A Voter’s Guide to Thinking

Scott Adams (the Dilbert guy) offers a Voter’s Guide to Thinking that is so good I am going to reproduce it in its entirety:

  1. If you are comparing Plan A to Plan B, you might be doing a good job of thinking. But if you are comparing Plan A to an imaginary situation in which there are no tradeoffs in life, you are not thinking.
  2. If you see quotes taken out of context, and you form an opinion anyway, that’s probably not thinking. If you believe you need no further context because there is only one imaginable explanation for the meaning of the quotes, you might have a poor imagination. Sometimes a poor imagination feels a lot like knowledge, but it’s closer to the opposite.
  3. If a debate lends itself to estimates of cost (in money or human suffering) and you aren’t willing to offer an estimate in support of your opinion, you don’t yet have an opinion.
  4. If you are sure you know how a leader performed during his or her tenure, and you don’t know how someone else would have performed in the same situation, you don’t actually know anything. It just feels like you do.

  5. If something reminds you of something else (such as Hitler, to pick one example) that doesn’t mean you are thinking. That just means something reminded you of something. A strong association of that type can prevent you from thinking, but it is not itself a component of reason.
  6. Analogies are not an element of reason. Analogies are good for explaining things to people who are new to a topic. If I am busy as a beaver, that does not imply that I also build dams by gnawing on wood. It just means I’m busy.
  7. If you think your well-informed and reasoned opinions as a voter are bringing up the average, let me introduce you to the 100% of other voters who believe they are bringing up the average as well.
  8. If your opinion is based on your innate ability to predict the future, you might be employing more magical thinking than reason. The exceptions would be the people who use data to predict the future, such as Nate Silver. That stuff is credible albeit imperfect by nature. Your imagination is less reliable.
Print Friendly, PDF & Email
Share

31 Responses to “A Voter’s Guide to Thinking”


  1. 1 1 J-man

    9. If you thought of sending this to someone else because you think it explains their thinking and actions, but have not thought how it applies to you…start reading from the beginning, slowly.

  2. 2 2 Doctor Memory

    Much of this is very good, but #4? Not only not true, but the polar opposite of true, and maybe Mr. Adams need to re-read C.S. Lewis. (Not even the heavy apologetics; merely the childrens’ books. Aslan was cogent on this subject.) If you “know” how someone else “would have” performed in the same circumstances, you are at best speculating but most often in politics you are reciting the press releases of an interested party in the enviable position of stating a position entirely without risk of having to implement it.

  3. 3 3 Alan Wexelblat

    First of all I think it is churlish of you to reproduce this in its entirety rather than just linking to it. I doubt much of Mr Adams’ income or satisfaction comes from site visit numbers but still this is a significant and unnecessary copyright violation.

    As to the specific points in the list, I wholeheartedly agree with Doctor Memory on point 4. We are none of us privileged to live in alternate worlds so at best we can speculate – we cannot “know”. Whether it’s politics (how would Gore have performed had he become president instead of Bush) or sports (how would Player A have performed had he been named club manager instead) or corporate leadership (who would have done better than Ballmer as Gates’ successor) we are all in the land of speculation.

    As a psychologist, I will further state that #5 is also flat-out wrong. We (humans) reason in significantly associative ways. I would go so far as to say associative reasoning is a dominant mode of cognition as it is clearly absolutely necessary for many major cognitive functions, not least of them vision. How else do we recognize a new “car” as a car except by comparing it to our previous experiences of “car objects”? A thing that reminds us of nothing else is a singularity and we simply could not function as thinking beings in a world of nothing but cognitive singularities. This is one of the modes of early infant cognition, and we rapidly grow out of it. People who lose this ability (e.g. through brain damage) rapidly become nonfunctional.

    As a philosopher (though not precisely a logician) I would also take strong objection to #6. Not only are analogies a part of formal logical reasoning systems, they are also a crucial component of reasoning in other systems. Take, for example, legal reasoning. The law approaches new things by determining the appropriate analogies with which to reason. Famously, law students are taught about the first cases involving searches of mobile homes where it was important whether the mobile home was analogous to a house (in which case it was afforded certain very strong 4th Amendment protections) or analogous to a car (in which case weaker protections applied).

