should be concerned about. Slightly less absurd than the idea that not mowing the grass would lead the grass to rape her, but only slightly. But responding to such an absurd assertion, “I’ll get over it,” the natural charitable reading is that she’s conceding this completely absurd point by refusing to challenge it and moving on to a different part that she does deem defensible. That’s not to say that she’s making a statement of agreement, either; that would be an unwarranted reading.

            But, like you said, she’s a kid, and also this is one exchange. Even if she weren’t a kid, such a careless concession in one brief exchange doesn’t deserve blame. People in conversation say accidentally all sorts of things all the time that do a poor job of conveying their actual inner beliefs. Heck, look at the flubbed line “Small step for man, giant leap for mankind.” If the media is trying to make this go viral in order to make the point “See, even pro-immigration people concede that they’ll be raped!” then they are dishonest. I think people who see this and carefully think about it will recognize the dishonesty and not pay much attention, but I’m also a little skeptical that a very high proportion of people who see this will carefully think about it.

        • Wrong Species says:

          A BLM activist goes up to a black person and says “we have to stop systematic racism in this country. The police could beat you up.”

          What is the appropriate reaction to that?

          • sandoratthezoo says:

            I have no idea, and don’t understand the point you’re making. I think that there are all kinds of appropriate reactions to your hypothetical.

          • Wrong Species says:

            I think you have an idea of the point I’m making. But if you really don’t, if the situation I laid out happened and a black guy blithely replied that they’ll get over it, you would be at least a little disgusted at the guy and his willingness to throw his own self under the bus to protect the police. But since Islam is supposed to protected at all costs, you have to go through these mental gymnastics to reconcile your contradictory beliefs.

          • sandoratthezoo says:

            Ah, I see. You imagined that I agreed with you and created a parallel that does not exist.

            See, I believe that it is factually true that the police are disproportionately likely to beat up black people. I do not believe that it is factually true that immigrants are likely to rape women.

            So I don’t interpret Myslikova’s comment as meaning that she’s willing to throw herself under the bus in order to protect immigrants, I interpret her as saying, “Shit happens, nobody is perfectly safe, I’m a survivor.” (She’s a teenager, she probably thinks she’s more invincible than she is).

            That said, if a black person told me any of the following:

            a. He did not agree that systematic racism existed and that he was no more likely to be beaten up by the police than I was.

            b. He agreed that he was more likely to be beaten up by the police than I was, but that he thought that the BLM movement was weakening the police in ways that were ultimately worse for him or the country than was worth the benefit of decreased chance of being beaten up.

            c. That he just wasn’t personally all that invested in this particular problem and that it wasn’t worth his time to try to change things, and he’d take the risk.

            Then I’d, in case a: disagree with him on the facts, but not consider him disgusting or a traitor or anything. Maybe kind of naïve. In case b: I don’t know much about BLM, if he had a reasoned argument, I think I’m convinceable. If he didn’t convince me, I wouldn’t think he was disgusting or a traitor. In case c: Presumptively he knows better than I do about what’s worth his time. I don’t have any right to even an argument with him unless we’re close buddies.

          • Mark says:

            Um… what if he said, “I’ll get over it” ?

          • sandoratthezoo says:

            The girl’s “I’d get over it” feels like it has a clearer meaning in the context of a showdown with a neo-nazi than a black guy’s “I’d get over it” has in the context of a BLM activist trying to engage him in a cause. But, if you’re asking me, would I be offended if a black guy said that? No. Perhaps in some cases puzzled. If I had to totally guess without further elaboration what the guy meant when he said “I’d get over it” in your hypothetical, I’d guess he meant “I’m not not interested in engaging this whole thing, I think it’s a waste of my time,” in which case, fair enough.

