One question, two answers, three interpretations

« previous post | next post »

My reactions to David Brooks' August 11 column "Harmony and the Dream" led me to look again at three books by prominent psychologists: Richard E. Nisbett's 2003 The Geography of Thought: How Asians and Westerners Think Differently and Why; and James R. Flynn's 2007 What is Intelligence? Beyond the Flynn Effect; and Alexander Luria's 1976 Cognitive Development: Its Cultural and Social Foundations.

I looked at Nisbett's book because it's the intellectual foundation of Brooks' column; and at the Flynn and Luria books because… well, you'll see.

There's no reference to Flynn in Nisbett's book; and Nisbett is not in Flynn's book either. Yet both are crucially concerned with how people in different places, times and social contexts interpret similarities, especially as judged by certain kinds of psychological test instruments. And both books draw important ideas, with attribution, from the same place: the research of a couple of Soviet psychologists, Lev Vygotsky and Alexander Luria, who studied the cognitive effects of modernization in Uzbekistan and Kirghizia (now the Kyrgyz Republic) in the 1930s.

Given Nisbett and Flynn's well-deserved prominence, and the importance of the various phenomena at issue, and the similarity (and common origin) of their ideas, I'm curious about the mutual lack of reference. But in this post, I'm not going to say anything more about this odd lack of explicit discussion of a strong implicit connection. Instead, I'll limit myself to a sort of catalogue of quotations from various relevant sources.

James R. Flynn's 2007 book What is intelligence? tries to understand and explain the "Flynn effect", which is a name for the fact that "the twentieth century saw massive IQ gains from one generation to another". Malcolm Gladwell's review in the New Yorker summarizes Flynn's conclusion this way:

The best way to understand why I.Q.s rise, Flynn argues, is to look at one of the most widely used I.Q. tests, the so-called WISC (for Wechsler Intelligence Scale for Children). The WISC is composed of ten subtests, each of which measures a different aspect of I.Q. Flynn points out that scores in some of the categories—those measuring general knowledge, say, or vocabulary or the ability to do basic arithmetic—have risen only modestly over time. The big gains on the WISC are largely in the category known as “similarities,” where you get questions such as “In what way are ‘dogs’ and ‘rabbits’ alike?” Today, we tend to give what, for the purposes of I.Q. tests, is the right answer: dogs and rabbits are both mammals. A nineteenth-century American would have said that “you use dogs to hunt rabbits.”

“If the everyday world is your cognitive home, it is not natural to detach abstractions and logic and the hypothetical from their concrete referents,” Flynn writes. Our great-grandparents may have been perfectly intelligent. But they would have done poorly on I.Q. tests because they did not participate in the twentieth century’s great cognitive revolution, in which we learned to sort experience according to a new set of abstract categories. In Flynn’s phrase, we have now had to put on “scientific spectacles,” which enable us to make sense of the WISC questions about similarities. To say that Dutch I.Q. scores rose substantially between 1952 and 1982 was another way of saying that the Netherlands in 1982 was, in at least certain respects, much more cognitively demanding than the Netherlands in 1952. An I.Q., in other words, measures not so much how smart we are as how modern we are.

Flynn himself grounds these ideas in an extended paraphrase of Luria (pp. 26-27 of What is intelligence?):

Today we have no difficulty freeing logic from concrete referents and reasoning about purely hypothetical situations. People were not always thus. Christopher Hallpike (1979) and Nick Mackintosh (2006) have drawn my attention to the seminal book on the social foundations of cognitive development by Luria (1976). His interviews with peasants in remote areas of the Soviet Union offer some wonderful examples. The dialogues paraphrased run as follows:

White bears and Novaya Zemlya (pp. 108-109):

Q: All bears are white where there is always snow; in Zovaya Zemlya there is always snow; what color are the bears there?
A: I have seen only black bears and I do not talk of what I have not seen.
Q: What what do my words imply?
A: If a person has not been there he can not say anything on the basis of words. If a man was 60 or 80 and had seen a white bear there and told me about it, he could be believed.

Camels and Germany (p. 112):

Q: There are no camels in Germany; the city of B is in Germany; are there camels there or not?
A: I don't know, I have never seen German villages. If  is a large city, there should be camels there.
Q: But what if there aren't any in all of Germany?
A: If B is a village, there is probably no room for camels.

