Why are human beings simultaneously capable of reasoning, and yet so bad at it? Why do we have such faulty mechanisms as the “confirmation bias” embedded in our brains, and yet at the same time, find ourselves capable of brilliant rhetoric and complex mathematical calculations?

According to Hugo Mercier, we’ve been reasoning about reason all wrong. Reasoning is very good at what it probably evolved to let us do—argue in favor of what we believe and try to convince others that we’re right.

In a recent and much discussed paper in the journal Behavioral and Brain Research, Mercier and his colleague Dan Sperber proposed what they call an “argumentative theory of reason.” “A wide range of evidence in the psychology of reasoning and decision making can be reinterpreted and better explained in the light of this hypothesis,” they write.

Given the discussion this proposal has prompted, Point of Inquiry wanted to hear from Mercier to get more elaboration on his ideas.

Hugo Mercier is a postdoc in the Philosophy, Policy, and Economics program at the University of Pennsylvania. He blogs for Psychology Today.

http://www.pointofinquiry.org/did_reason_evolve_for_arguing_hugo_mercier/

I’ve read some of this debate and I’m unimpressed. I’ve devoted considerable thought to this problem, and, while I’m not completely confident of my results so far, I think my own hypothesis on this matter is much better than Mercier’s. The big flaw in his reasoning is that he fails to recognize that rationalism is very much a Western concept, invented by the Greeks. Perhaps he should not have used the term “reason”, preferring “argumentation” instead. In any event, should anybody like to read my own hypothesis on the matter, they can start by following this sequence. I’m sorry about the indirection, but I can’t just paste the URL in, because it includes space characters, which once upon a time were usable, but nowadays are absolutely verboten. I’m in the process of revising my website, so it’ll get fixed eventually. But for now, you have to do the following:

Go to this page and then click on the link to “A History of Thinking”. Sorry about that.

Thank you for your comment. Here’s a paragraph from my website on that issue:

” It is also interesting to notice that argumentation is not a modern or a Western phenomenon. Some might have us believe that illiterate people cannot reason, or that Easterners never argue (this is too extreme, no one serious supports these views anymore, but run down versions are still common). This is not true. People argue all over the world, and all the available evidence indicates that they do it well. Also, people seem to reason better in groups everywhere—where data is available at least. And Cicero would have little to reproach to the Easterners’ tradition of argumentation and persuasion. (This is described there: Mercier, H. (in press). On the universality of argumentative reasoning. Journal of Cognition and Culture.)”

The paper can be found there:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1784902

Mr. Mercier, I believe that our difference lies in the definitions of the verb “reason” that we use. You use it as a synonym for “argue”. I use it for a particular subset of argumentation characterized by an attempt to remain within the confines of logical rigor. Certainly none of the ancient Chinese writings, and none of the ancient Indian writings satisfy my definition of reason. You are welcome to your own definition, of course, but my definition recognizes the profound differences between Western thought and Eastern thought. I’ll also point out that the huge differences between the history of science in Western civilization and all the other civilizations is readily addressed by my own definition, and not addressed by yours.

I’m mostly interested in psychological mechanisms. From that point of view, it’s the same reasoning abilities that underlie what you call ‘reason’ and other types of arguments. So whether there is an interesting distinction to be drawn between different types of arguments, some which we could call ‘reason’ and others, it’s not really at issue here. For what it’s worth, I very much doubt there’s anything to this distinction, at least in normative terms. If anything, those who follow ‘reason’ too strictly are more likely to end up doing the opposite of science, developing formal systems with no connection with the world.

I too am interested in the psychology of this issue—the explanation of my hypothesis on this issue starts with the first nervous systems over 600 million years ago and traces the development of the neural mechanisms underlying this broad kind of processing right up through the development of language. At that point, my explanation switches to historical considerations, and zeroes in on the recovery of Greek society from the post-Bronze Age dark ages around 900 BCE. That society differed from all others in that its basic power structure was mercantile, because agricultural land (which can be dominated by an aristocracy) was in short supply and was better used for tradable commodities (wine and olive oil) than for cereals. There’s a profound difference between mercantile societies and aristocratic societies: logic dominates the former and is irrelevant to the latter. An aristocratic king can be incorrect and still rule; an incorrect merchant goes bankrupt. That’s why Greece developed rationalism, and their devotion to rationalism bursts out of their writings to a degree that never shows up in Chinese or Indian writings. The Chinese adulate the notion of “the gentleman”, but that ideal is distinguished by moral virtues, not rationalism.

I’m surprised that you don’t perceive the vast difference in argumentation styles between Western thinkers and others. It really is a huge difference in the entire mentality, and has been noted by many, many observers. Some characterize it as the difference between holistic thinking and analytical thinking. There have even been psychological studies showing that Chinese subjects are more likely to detect changes in background imagery, while Western subjects are more sensitive to foreground imagery.

True, logic can be overdone; that’s one of the conclusions of my own explanation of rationalism. But there’s no denying that rationalism is the driving force that made the difference between Western civilization in, say, 1900 CE and Chinese civilization at the same time so stark.