    I understand the humor value of saying that the entire legal system is not based on reason, but in serious conversation that would be a counterfactual statement.

    Having struck three of the eight items for being simply wrong I find myself less impressed by this list than I was reading the first two points.

  4. 4 4 Rainmount

    Scott Adams: “If you judge a candidate’s performance in some situation without a baseline then you are not thinking”

    Doctor Memory: “If you judge a candidate’s performance relative to an unverifiable baseline supplied by a partisan pundit then you are not thinking”

    They both look right to me.

  5. 5 5 Ahmed

    The problem with voting is uninformed voters. You can study the issues to death and on election day, ten uninformed voters will show up and dilute away the value of your vote.

    Worse yet, it is these uninformed voters that the politicians pander too. Most issues we vote on are economic in nature and most people have a poor grasp of economics, often voting for things which are against their interest. As the saying goes, economics is the queen of the counter-intuitive sciences.

    Democracy sucks. It means your life is governed by idiots.

  6. 6 6 Jonathan Kariv

    Two quotes come to mind.

    1. The best argument against is a five minute conversation with an average voter.

    2. Democracy is the worst system ever tried, except for all the others.

  7. 7 7 Ken B

    Hmmm. I’m a little skeptical of #4. Well, more than a little!

    I know how Hitler performed in his tenure. I’m pretty sure Trump would do better. That unthinking?

    I took a course in economics taught by a three year old chimpanzee. It did not go well, and I think Steven Landsburg might be a better choice as teacher the next time. Unthinking?

  8. 8 8 Roger

    Adams has been predicting Trump will win big in Nov. 2016.

    Nate Silver has been predicting that Trump has very little chance of winning the nomination.

    Much of the criticism of Trump consists of comparing him to Hitler, or calling him a fascist, or other indications of not thinking.

  9. 9 9 Jonathan Kariv

    Meant
    1. The best argument against *democracy is a five minute conversation with an average voter.

    Sorry for the double post

  10. 10 10 Ken B

    Alan W’s comment on 5 is obtuse. Adams is not denying intuition. He is denying a certain license to let one’s prejudices run free. It is a warning about confirmation bias.
    And seriously, you deny that a strong emotional reaction based on a vague feeling can be a hindrance to thought?

  11. 11 11 Alan Wexelblat

    @Ahmed I recommend you read some of the writings of noted (conservative) legal scholar Ilya Somin on the issues of uninformed voters. I’m not able to recapitulate here the last 7+ years of Somin’s work but suffice it to say that he argues strongly that uninformed voters are not only not a problem, but completely rational and logical.

  12. 12 12 Ken

    0. Should politicians be given power over citizens on this particular subject is the only place to properly start. If you’re starting the conversation by merely assuming that politicians have power over whatever topic you’re talking about, you’re probably not thinking.

  13. 13 13 David

    Adams’ hypothesis on #4 states “and you don’t know how someone else would have performed in the same situation” The counterexample offered by Ken B clearly doesn’t refute #4, as we actually do know how Steve would perform teaching an economics course.

  14. 14 14 David

    @Alan – I think perhaps you are being too literal. What I believe Adams is pointing out in #5 is not that no cognition occurs when we recognize similarities, but rather that if some aspect of A reminds us of B, a conclusion that A and B are also similar in other respects is not a thoughtful one.

    Similarly, in #6 Adams is pointing out that an analogy, by definition, is two *different* things that share a commonality. When you pretend you don’t understand that fact and accusing an analogist of arguing that A and B are in fact “the same thing” you are not engaged in thoughtful analysis.

  15. 15 15 Advo

    Thankfully, for the general election it boils down to one very simple question for every voter:

    Do I want another financial meltdown within the next decade or two?

    Yes > vote GOP
    No > vote for someone else

  16. 16 16 Harold

    #2 and #3. No. 4 says “if you know how someone performed…” That is you have experience of that persons performance. Then “and you don’t know how someone else would have performed…”

    Dr. memory and Alan are saying that you never know how someone else would have performed, so this always applies.

    Whilst it is true that you never know exactly how someone else would have performed, you can get a pretty good idea if you have experience of them performing in similar situations. So in Ken B’s example of Landsburg as an economics teacher, we do have exeperience of this, so we know approximately how he is likely to perform.