    • rlms says:

      I don’t think that was the story that went viral. As far as I can tell, no-one said the things you mention. According Pamela Gellar, “When the Neo-Nazi asserted the teen would be ‘violated by those she defends,’ Myslikova replied, ‘Even if something happened to me, the physical wounds always end up healing.’.”. I don’t really trust her to accurately translate the French. Even if she did, and we accept that that the Scout interpreted the skinhead as saying a migrant would rape her (rather than e.g. mishearing/misinterpreting it as a violent threat from him), it seems vastly more likely to be the kind of instinctive reaction people make in shouty arguments without thinking if it makes sense, rather than indicative of anything.

    • keranih says:

      David Friedman can probably place the reference better than I can, but there was a castle under siege, defended by troops commanded not by the lord/duke/whatever, (out of town at the time) but his lady wife.

      The army outside had caught the son of the duke, and brought him up (under guard) and sent a message to the castle defenders that if they didn’t surrender, they were gonna cut the boy’s throat then and there. They call for the mother (looking down on them from the wall) to surrender or they will really do it.

      The duke’s wife hikes up her skirt, points at her crotch and says “I can make more sons, you cowardly gits” (note: not an exact quote). IIRC, the besiegers eventually went away, and the boy lived.

      I get much the same impression here – part bluster, part real refusal to surrender, part refusal to accept responsibility for the (bad) actions of others.

      It’s not the Scout’s support for open borders/progressiveness/racial equality/whatever that leads to her being raped, it’s the criminal actions of the rapist. You could argue that it’s not reasonable for her to fail to take precautions like supporting limiting immigration, but that still doesn’t make her the criminal assaulter here.

      • Anonymous says:

        It’s not the Scout’s support for open borders/progressiveness/racial equality/whatever that leads to her being raped, it’s the criminal actions of the rapist. You could argue that it’s not reasonable for her to fail to take precautions like supporting limiting immigration, but that still doesn’t make her the criminal assaulter here.

        Of course not. She’s not part of the government (which invites and fosters migrants) in any reasonable way, and therefore not responsible for protecting the citizens (like herself). If she’s 16, like people mentioned above, she isn’t even voting (yet). Her supporting anything is pretty meaningless. And being imprudent is not a crime.

      • axiomsofdominion says:

        That’s actually most likely a myth; its about Caterina Sfroza.

      • psmith says:

        The version I’m familiar with is as follows:

        When King Stephen besieged Newbury Castle in 1152, according to William’s biographer, he used the young William as a hostage to ensure that John kept his promise to surrender the castle. John, however, used the time allotted to reinforce the castle and alert Matilda’s forces. When Stephen ordered John to surrender immediately or William would be hanged, John replied that he should go ahead saying, “I still have the hammer and the anvil with which to forge still more and better sons!” Subsequently, there was a bluff made to launch William from a pierrière, a type of trebuchet towards the castle. Fortunately for the child, Stephen could not bring himself to harm young William.[5] William remained a crown hostage for many months, only being released following the peace that resulted from the terms agreed at Winchester on 6 November 1153 that ended the civil war.

      • @Karanih:

        I don’t recognize your incident. It might be the story from when William Marshal, as a child, was a hostage to Stephen, distorted in transmission to change, among other things, the gender of the person commanding the castle, or it might be a similar incident I don’t know about.

        • Deiseach says:

          There is also a version of this about the Knight of Glin, which I remember learning (in the bowdlerised version) in primary school history; Wikipedia has the account more or less the same:

          According to another legend, in the early 16th century under Elizabeth I, England set about enforcing loyalty in the western parts of Ireland. When one of her ships came up to the Knight of Glin’s castle on the Shannon Estuary, a fierce battle ensued. The ship’s captain managed to capture one of the Knight’s sons and sent the Knight a message that he should surrender or else the son would be put in one of the ship’s cannons and fired against the castle wall. He replied that as he was virile and his wife was strong, it would be easy to produce another son.

