The culture of reasoning: the Ummah shall not agree upon error

Because I watch Screen Junkies‘ “Honest Trailers” I get recommendations like the above from Looper, The Real Reason Why Valerian Flopped At The Box Office. Of course no one knows the ‘real reason’ Valerian flopped, aside from “it didn’t seem like a good movie.” The reality is that Valerian and the City of a Thousand Planets based on a French comic book and cast a 31 year old actor who looks like a haunted 15 year old. That’s all there is to say definitively. All the various failure points are overdetermined. But the video above gives you a lot of “reasons” if you want them in a list format in a British accent. All in the service of infotainment.

Hugo Mercier and Dan Sperber’s The Enigma of Reason offers up an explanation for you have things like “top 10 reasons” for pop culture artifacts of an ephemeral nature (for a preview, The Function of Reason at Edge).

I’ve mentioned this book a few times. I finished while in the Persian Gulf (I’ll blog that at some point soon), and have been ruminating on its implications, and whether to mention it further. The issue I’m having is that I am very familiar with Sperber’s work, and those who he has influenced, and research domains complementary to his. Even if I didn’t know all the details of the argument in The Enigma of Reason, in the broad sketches I knew where they were going, and frankly I could anticipate it. I suppose somewhat ironically I managed to infer and reason ahead of the narrative since I had so many axioms from earlier publications.

The Enigma of Reason comes out of a particular tradition in cognitive anthropology. What Dan Sperber terms the “naturalistic paradigm” in anthropology. This is in contrast to the more interpretative framework that you are probably familiar with in the United States. No one would deny that the naturalistic paradigm has scientific aspirations. That is, it draws from natural science (in particular cognitive anthropology as well as the field of cultural evolution), and conceives of itself as the study of natural phenomenon.

Scott Atran’s In Gods We Trust: The Evolutionary Landscape of Religion comes out of this tradition, and some of the experimental literature in The Enigma of Reason seem very familiar from the earlier book (as well as Pascal Boyer’s Religion Explained). This is due to the fact that Atran goes to great lengths to show the ultimate nature of religion does not have to do with rational inferences as we understand them. That is, theological is a superstructure overlain atop a complex phenomenon which is not about philosophical reflection at all.

Of course the flip side can be true as well. When I was a teenager and younger adult I explored the literature on the existence of God a bit, from old classics like Thomas Aquinas’ arguments in Summa Theologica, to more recent and contrasting proofs of Norman Malcolm and Richard Swinburne (Michael Martin’s Atheism: A Philosophical Justification was actually a good sourcebook for high level arguments to theism).

When I read In Gods We Trust I realized that my earlier explorations were primarily intellectual justifications, and had little relationship why most people around me believed in God. And yet how did I become an atheist? For me this is a flashbulb memory. I was eight years old, in the public library. I was thumbing through the science books in the children’s section (particular, books on biology and medicine). The third row from the front of the stacks. And all of a sudden I had the insight that there wasn’t a necessary reason for the existence of God.

It all happened over the course of a minute or so. Mind you, I was never really religious in any deep sense. Something I’ve confirmed when talking to religious friends about their beliefs and how it impacts them. Though I nominally adhered to my parents’ religion when I was a small child, I was fascinated much more by science, and that really engaged most of my thoughts and guided my actions (contrastingly, going to the mosque was one of the most horribly boring things I recall doing as a small child).

My point here is that many of our beliefs are arrived at in an intuitive manner, and we find reasons to justify those beliefs. One of the core insights you’ll get from The Enigma of Reason is that rationalization isn’t that big of a misfire or abuse of our capacities. It’s probably just a natural outcome for what and how we use reason in our natural ecology.

Mercier and Sperber contrast their “interactionist” model of what reason is for with an “intellectualist: model. The intellecutalist model is rather straightforward. It is one where individual reasoning capacities exist so that one may make correct inferences about the world around us, often using methods that mimic those in abstract elucidated systems such as formal logic or Bayesian reasoning. When reasoning doesn’t work right, it’s because people aren’t using it for it’s right reasons. It can be entirely solitary because the tools don’t rely on social input or opinion.