The peasants are entirely correct. They understand the difference between analytic and synthetic propositions: pure logic cannot tell us anything about facts: only experience can. But this will do them no good on current IQ tests. As for the effect of attachment to the concrete on classification, the kind of thing required in the Similarities subtest, Luria (1976) serves to drive the point home:

Dogs and chickens (pp. 81-82):

Q: What do a chicken and a dog have in common?
A: They are not alike. A chicken has two legs, a dog has four. A chicken has wings but a dog doesn't. A dog has big ears and a chicken's are small.
Q: Is there one word you could use for them both?
A: No, of course not.
Q: Would the word "animal" fit?
A: Yes.

Fish and crows (p. 82)

Q: What do a fish and a crow have in common?
A: A fish — it lives in water. A crow flies. If the fish just lies on top of the water, the crow could peck at it. A crow can eat a fish but a fish can't eat a crow.
Q: Could you use one word for them both?
A: If you call them "animals", that wouldn't be right. A fish isn't an animal and a crow isn't either. A crow can eat a fish but a fish can't eat a bird. A person can eat fish but not a crow.

Note that even after an abstract term is suggested, that kind of answer is still alien. Today we are so familiar with the categories of science that it seems obvous that the most important attribute things have in comon is that they are both animate, or mammals, or chemical compounds. However, people attached to the concrete will not find those categories natural at all. First, they will be far more reluctant to classify. Second, when they do classify, they will have a strong preference for concrete similarities (two things look alike, two animals are functionally related, for example, one eats the other) over a similarity in terms of abstract categories.

(For more on the ways and means of discoursing about unseen white bears, see the passage from Tristram Shandy quoted and discussed here.)

Luria (1976) is the book Cognitive Development: Its Cultural and Social Foundations, which was published in English translation in 1976. The author was Alexander R. Luria (1902-1977), a Soviet psychologist who made contributions in a wide variety of fields, from physiological psychology to clinical neurology. The research in question is described in the Wikipedia as follows:

In 1924, Luria met Lev Vygotsky, who would influence him greatly. Along with Alexei Nikolaevich Leont'ev, these three psychologists launched a project of developing a psychology of a radically new kind. This approach fused "cultural," "historical," and "instrumental" psychology and is most commonly referred to presently as cultural-historical psychology. It emphasizes the mediatory role of culture, particularly language, in the development of higher mental functions in ontogeny and phylogeny.

Luria's work continued in the 1930s with his psychological expeditions to Central Asia. Under the supervision of Vygotsky, Luria investigated various psychological changes (including perception, problem solving, and memory) that take place as a result of cultural development of undereducated minorities

[For more on the political and intellectual context of the Vygotsky/Luria research in Central Asia, see Cosma Shalizi on "The Neuropsychology of Praxis", 3/30/2007.]

Luria's work was also featured in Walter J. Ong's Orality and Literacy, pp. 50-51:

[In their research in Uzbekistan and Kirghizia in the 1930s] Luria and his associates gathered data in the course of long conversations with subjects in the relaxed atmosphere of a tea house, introducing the questions for the survey itself informally, as something like riddles, with which the subjects were familiar. Thus every effort was made to adapt the questions to the subjects in their own milieu. [...] Among Luria's findings the following may be noted as of special interest here.

(1) Illiterate (oral) subjects identified geometrical figures by assigning them the names of objects, never abstractly as circles, squares, etc. A circle would be called a plate, sieve, bucket, watch, or moon; a square would be called a mirror, door, house, apricot, drying-board. Luria's subjects identified the designs as representations of read things they knew. They never dealt with abstract circles or squares but rather with concrete objects. Teachers' school students on the other hand, moderately literate, identified geometrical figures by categorical geometric names: circles, squares, triangles, and so on (1976, pp. 32-9). They had been trained to give school-room answers, not real-life responses.