    The closer the situation to one we have direct experience of, the better our estimation of the individual’s performance is likely to be.

    #5 is wrong if taken literally. Clearly a new car does remind us of cars we are familiar with, so we are justified in making certain assumptions about what this object is. We can also be very wrong – think inflatable tanks used to deliberately mis-inform the enemy. It looks like a tank, but it does not share any of the properties we associate with a tank except a general size and shape. The important properties – i.e. guns, armour and mobility, are not in fact shared, so we get that wrong, with disasterous consequences.

    The error comes when we assume properties that we have no good reason to assume. The cat with the appearance of a toothbrush moustache and black parted hair is not necessarily like Hitler. That is the mistake made by the “Cats that look like Hitler” website when they say:

    “Does your cat look like Adolf Hitler? Do you wake up in a cold sweat every night wondering if he’s going to up and invade Poland?… If so, this is the website for you.”

    http://www.catsthatlooklikehitler.com/cgi-bin/seigmiaow.pl

    In this case the association is not valid – the appearance of Hitler in a cat does npt actually make the cat likely ot invade poland.

    So how do we know whether our association is valid or not? Quite simply, we don’t, and we certainly don’t unless we question them. So SCott Adams may not be technically right, but bearing point 5 in mind will make you a better thinker.

    Ahmed #5 “ten uninformed voters will show up and dilute away the value of your vote.” Landsburg has taught us that the rational person will not vote anyway, at least withthe intention of affecting the result. So pretty much everyone that votes is irrational. What a system. But maybe better than the alternatives – except Australia.

  17. 17 17 nobody.really

    [T]he appearance of Hitler in a cat does not actually make the cat likely to invade poland.

    Have you been to Poland? Have you seen the cats there?

    Maybe we should withhold judgment until we have better data….

  18. 18 18 nobody.really

    3. If a debate lends itself to estimates of cost (in money or human suffering) and you aren’t willing to offer an estimate in support of your opinion, you don’t yet have an opinion.

    Eh. Yeah, an absolute estimate of cost is nice. But when comparing plans, sometimes it will suffice to have an ordinal estimate of cost, e.g., I don’t know how many lives we’ll lose sending in a tactical team, but it will be less than the number we’d lose by sending in a full ground invasion.

    4. If you are sure you know how a leader performed during his or her tenure, and you don’t know how someone else would have performed in the same situation, you don’t actually know anything. It just feels like you do.

    Fine. And yet, pretty much all policy discussions involve counterfactuals. We compare how Plan A would operate relative to Plan B, even though we actually implemented only Plan A. Thus, Adams’s claim boils down to a plea for humility – but it doesn’t really provide any basis for us to avoid making judgments about counterfactuals.

    6. Analogies are not an element of reason. Analogies are good for explaining things to people who are new to a topic. If I am busy as a beaver, that does not imply that I also build dams by gnawing on wood. It just means I’m busy.

    To the contrary, I think analogies – the tendency to generalize from learned and observed phenomena – are the foundation of science. Can people draw inappropriate conclusions based on analogy? Sure. But what alternative do we have?

    Yesterday my Hitler-like cat pushed my pen off the counter. It fell, just as it did the prior day. I can generalize from these experiences to conclude something about gravity. But how do I know that what happens with my pen would also happen with my pencil? Or what happened yesterday will also happen today? I can only observe the similarities, and generalize.

    I know of no rule to demonstrate the appropriate limits of generalization. Does anyone?

    Ironically, in #4 Adams disparages the practice of making judgments without establishing a benchmark against which to judge – yet in #6 he fails to articulate an appropriate benchmark for generalizations.

  19. 19 19 Ken B

    @David 11: Not so. Not THIS economics course, only previous ones. Maybe we can’t generalize in this case.

    You have fallen into my trap. We can and do infer from performance in one endeavour to success in another. Adams #4 simply denies we can do this, without any caveat.

  20. 20 20 Ken B

    Advo’s response tickles my sardonic irony sensors.

  21. 21 21 iceman

    God I hope 15 was supposed to be ironic

  22. 22 22 Ken B

    @iceman 21
    See 20.
    And my best guess, based on Advo’s history here, is no, it’s not ironic.