          …The garrison of the castle, according to tradition, was divided into two sections, one of which was commanded by Donall na Searrach Culhane and the other by Tadhg Dore. Before the siege began, Carew[Sir George Carew, President of Munster], who had the knight’s child as hostage, sent an order to the knight to surrender the castle at once or else he would blow the child out of the mouth of the cannon. The knight’s answer was remembered but can only be rendered here by algebraic symbols: Gread leat. Ta X go meidhreach fos agus Y go briomhar. Is fuiriste leanbh eile do gheiniuint.

          The Irish translates as “Be off with you! X (presumably a reference to the lady’s parts) is lively (lit. merry, playful) yet and Y (his parts) strong/vigorous. It is easy to procreate another child”.

          This seems like one of those stories that may be based on an actual incident but which is so good, it gets passed down and attached to various events distant in time and space, rather like those quotations that are attributed to anyone and everyone who is a Famous Historical/Literary Figure.

          • This seems like one of those stories that may be based on an actual incident but which is so good, it gets passed down and attached to various events distant in time and space

            The William Marshall version appears in the History of William Marshall, written shortly after his death, so about sixty years after the events described.

      • LHN says:

        axiomsofdominion is right that the story was told about Caterina Sforza, e.g., in Machiavelli’s Discourses on Livy (Book 3, Chapter 6):

        Some conspirators of Furli killed the Count Girolamo, their lord, and took his wife and children, who were of tender age, prisoners. Believing, however, that they could not be secure if they did not obtain possession of the castle, which the castellan refused to surrender, the Lady Catharine, as the Countess was called, promised to the conspirators to procure its surrender if they would allow her to enter it, leaving them her children as hostages. Upon this pledge the conspirators consented to let her enter the castle; but no sooner was she within than she reproached them for the murder of the Count, and threatened them with every kind of vengeance. And to prove to them that she cared not for her children, she pointed to her sexual parts, calling out to them that she had wherewith to have more children. Thus the conspirators discovered their error too late, and suffered the penalty of their imprudence in perpetual exile.

        (Though it’s interesting that in both this and the William Marshal case, contra what one would expect in something by George R.R. Martin, the hostage children were spared.)

    • Anonymous says:

      Sounds a lot like #inteerkvinna.

  32. Nancy Lebovitz says:

    Speaking of the Juicero thread that just went up, there doesn’t seem to be a kitchen gadget which carmelizes onions. It doesn’t seem like it would be technically complex. Have I missed something?

    • Douglas Knight says:

      Here is a kitchen gadget for caramelizing onions safely. (Other people say that a different gadget works better. But I understand the theory behind the first.)

      • Eric Rall says:

        I bake them inside a cast iron dutch oven ($20-50, depending on size and style). Any oven temperature between 275 and 400 will work, with the lower temperatures taking longer but the higher temperatures requiring regular checking and stirring (about every 15 minutes) to make sure they don’t burn. I usually do around 300 or 325, and I usually add a little bit of baking soda (about 1/4 tsp per pound of onions) to speed up the caramelization.

        In general, a dutch oven can do anything a crock pot can do with about the same amount of work and a moderate amount of additional knowledge, with the advantage of giving you more options for temperature and they let you sear your food before the slow-cook without getting a second pan dirty.

    • CatCube says:

      If there’s something relatively cheap, I might give it a try. I tried to carmelize onions for the first time on Tuesday, and they came out awful. I’m not sure what I did wrong.

      • Nancy Lebovitz says:

        In what way were they awful? The answer should supply some clues.

        Here’s the cheap way to do it. Put chopped onions and butter and/or olive oil in a pan at a very low heat.

        You don’t need to stir them constantly, but you do need to stir them every minute or two. Stop when they are brown and sweet.

        This is cheap but time-consuming.

        • CatCube says:

          Your cheap way is exactly what I did. (I was trying to follow the directions from here.)

          I used a 12″ stainless steel pan with a little olive oil in the bottom, put it on medium heat, and put a chopped onion in and sprinkled a little sugar on it. I stirred it with a metal spatula, scraping the bottom, every few minutes at first, then every 7-8 minutes later on.