The differences you mention between “Eastern” and “Western” mentalities are probably not as strong as you suggest. This is discussed at length in the paper linked to in my first comment. And I’m not sure many people still think that rationalism drove the industrial revolution.

I’m still reading your paper, so I’ll have to defer extended comment on that point until I complete that—and I have a variety of meetings and appointments throughout the rest of today, so it might not be until tomorrow that I can comment. However, for the moment I’ll point out the vast difference in technological achievement between Eastern and Western thought. Given the larger size of the Sinic economy and population, we would expect it to have outpaced the West in technological advance, but its almost total lack of science (as opposed to its engineering, which was developed entirely by trial and error) prevented it from coming anywhere near the West in science. The most astounding difference here is in mathematics. The calculations carried out by Copernicus were utterly beyond the comprehension of any Sinic thinker until perhaps the 19th century, possibly later. And the work of Western mathematicians such as Legendre, Pascal, Newton, Euler, and others simply leaves Sinic thought in the dust. The chasm between Western achievement and Chinese achievement in mathematics is stupendous.

As to the Industrial Revolution, it was only one component of a long process that began in 900 BC. By 1800 CE, Western rationalism had developed so far that it acted only as an enabling factor to the complicated constellation of factors that led to the Industrial Revolution. Have you read A Farewell to Alms? Great stuff, that!

I agree, on the whole) with your statements about mathematics. Yet it seems as if Eastern populations have not much problem catching up in math, even producing important mathematicians (like Ramanujan). So it seems as if the difference may be more institutional than one of mentalities.

As for A Farewell to Alms, yes, and it has a good discussion of the Malthusian trap, but the whole idea of genetic changes driving the IR is absurd (see for instance the discussion in McCloskey’s Bourgeois Dignity)

Perhaps I’m missing something, but the theory doesn’t seem to leave room for nonhuman or preverbal hominid reasoning.  Dr. Mercier is proposing that reason serves argument or, at least, social interaction leaving little room for the enormous benefit of individual, contemplative reasoning in pre or proto verbal groups that were incapable of the types of arguments he presents.  In my opinion, this theory requires another which addresses how language could have gotten off the ground in a hominid brain that had little or no capacity for reason.  The anecdote about the Piraha is as close as he gets and is far from compelling.

Thank you for your comment. I think the problem is that you may rely on a wider definition of reasoning that includes most inferences. With our definition (see for instance here:

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8242469&fulltextType=RA&fileId=S0140525X10000968

) it’s fairly clear that reasoning is not necessary for language.

Hugo Mercier - 16 August 2011 10:33 AM

Thank you for your comment. I think the problem is that you may rely on a wider definition of reasoning that includes most inferences. With our definition (see for instance here:

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8242469&fulltextType=RA&fileId=S0140525X10000968

) it’s fairly clear that reasoning is not necessary for language.

I read the paper and thanks for answering my question by pointing at it.  You define “reasoning” as not belonging to system 1 reasoning (fast, frugal, and unconscious reasoning) but instead, yours is everything else; system 2 reasoning defined as the “negative of the former” and then state that the distinction is based on the conventions of almost every sub-discipline in the field of cognitive psychology, as if researchers in these fields share a common view on this distinction. Until I know where these scientists, or even a subset of them, have drawn the line (citation?), I don’t have your definition.  I am also not exactly sure how this definition makes it obvious that reasoning is not necessary for language when the definition of system 2 reasoning includes many elements of cognition - like attention, memory, and learning - that clearly are.  I really don’t think that my definition of reasoning is wider than yours but I simply don’t have enough information to be sure.

I do thank and commend you for addressing the comments.  Regardless of any differences in opinion - or reasoning - among us, I’m sure that Chris and the community appreciate it.

It’s my pleasure to engage with people who find some interest in my work. I’m sorry we didn’t do a better job at making our definition clearer. What you’re referring to is actually not our definition but the standard definition of system 1 and system 2, or intuition and reasoning. Ours is quite a bit different, in that reasoning is much more intuitive than most people think. The next post on my blog will be exactly on that topic, so if you’re interested you can check it out, hopefully it’ll be written by the end of the week:

http://www.psychologytoday.com/blog/social-design

cheers

Hugo Mercier - 16 August 2011 10:33 AM

Thank you for your comment. I think the problem is that you may rely on a wider definition of reasoning that includes most inferences. With our definition (see for instance here:

http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=8242469&fulltextType=RA&fileId=S0140525X10000968

...

Forgive me. That is not my field; I am here accidentally. You wrote:

“Reasoning is generally seen as a means to improve knowledge and make better decisions. However, much evidence shows that reasoning often leads to epistemic distortions and poor decisions. This suggests that the function of reasoning should be rethought. Our hypothesis is that the function of reasoning is argumentative. It is to devise and evaluate arguments intended to persuade.”

Suppose the ending is changed slightly; suppose it becomes ... to persuade ourselves or others. In that case the difference between two functions (making better decisions versus becoming better at arguing) disappears.