The interactionist model holds that reasoning exists because it is a method of persuasion within social contexts. It is important here to note that the authors do not believe that reasoning is simply a tool for winning debates. That is, increasing your status in a social game. Rather, their overall thesis seems to be in alignment with the idea that cognition of reasoning properly understood is a social process. In this vein they offer evidence of how juries may be superior to judges, and the general examples you find in the “wisdom of the crowds” literature. Overall the authors make a strong case for the importance of diversity of good-faith viewpoints, because they believe that the truth on the whole tends to win out in dialogic formats (that is, if there is a truth; they are rather unclear and muddy about normative disagreements and how those can be resolved).

The major issues tend to crop up when reasoning is used outside of its proper context. One of the literature examples, which you are surely familiar with, in The Enigma of Reason is a psychological experiment where there are two conditions, and the researchers vary the conditions and note wide differences in behavior. In particular, the experiment where psychologists put subjects into a room where someone out of view is screaming for help. When they are alone, they quite often go to see what is wrong immediately. In contrast, when there is a confederate of the psychologists in the room who ignores the screaming, people also tend to ignore the screaming.

The researchers know the cause of the change in behavior. It’s the introduction of the confederate and that person’s behavior. But the subjects when interviewed give a wide range of plausible and possible answers. In other words, they are rationalizing their behavior when called to justify it in some way. This is entirely unexpected, we all know that people are very good at coming up with answers to explain their behavior (often in the best light possible). But that doesn’t mean they truly understanding their internal reasons, which seem to be more about intuition.

But much of The Enigma of Reason also recounts how bad people are at coming up with coherent and well thought out rationalizations. That is, their “reasons” tend to be ad hoc and weak. We’re not very good at formal logic or even simple syllogistic reasoning. The explanation for this seems to be two-fold.

First, reason is itself an intuitive process.

For the past few weeks we’ve had an intern at the office. I’ve given them a project using Python…a language they barely know. One of the things that is immediately obvious when going through pitfalls is that a lot of the debugging process relies on intuition one accrues over time, through trial and error. When someone is learning a programming language they don’t have this intuition, so bugs can be extremely difficult to overcome since they don’t have a good sense of the likely distribution of probabilities of the errors they’d introduce into the system (or, to be concrete, a novice programmer might not even recognize that there’s an unclosed loop, when that is one of the most obvious errors to anyone).

Second, reason is an iterative process which operates optimally in a social context. While  The Enigma of Reason reviews all the data which suggests that humans are poor at formal logic and lazy in relation to production of reasons, the authors also assert that we are skeptical of alternative models. This rings true. I recall an evangelical Protestant friend who once told me how ridiculous the idea of Hindu divine incarnations were. He was less than pleased with I simply switched his logic to a Christian context. But Mercier and Sperber suggest that these two features of loose positive production of reasons and tighter negative skepticism of those reasons come together in a social context to converge upon important truths which might increase our reproductive fitness.

The framework above is fundamentally predicated on methodological individualism, focusing in natural selection at that level. The encephalization of humans over the past two million years was driven by increased social complexity, and this social complexity was enabled by the powerful ability to reason and relate by individual humans. In some ways  The Enigma of Reason co-opts some of the same arguments presented by Robin Dunbar over ten years ago in Grooming, Gossip, and the Evolution of Language, except putting the emphasis on persuasion and reasoning.

At this point we need to address the elephant in the room: some humans seem extremely good at reasoning in a classical sense. I’m talking about individuals such as Blaise Pascal, Carl Friedrich Gauss, and John von Neumann. Early on in The Enigma of Reason the authors point out the power of reason by alluding to Eratosthenes’s calculation of the circumference of the earth, which was only off by one percent. Myself, I would have mentioned Archimedes, who I suspect was a genius on the same level as the ones mentioned above.