          I was expecting the times found there to be roughly accurate, but it “felt” like it was cooking a lot faster. I kept turning the heat down, until it was practically at the point where I simmer sauces, and still called it at about 40 minutes. I don’t think it tasted burned, but it was just generally bad (I was having it with a steak and a baked potato. I’ve been sauteeing an onion with that meal and wanted to try something different. I ended up trying two forkfuls and throwing the rest away.)

          • Nancy Lebovitz says:

            Differences which may have made a difference.

            I used more than a little oil. I started at a low heat and stayed there. It seemes unlikely to have mattered, but I used a cast iron pan.

            I have no idea what could have gone wrong which led to such a bad outcome. What sort of bad was the onion?

          • CatCube says:

            It was just a regular yellow onion. Like I said, normally I’d sauté it instead, and it always came out fine; these were from the same package.

            There’s two things that I can think of. One is that I really overcooked it to the point of drying it out; a lot of comments on recipes tended to be “you need like an hour to carmelize an onion” so I was pretty anchored to the longer time, and ignored signs that it was done earlier.

            The other is that my stove seems to really tend towards hot. Packaged food with instructions to use “medium heat” will burn easily if I actually put the dial in the middle.

          • Iain says:

            If your caramelized onions are drying out, your pan is probably too hot.

    • Eric Rall says:

      The Juicero-level option is a Thermomix. It’s a food processor with a heating element and setting to use the food processor motor to stir stuff while it’s being heated, which is pretty much perfect for caramelizing onions (as well as doing a variety of other cooking things that normally require labor-intensive combinations of heating, chopping, and stirring). I hear it’s really good at doing what it does, but for onions it strikes me as “shooting down mosquitos with an anti-aircraft gun” level of overkill.

      • Paul Zrimsek says:

        More entertaining introduction here (NSFW: language).

      • bean says:

        I hear it’s really good at doing what it does, but for onions it strikes me as “shooting down mosquitos with an anti-aircraft gun” level of overkill.

        Why does google not know about the effects of explosions/shockwaves on mosquitos? I really want to find out how well this would work.

        • Paul Brinkley says:

          Finally found a reason to revive battleships, eh?

          • LHN says:

            Or at least do some testing using Iowa’s guns.

          • bean says:

            The problem is that there aren’t that many mosquitoes in SoCal, and the authorities here have all sorts of weird taboos against the use of live rounds. The last time we asked about just firing HE from the 5″ guns, they told us no. Something about not blowing away the Queen Mary. We were very reasonable, and didn’t even ask about the 16″ guns.
            To get enough mosquitoes, we’d probably have to go to Alabama or North Carolina. I expect the authorities there to be less rigid about the use of explosives, too.

          • Paul Zrimsek says:

            I suggest a really big can of Doolittle Raid.

        • Eric Rall says:

          Now you got me curious, and after some reference-chasing, I found a study on the relationship between body mass and LD50 blast overpressure. Their experimental data points were all mammals, not insects, but that’s at least a starting point.

          Plugging in a mosquito’s typical mass (2.5 mg) to their “short duration reflected pressure” regression equation (caption of Figure 7: log (LD50 PSI) = 1.7055 + 0.2502 * log (mass in KG)) gives an overpressure of 2.01 PSI to kill half the mosquitos in the radius. Eyeballing Figure 4 indicates that LD99 is pretty uniformly about 10 PSI more than LD50.

          Bean has a better idea how to track down reliable figures for the explosive charge in various AAA shells than I do, but I think I found a number for the 40 mm guns on the USS Iowa as having 92g of Hexolite, which seems to be equivilent to about 120g of TNT. And if I’m correctly plugging numbers into the Sadovsky formula for blast wave from TNT explosion, then that will create the LD50 overpressure for mosquitos at a radius of just under 1 meter, and will hit the LD99 overpressure at about 47 cm.

          I’m a bit out of my depth here, so I probably made several mistakes. Double check my work before using it for anything important.