Talking to myself, by writing a diary, used to an important part of my attempts to make good decisions (to persuade myself).
.

That’s an interesting suggestion. The problem is that it’s not clear what are the evolutionary payoffs of convincing ourself: if you believe something, they you accept it. Empirically, it also seems that people reasoning on their own, even if they think they are being objective and consider different points of view, in fact have a rather massive confirmation bias. Granted there are exceptions, cases in which we appropriately take another point of view, anticipate counter-arguments and change our mind, but that’s what they are: exceptions. In general, it’s just easier and safer to reason with others: they are likely to be more motivated to provide rejoinders and more knowledge to do so.

Well, I had some long gaps during my meetings, and I stayed up late writing this exceedingly lengthy reaction to your paper. I’ll attempt to post it all in one go, but I expect that the editor will choke on the size of the text and I’ll have to break it up into parts. Here goes:

I would like to echo shawnpat’s appreciation of your presence here; it’s always much more productive to discuss an idea with its primary creator. Moreover, I am intensely interested in a closely related topic and so I am eager to benefit from your expertise. I expect, of course, that you might enjoy some benefits by being grilled by this astute group; and I further expect you to depart as soon as you believe that you are no longer deriving any benefits—as would I.

Academics versus intellectualism
I’d like to begin my discussion of your paper with three broad observations. First, this is an academic paper whose ultimate purpose is to advance your academic career. This places a number of important constraints upon it. It must confine itself to those problems that are currently of interest in your field; should you stumble upon a major discovery in microbiology, say, or unearth topless photos of Michelle Bachmann, or reach a startling realization in Old Testament analysis, it would not be of any value to you to publish any of this information. Less sensationally, issues only peripherally related to your field would be of no value even though they might reveal truths that are important in some other field (such as paparazziology, for example).

This constraint can be harmful in highly cross-disciplinary fields such as yours. My own researches on the development of human cognition have wandered through evolutionary theory, evolutionary psychology, linguistics (especially the development of human language), history of science, neurophysiology, economic history, digital electronics, Cicero, the soils and climate of Greece, history of the rise of Greece, cognates of compound conjunctions in Indo-European languages, computer science, St. Thomas Aquinas… it’s a long list. Some of that material is only peripheral to your own investigations, and since you labor under such great pressure to publish, you must be quite discriminating in the material you examine. While this is indeed the most efficient way to proceed, it can rob you of the opportunity to explore some material that can ultimately have utility in this research. Few academics get the opportunity to engage in broad research of this type until they get tenure.

Parochialism
Second, you go to great lengths to combat Western parochialism, and I heartily applaud those efforts. I myself really don’t believe in “better” in the general sense; I instead concentrate on specifics. Yes, Western mathematics was definitely way better than Chinese mathematics throughout most of the last half-millenium. The same can be said for Western science and Western technology. But those are only a few indeces of achievement. It’s ridiculous to get parochial and declare some sort of generalized superiority. I’m surprised that you felt a need to combat such an obvious prejudice, but apparently that problem still haunts your field.

Terminology
Third, you live and breathe a highly specialized argot that concentrates your thinking along one set of lines. This argot makes it easier for you to directly address the issues of interest to your peers and judges, but it can also make it more difficult for you to break out of some confining assumptions. The most obvious example of this arises in your definition of “reasoning” as that form of linguistic expression required to convince others. In Medieval Europe, that was called “rhetoric”. There’s no gain for us to quibble over the “true” meaning of any term; if you want to use “reasoning” to be synonymous with “rhetoric”, that’s fine with me. I will certainly not quibble with your notion that reasoning is that second mode of cognition (analytic as opposed to intuitive). However, I would like to offer you a different way to slice the semantic pie that provides more analytical utility.

My own thinking on this matter divided the pie into “pattern-based cognition” versus “sequential cognition”. I now think that a cleaner terminology is “parallel versus serial”. While it doesn’t quite cover the neurophysiology well, it’s close enough to serve the purpose, and it’s a very clear distinction. Thus, what your tribe calls “intuitive cognition” I call “parallel cognition”; what your tribe calls “analytical cognition”, I call “serial cognition”.

The advantage of my terminology is that it really zeroes in on the mental processes at work. Serial cognition is a neurophysiological kluge, something that neurons really don’t do well, and that’s why it’s so damned difficult. That’s also why mammalian brains are so damned big: it takes a hell of a lot of neurons (which are essentially parallel devices) to cobble together a serial-processing system. And that’s also why human brains are so much bigger than mammalian brains: we do a LOT more serial processing. Language is the primary form of high-intensity serial processing that we do, and language gobbles up neurons by the ganglion.

This line of thinking clearly shows that formal logic, the kind of rigorous serial thinking that made Western math, science, and technology possible, is the most extreme form of serial cognition. It’s also the most difficult to do; we really have to push our brains hard to carry out that processing. My cutesy way of expressing this point is that “serial logic does to the mind what yoga does to the body—you really have to twist yourself around to get it to work.”

Nope, it cut it off. On to part II.