(2) Subjects were presented with drawings of four objects, three belonging to one category and the fourth to another, and were asked to group together those that were similar or could be placed in one group or designated by one word. One series consisted of drawings of the objects hammer, saw, log, hatchet. Illiterate subjects consistently thought of the group not in categorical terms (three tools, the log not a tool) but in terms of practical situations — 'situational thinking' — without adverting at all to the classification 'tool' as applying to all but the log. If you are a workman with tools and see a log, you think of applying the tol to it, not of keeping the tool away from what it was made for — in some weird intellectual game. A 25-year-old illiterate peasant: 'They're all alike. The saw will saw the log and the hatchet will chop it into small pieces, If one of these has to go, I'd throw out the hatchet. It doesn't do as a god a job as a saw' (1976, p. 56). Told that the hammer, saw and hatchet are all tools, he discounts the categorical class and persists in situational thinking: 'Yes, but even if we have tools, we still need wood — otherwise we can't build anything' (ibid.). [...]

… A barely literate worker, aged 56, mingled situational grouping and categorical grouping, though the the latter predominated.  Given the series axe, hatchet, sickle to complete from the series saw, ear of grain, log, he completed the series with the saw — 'They are all farming tools' — but then reconsidered and added about the grain, 'You could reap it with the sickle'. [...]

At points in his discussions Luria undertook to teach illiterate subjects some principles of abstract classification. But their grasp was never firm, and when they actually returned to working out a problem for themselves, they would revert to situational rather than categorical thinking. They were convinced that thinking other than operational thinking that is, categorical thinking, was not important, uninteresting, trivializing.

Ong emphasizes the role of literacy, but Luria was interested in "the sociohistorical evolutionary of mental processes" construed more broadly, and in an explicitly Marxist framework. As he says in the first chapter of Cognitive Development:

Soviet Psychology, using the notion of consciousness as “conscious existence” (das bewusste Sein) as a starting point, has rejected the view that consciousness represents an “intrinsic property of mental life,” invariably present in every mental state and independent of historical development. In line with Marx and Lenin, Soviet psychology maintains that consciousness is the highest form of reflection of reality; it is, moreover, not given in advance, unchanging and passive, but shaped by activity and used by human beings to orient themselves to their environment, not only in adapting to conditions but in restructuring them.

It has become a basic principle of materialistic psychology that mental processes depend on active life forms in an appropriate environment. Such a psychology also assumes that human action changes the environment so that human mental life is a product of continually new activities manifest in social practice.

The way in which the historically established forms of human mental life correlate with reality has come to depend more and more on complex social practices. The tools that human beings in society use to manipulate that environment, as well as the products of previous generations which help shape the mind of the growing child, also affect these mental forms. In his development, the child's first social relations and his first exposure to a linguistic system (of special significance) determine the forms of his mental activity. All these environmental factors are decisive for the sociohistorical development of consciousness. New motives for action appear under extremely complex patterns of social practice. Thus are created new problems, new modes of behavior, new methods of taking in information, and new systems of reflecting reality.

Nisbett's book doesn't discuss Vygotsky and Luria as extensively as Flynn's did, but on p. 85 of The Geography of Thought Nisbett writes that "the school of psychology that I find myself belatedly belonging to is the historical-cultural one established by the Russian psychologists Lev Vygotsky and Alexander Luria".

For the relevance of the methods as well as the ideas, you can look at some of the papers surveyed in my earlier post ("David Brooks, social psychologist", 8/13/2008), or read e.g. Richard E. Nisbett and Takahiko Masuda, "Culture and Point of View", PNAS 100(19): 11163-11170.

Summarizing briefly: We have a single type of psychological testing method — asking which things do or don't belong together — where subjects might answer on the basis of abstract taxonomic categories, or on the basis of more concrete functional relationships. This single situation has at least three very different interpretations. The developers of intelligence tests saw shared taxonomic categories as the correct choice, and used an individual's propensity to answer that way as a measure of intrinsic mental ability. Vygotsky and Luria interpreted similarity judgments based on abstract taxonomic categories, as opposed to concrete functional relationships, as a psychological correlate of social modernization. Flynn agrees with them, and sees the last century's secular trend in IQ scores as a consequence. But Nisbett and his colleagues interpret similar differences as reflecting an enduring cultural distinction between the West and the East.

I guess these interpretations could all be correct. But I wonder.

If Flynn is right about the reasons for the Flynn effect, and if Nisbett is right about the enduring difference between East and West in affinity for abstract categorization, then Asians should have a systematic disadvantage in IQ testing. Which doesn't seem to be true.