          Other references:
          Shell info
          TNT equivalents (I assumed Hexolite was the same thing as HBX, based on various descriptions I found)
          Blast effect calculations

          • bean says:

            We don’t have even a single 40 mm gun today, sad to say, so I can’t use that. (On the other hand, Alabama has a lot, so it might work for them.) Working from the canonical reference the typical shell has 7.25 lb of Composition A, with a relative effectiveness of 1.6. Equivalent to 5.16 kg of TNT.
            Doing the math using the same system, it looks like you’d get an LD50 radius of 15 m and an LD99 radius of 5.25 m. I’m rather suspicious of all of the numbers involved, but it does make me wonder why soldiers in the South Pacific never tried it. The mosquitoes there were terrible.
            Actually, let’s see what the most efficient way to remove mosquitoes would be. A typical ROF for the 40 mm gun would be 120 rpm per barrel. So a quad 40 would have an LD 99 volume of 208.75 m3/min and an LD 50 volume of 2010.62 m3/min. A 5″/38 mount would have a typical ROF of 15 rpm per gun, while 22 was high. I’ll split the difference and call it 18. That gives an LD99 volume of 21820.7 m3/min, and an LD50 volume of 508938 m3/min. A single quad 40 weighs about 10.75 tons. Iowa’s 5″ mounts weigh 77 tons each. (Other twin 5″ mounts were lighter.) So the 5″ is definitely the way to go for killing mosquitoes, and could probably be improved with modern ammo over my numbers here.
            Or is it? While they’re not technically AA guns, there was an AA fire-control table produced for the 16″ guns. So I’m going to see what their HC shells can do. The burster is 69.7 kg. It looks like they also used Composition A, so the 1.6 ratio holds. LD99r comes to 14.6 m and LD50r is 42 m. A typical rate is 2 rpg, and the turret as a whole weighs 1730 tons. (All tons are metric, if anyone tries to check my math.) So we have 78216.5 m3/min LD 99 and 1862030 m3/min LD50. By weight, the 5″ wins.
            (As a side note, the WW2-era 40 mm shells were 68g TNT. They’ve apparently come a long ways since then.)

          • Loquat says:

            I have almost no idea what I’m talking about when it comes to military-grade heavy weaponry, but how would you actually hit the mosquitoes? Just start bombing a given mosquito-infested area, or what?

          • bean says:

            @Loquat
            You couldn’t target individual mosquitoes. They’re way too small. They’re even too small for the proximity fuse to work. (Although some WW2 models were sensitive enough to be set off by seagulls.)
            It’s proving surprisingly hard to find data on mosquito swarm density (although I’m sure someone will find a paper I should have found in a few seconds), but it looks like a high mosquito rate might run into 100 insects per cubic meter or so. Which means a single 16″ shell could theoretically kill 16.8 million mosquitoes. (That’s rather absurdly high, but it’s a nice number.)

            Oh, why am I now seeing ‘the world’s deadliest animal vs the world’s deadliest ship’ as the trailer for the movie Battleship vs Mosquito?

          • Trofim_Lysenko says:

            Hmmm, in that case enough overlapping 2+psi overpressure wavefronts should get the job done relatively well without damage to the humans from the pressure itself.

            Unfortunately that’ll still shatter windows and toss debris around, so once you add in any fragments from the bursting shells I still can’t sell this as a more entertaining way of keeping my pool parties mosquito free this summer.

            Hmmm…maybe if I pass out complimentary flak vests and K-pots….

          • bean says:

            @Trofim_Lysenko
            That sounds much more fun than a normal pool party. That might actually get me to one. I can see it now:
            “The mosquitoes are getting bad. Everyone to the bunker!”

  33. lycotic says:

    This is apropos of the thread on the Silicon Valley post that migrated into the perfidy of banks during the financial crisis.

    Wells Fargo. Why is it still a going concern? Why do people leave their money there?

    Why was the board only *almost* replaced?