Mercier and Sperber state near the end of the book that math in particular is special and a powerful way to reason. We all know this. In math the axioms are clear, and agreed upon. And one can inspect the chain of propositions in a very transparent manner. Mathematics has guard-rails for any human who attempts to engage in reasoning. By reducing the ability of humans to enter into unforced errors math is the ideal avenue for solitary individual reasoning. But it is exceptional.

Second, though it is not discussed in The Enigma of Reason there does seem to be variation in general and domain specific intelligence within the human population. People who flourish in mathematics usually have high general intelligences, but they also often exhibit a tendency to be able to engage in high levels of visual-spatial conceptualization.

One the whole the more intelligent you are the better you are able to reason. But that does not mean that those with high intelligence are immune from the traps of motivated reasoning or faulty logic. Mercier and Sperber give many examples. There are two. Linus Pauling was indisputably brilliant, but by the end of his life he was consistently pushing Vitamin C quackery (in part through a very selective interpretation of the scientific literature).* They also point out that much of Isaac Newton’s prodigious intellectual output turns out to have been focused on alchemy and esoteric exegesis which is totally impenetrable. Newton undoubtedly had a first class mind, but if the domain it was applied to was garbage, then the output was also garbage.

A final issue, which is implicit in the emergence of genius is that it exists in can only manifest in a particular social context. Complex societies with some economic surplus and specialization are necessary for cognitive or creative genius to truly shine. In a hunter-gatherer egalitarian society having general skills to subsist on the Malthusian margin is more critical than being an exceptional mind.**

Overall, the take-homes are:

  • Reasoning exists to persuade in a group context through dialogue, not individual ratiocination.
  • Reasoning can give rise to storytelling when prompted, even if the reasons have no relationship to the underlying causality.
  • Motivated reasoning emerges because we are not skeptical of the reasons we proffer, but highly skeptical of reasons which refute our own.
  • The “wisdom of the crowds” is not just a curious phenomenon, but one of the primary reasons that humans have become more socially complex and our brains have larger.

Ultimately, if you want to argue someone out of their beliefs…well, good luck with that. But you should read The Enigma of Reason to understand the best strategies (many of them are common sense, and I’ve come to them independently simply through 15 years of having to engage with people of diverse viewpoints).

* R. A. Fisher, who was one of the pioneers of both evolutionary genetics and statistics, famously did not believe there was a connection between smoking and cancer. He himself smoked a pipe regularly.

** From what we know about Blaise Pascal and Isaac Newton, their personalities were such that they’d probably be killed or expelled from a hunter-gatherer band.

3 thoughts on “The culture of reasoning: the Ummah shall not agree upon error

  1. I would normally pass this by, but, since reference was made to both Graham Priest’s Logic: A Very Short Introduction and Michael Martin’s Atheism: A Philosophical Justification, I would note that their summaries of Classical theistic arguments are so bad as to be almost beyond belief.

    Anyway, Edward Feser has a new book out in a couple weeks on the philosophical arguments for God, which summarizes the arguments as they are in exceptionally clear and concise language.
    Feser says it should be accessible even without a background in Classical philosophy, but if anyone wants a quick primer on the concepts likely to show up, I’d recommend his short book on Aquinas.

    Feser talks about Martin and Priest in these two blog posts:

    Incidentally, Feser agrees that modernist arguments for God of the Swinburne or Plantinga-type are typically every bit as bad as they are said to be:

    Not saying anybody has to believe the Classical arguments, but it would be good to know what they actually are.

  2. I read this book based on your recommendation and really enjoyed it. I agree, though, that it didn’t delve much into the question of (1) how much variation is there in reasoning ability and (2) how much is it related to general intelligence, especially since they go out of the way to show that reason is a universal human trait. I suspect based on the stories like Newton, Pauling, etc. that solo reasoning ranges from mostly terrible to moderately terrible with occasional exceptional ability in one narrowly defined area.

    I also found it incredible that they introduced the OJ Simpson trial as an example of different standards of proof rather than bringing it up a few pages earlier as a strong counter-example to the claim that reasoning in a social context leads to better decisions. May be an example of the normative disagreements you describe.

Comments are closed.