  23. 23 23 Ken B

    This thread is a little too quiet. Time for a little provocation. Several posters have mentioned democracy, and it’s presumed failings. But what, at its essence, is democracy? I heard a definition once I think is right:

    Democracy is government by talking about it.

    That is why the vote exists and is important. Not to harness the wisdom of the masses in making policy; not because all opinions are valuable; not because elections lead to the best results and the best decisions. If everyone has a vote then they can at least be heard, they can invigilate their rights and interests, they can try to guard against the worst result.

    Not all voices will be equal, or should be equal. But if you cede the discussion, not just the power but the entire debate, to one segment of society, the priests, the mullahs, the billionaires, the whites, the aristocracy, the professariate, then you end up with something other than a full discussion, and with some rights and interests ignored.

  24. 24 24 David

    @Ken B 19: I think you’re missing his point. Adams isn’t arguing that you can’t make inferences. He’s simply pointing out that if, for example, the year is 2007 and you disagree with Bush’s leadership choices about Afghanistan, drone strikes, the national debt, civil liberties, bank & auto bailouts, etc and are certain that when faced with the same information, incentives and constraints, Obama would make significantly different choices, perhaps your certainty is erroneous.

  25. 25 25 Ken B

    @David 24
    I agree your hypothetical falls under Adams’s rule 4. But so do many others. There is after all only one person eligible to be president who has already been president. So according to Adams, you could compare Obama to Bush 1 in 2012, but not to Romney. Romney hasn’t been in the same position. I might have an opinion about his administrative and leadership skills, but according to rule 4 I am just fooling myself. I don’t think that’s right.
    Hence my example of Steve and teaching. We can make a pretty good guess. But it isn’t a case of knowing, of certainty, is it? There is a scale, other factors do allow reasonable inferences; it is not unthinking to draw such inferences.

  26. 26 26 Ken B

    This may be a bit off topic, but Steve did apply the label Bad Reasoning to this post.

    Scott Alexander calls this the scariest stats question you’ll ever see. He’s right. https://stats.stackexchange.com/questions/185507/what-happens-if-the-explanatory-and-response-variables-are-sorted-independently

  27. 27 27 David

    Ken B 25: I agree there is a scale. I agree that other factors allow reasonable inferences. I agree it is not unthinking to draw such inferences. I disagree that #4 contradicts any of the above. I disagree that Adams is even addressing any of the above. To return to my “hypothetical,” (which actually wasn’t hypothetical at all) he’s addressing those who *know* that Obama is going to get us out of Afghanistan, etc, etc and use this knowledge to inform their vote in 2008.

  28. 28 28 Harold

    #26 – we are not told what sector the correlators are working in- I think we really need to know. However, I am sure I will find this wonderful new data analysis technique useful at some point in the future.

    #27 #4 says you know how a president did in a situation – that is you know how Bush did. You don’t know how Obama would do in the same situation. You don’t know anything, but it feels like you do.

    This is fine up until the “it feels like you do” Why does it feel like you know something? To make sense, there is an implied belief that you know how Obama would do in that situation. Otherwise it would not feel like you knew anything.

    Perhaps it should read: If you know how a leader did in a situation, and you think you know how someone else would do in the same situation but you have never actually seen that person in a similar situation, then you don’t know anything, it just feels like you do.

    The difficulty here is how similar does the situation have to be before you can say you do know something? If it has to be exactly the same situation, then we know nothing about anything – we cannot say that Landsburg will give a better economics lesson than a monkey. If it means slightly similar, then we can predict the economics lesson, but we also might erroneously predict Obama’s response to Afganistan.

    Actually, we know a bit about how anyone will respond to new situations based on their previous behavior. The best prediction of future behavior is probably past behavior. We should be particularly cautious when the behavior we are trying to predict is in a very new situation – such as being president for the first time.

  29. 29 29 Michael Rulle

    Interesting by Adams. Agree with Dr.Memory. I still smell bias however.

  30. 30 30 Ken B

    To all Christians, Happy Xmas.
    To everyone else, Merry Christmas.

  31. 31 31 RG

    Thinking is hard. The alternatives are easier. There are too few among us who choose hard over easy.

Leave a Reply