    • keranih says:

      Why was the board only *almost* replaced?

      Because the people doing the replacing aren’t any better (and probably worse) at getting money out of “Wells Fargo” than the people already on the board.

      Investors don’t want your stuff or your job, they want the money that you can make from your stuff at your job. Turning stuff into money is a non-universal skill.

      • lycotic says:

        Fair enough, I shouldn’t have bothered with the last question, because it was implied by the previous two.

        The upshot is that the market signal that pushes banks toward good practices is extremely weak (although not nonexistent in this case, as WF did experience *some* loss of deposits). Part of that is low expectations (“no more perfidious than the rest of the industry”).

        As long as bad behavior is met with the equivalent of a shrug, we probably can’t expect much to change from market signals.

    • The Nybbler says:

      Why Wells Fargo, specifically? As far as I know they didn’t lose depositors’ money, and seem no more perfidious than the rest of the industry. I use Wells Fargo now, as a result of the Wachovia purchase; the alternative for Wachovia at the time was Citigroup, which doesn’t seem better.

      If you mean the fake account scandal, that was mostly perpetrated _on_ Wells Fargo corporate by its employees; they created for themselves a whopper of a principal/agent problem and got bit.

    • My bank accounts are with Wells Fargo. Why would I change? The whole scandal didn’t affect me at all. It’s hard for me to judge if the whole thing was even significant, or simply overblown like journalists and activists normally do. Every large company is surely doing some immoral things if you dig deeply enough, simply because there are so many employees.

  34. Anonymous says:

    If you baked a multivitamin pill in 180 C, how much of its vitaminous properties would be lost? Would it become poisonous?

  35. Nancy Lebovitz says:

    http://existentialcomics.com/comic/13

    Choice, will power, obedience, self extended through time, etc.

    I know someone who stopped smoking by deciding he was going to stop on a significant date about six months from when he made the decision.

  36. Tekhno says:

    Are there any sci-fi stories set in IC 1101?

    • andrewflicker says:

      None that I know of. See: https://en.wikipedia.org/wiki/Galaxies_in_fiction

      I’m guessing you’re interested in the fictional opportunities offered by a particularly large galaxy?

    • John Schilling says:

      Is there a fiction writer anywhere who has made full use of even an average galaxy? Most SF of “galactic scope” would be unchanged if the galaxy in question were four or five orders of magnitude smaller than the Milky Way, and indeed would hold up better to the sort of analysis writers would prefer fans not get up to.

      • Paul Zrimsek says:

        I usually count myself lucky if an SF author manages to describe an individual planet in a way that makes it sound at all planet-sized.

      • sflicht says:

        I feel like there are some Drake Equation assumptions hidden in this critique. Like, to take Star Wars as an example, the fact that only O(10) systems are plot-relevant (and only O(10^4?) systems seem to be represented in the Galactic Senate) is probably consistent with known constraints on Drake factors. (So little is known, it’s hard not to be consistent.) Maybe you also need some standard assumptions about Pareto distribution of planetary resources, and perhaps certain assumptions about interstellar economics, in order for the rarity of Coruscants to be sensible.

      • Sfoil says:

        The Count to a Trillion books at least make a sincere attempt to describe galactic-scale civilizations/organisms/intellects in a manner that properly conveys their scale. I think they’re great but I don’t like talking them up too much as they won’t appeal to everyone.

        • John Schilling says:

          I think the first one was the most annoying libertarian nerd wish fulfillment fantasy this side of L. Neil Smith, and he’s tried to keep that element alive even as the book departs from any domain in which merely human politics or neurotypicality could be plausibly relevant. But you’re right that he does Grand Cosmic Scope like almost no one else in the business, and for that alone I’ll keep reading.

          • Deiseach says:

            I think the first one was the most annoying libertarian nerd wish fulfillment fantasy this side of L. Neil Smith

            If you’re still talking about the “Count to the Eschaton” series, the books I found most annoying in that regard were the Golden Oecumene trilogy as I don’t find those politics palatable, but the Cosmic Scope and Big Dumb Objects and Sensawunda were enough to keep me reading (also Atkins! Who does not love Atkins?)

            I don’t mind the same in Count to a Trillion et al. because Menelaus is such an idiot, it’s enjoyable 🙂 (Also, I am hoping so hard for a redemption arc for del Azarchel, even though he keeps doing his damnedest to blow any such chances out of the water).

            The bit in “The Judge of Ages” where Menelaus and del Azarchel are gearing up for their huge final climatic (yet again) battle when the wind is taken out of their sails by the new successor civilisation that says “We’re fed up of you two running and ruining everything, we’re in charge now and you don’t get to have your big fight” is fantastic.

            People I have fallen hopelessly in love with in the Count to the Eschaton series: Sir Guiden, Mickey the Witch

            People I find myself liking despite everything and not wanting to like them one bit because they are dyed-in-the-wool villains: Ximen “Blackie” del Azarchel, Reyes y Pastor (if he hadn’t died a martyr’s death, I would have blamed him for everything)

            People who are such thundering idiots I can’t help but like them even as I despair of them: Menelaus Montrose, Our Hero 🙂

          • John Schilling says:

            the new successor civilisation that says “We’re fed up of you two running and ruining everything, we’re in charge now and you don’t get to have your big fight” is fantastic.

            That was indeed fantastic. I’m hoping for an even better version when Princess Rania gets back and finds out what her suitors have been up to for the past eighty thousand years.

  37. Comments above have mentioned a defining feature of Gray Tribe as being “conspicuously upset about the War on Drugs.”

    After fading in recent years, with major de-escalation under Obama, it appears that the Trump Administration is reviving it:

    * Welcome to the ‘War On Drugs,’ Redux: Punishing low-level drug offenders is back in style.

    * Sessions Takes First Big Step Toward Bringing Back the War on Drugs

    * Jeff Sessions might bring back the war on drugs with harsher sentences for low-level offenders

    I’m not going to talk about my own reaction. Rather, my point is about the cognitive dissonance among those who (1) opposed the War on Drugs, and (2) voted for Trump.

    My theory is that this will work exactly like the abortion issue.

    As the parties polarized on social issues, there used to be many dire/gleeful predictions that Republicans coming out strongly in favor of outlawing abortion would drive pro-choice Republicans out of the party, while Democrats coming out strongly in favor of abortion rights would drive pro-life Democrats out of the party.

    Probably there was a little of that. But overwhelmingly, people forced to choose between party identity and abortion views put party first, and adjusted their views to match their partisanship.

    We’re still in the early stages of this new transformation, but I don’t expect the administration to back off on anti-drug policy. As we get into the 2018 election cycle, and more drug users are arrested and imprisoned, the advisability and extent of the crackdown will become an even more partisan issue, something Democrats will attack and Republicans will defend. People will line up in predictable ways.

    I predict we will see a lot of new support for cracking down on drug users, from Trump supporters who formerly espoused libertarian views on the subject. Meanwhile, Democrats who used to be drug warriors will oppose it.

    None of these views, either before or after 2016, were insincere. It’s just that political polarization is a powerful force.

    • Well... says:

      Meanwhile, Democrats who used to be drug warriors

      You mean like Joe Biden?

      • BBA says:

        Like every prominent Democrat before 2004 or thereabouts. I remember in ’09 how Obama, certainly the most “dovish” president on the drug war since it became a thing, treated a question about marijuana legalization as a joke. It’s incredible how quickly public opinion has moved on this.

        Or maybe it’s incredible how public opinion hasn’t moved on abortion. Polling has been basically unchanged since before Roe, despite the massive swings in policy.

        • Well... says:

          Joe Biden has exceptionally bloody hands from the war on drugs, at least for a politician of his prominence.

          Leftist progressives started and quietly perpetuated the war on drugs; I am certain it will be people fundamentally very ideologically different from them who end it.

          • Incurian says:

            Leftist progressives started and quietly perpetuated the war on drugs; I am certain it will be people fundamentally very ideologically different from them who end it.

            Why? If it seems like a right-wing issue now, it’s a good way to score points regardless of who started it.

        • cassander says:

          I assume you’re referring to Roe as the massive swing in policy? because there haven’t been any massive swings since.

    • BBA says:

      I suspect this is true, but what happens in deeply Republican Alaska? Do they reverse course on marijuana, or does the drug war bifurcate into “marijuana can be legal in some states, everything else gets life imprisonment for possession”?

    • Nornagest says:

      Time will tell. I’m kinda skeptical that this will shake out to anything in the long run, both for demographic reasons — the favorables for marijuana are pretty clearly headed upward — and because these days I assume by default that anything reported about Trump is at least two orders of magnitude less sinister than the reporting makes it out to be.

    • The Nybbler says:

      I don’t think the states will get on board, so this will be end up being a very pale imitation of the Reagan War on Drugs. There’s a lot of muttering from the press (some of it preceding the election) about the opoid epidemic, so some amount of Doing Something was probably inevitable. Not sure what the Democratic version would have looked like, but likely not like this.

      Although, the last bit of drug-related muttering from the press took away the only effective decongestant we have, thank you Daily Oregonian for many days of a stuffed-up nose and sinus headaches. So it could do some serious damage.

    • Deiseach says:

      I wouldn’t see cracking down on drug users (if that means the dealers who use as well as deal) as necessarily a bad thing, because I’ve already seen one instance of drift (I won’t call it slippery slope because of all the times people have appeared out of the woodwork to insist that there is no such thing as a slippery slope) on the topic; someone was complaining about “and so people who only have two charges of dealing heroin or meth have their lives ruined” (or something to that effect) about tough laws sending dealers to prison. My opinion, from seeing the effects of heroin use, is “Good, send the bastards to jail for ten years!”

      You see? Already we’ve moved on from “it’s ridiculous that people who deal small amounts of harmless drugs like weed are sent to jail like hardened criminals” and “it’s wrong that a one-time dealing charge is treated this way”. We’re on to harder drugs like meth and heroin, and people who’ve already got one chance and blew it by dealing a second (or more) times.

      So yeah, I do think decriminalisation and legalisation campaigns are the camel’s nose under the tent; give in on soft drugs, then you have to give in on multiple charges of dealing, then you have to give in on hard drugs because it was only a one-off, then…

      I agree it will be like abortion: only in the case of immediate risk to the physical life of the mother, only in rare cases, then in cases of rape or incest, then in cases of foetal abnormality where there would be a stillbirth or death soon after delivery, then things like Down’s Syndrome were included, then ‘threat to the life of the mother’ became ‘to the health’, and that included mental health, then that expanded to “why ruin a young woman’s life by forcing her to drop out of education”, term limits were set, those got over-ridden by “the necessity for abortion in cases of fatal foetal abnormality that is only diagnosed at a later stage” and now it’s heading for “at any time for any or no reason so long as the woman decides she does not wish to continue the pregnancy”.

      • You take it for granted that fully legalizing drugs would be a bad thing. Do you feel similarly about alcohol? Prohibition in the U.S. is the obvious previous round of the same issues, we abolished it and no catastrophe followed.

      • Nancy Lebovitz says:

        https://en.wikipedia.org/wiki/Drug_policy_of_Portugal

        “The drug policy of Portugal was put in place in 2001, and was legally effective from July 2001. The new law maintained the status of illegality for using or possessing any drug for personal use without authorization. However, the offense was changed from a criminal one, with prison a possible punishment, to an administrative one if the amount possessed was no more than a ten-day supply of that substance.[1]

        “In April 2009, the Cato Institute published a comprehensive case study of the decriminalization of drugs in Portugal.[2] Empirical data from that report indicate that decriminalization has had no adverse effect on drug u