And if Vygotsky, Luria and Flynn are right about the role of modernization in changing the salience of abstract taxonomies as opposed to concrete functional interrelationships, then you'd expect to see changes over time in the East/West differences that Nisbett and colleagues observe. They explicitly deny this, though some of their results seem to me to point in that direction.



  1. John Cowan said,

    August 14, 2008 @ 10:42 am

    To introduce a skoosh of Marxism into the debate:

    It seems to me that from the viewpoint of these Russian peasants, what they saw was a couple of crazies come from Moscow (or wherever) to ask them what were plainly trick questions; and as everyone knows, when Authority asks you trick questions, the safest and best strategy is to stick to what you know and have observed for yourself, as if you were in court.

    So it was not so much a difference in education as a difference in power, a class distinction, that induced these oh-so-concrete-operational responses. If the peasants were really as concrete as Luria makes them look, it seems to me they'd have an awful lot of trouble understanding their own folktales, which are full of strange categorical juxtapositions.

  2. Mark Liberman said,

    August 14, 2008 @ 11:44 am

    John Cowan: "…so it was not so much a difference in education as a difference in power…"

    Cosma Shilizi quotes one of the original Luria passages that Flynn paraphrased, and makes a similar point:

    The following syllogism is presented: There are no camels in Germany. The city of B. is in Germany. Are there camels there or not?
    Subject repeats syllogism exactly.
    So, are there camels in Germany?
    "I don't know, I've never seen German villages."
    Refusal to infer.
    The syllogism is repeated.
    "Probably there are camels there."
    Repeat what I said.
    "There are no camels in Germany, are there camels in B. or not? So probably there are. If it's a large city, there should be camels there."
    Syllogism breaks down, inference drawn apart from its conditions.
    But what do my words suggest?
    "Probably there are. Since there are large cities, there should be camels."
    Again a conclusion apart from the syllogism.
    But if there aren't any in all of Germany?
    "If it's a large city, there will be Kazakhs or Kirghiz there."
    But I'm saying that there are no camels in Germany, and this city is in Germany.
    "If this village is in a large city, there is probably no room for camels."

    Luria's interpretation was that Nazir-Said had difficulty with hypothetical syllogistic reasoning, as opposed to more concrete inferences in practical situations, difficulties typical of those "whose cognitive activity was formed by experience and not by systematic instruction or more complex forms of communication" (p. 115). But it's also easy to interpret this as Nazir-Said parrying the question with a perfectly valid, if enthymemic, syllogism ("Every large city has camels; B. is a large city; therefore B. has camels"), and then supporting his major premise with another valid syllogism ("Every large city has Kazakhs or Kirghiz; Kazakhs and Kirghiz always have camels; therefore every large city has camels"). The greater success of members of collective farms in "solving" the syllogisms might just reflect their greater willingness to cooperate with the Russians. In other words, there is a whole layer of issues here, involving the social relations between the scientists and their subjects, to which Luria turned a blind eye…

  3. Elihu M. Gerson said,

    August 14, 2008 @ 12:14 pm

    There's still another interpretation possible: the modernization effect operated from, say, 1870 – 1970, but has become weaker in the last generation. That is, at least some of the differences are attributable to IQ test design biases associated with a particular historical period. It'd be hard to test this though.

  4. Jonathan Mayhew said,

    August 14, 2008 @ 12:39 pm

    The ability to think in syllogisms vs. the ability to imagine conditions obtaining in an imaginary universe, one in which large cities do not have Kazahks living in them? The willingness OR ability of someone to play a game with a government bureaucrat about conditions obtaining in imaginary worlds?

    Any large city will have Kazahks; that is pretty much a given. I'm sure Cleveland has some too. So there is the basic knowledge/intelligence that once a town reaches a certain size, there will be people of different ethnicities passing through or living there. (A village will be monoethnic.) There's a kind of basic probabilistic intelligence working here too: the greater the size of the town, the more likely there will be Kazahks, and thus camels, given the metonymic connection between them.

    Germany, with no context given, is irrelevant. Were the subjects told that Germany is not in central Asia?

    Plus, I'm pretty sure that there are camels in Germany.

  5. jlr said,

    August 14, 2008 @ 1:19 pm

    This reminds me of an incident in college where a bunch of us were discussing some ethics problem, akin to the Trolley Problem, which involved possibly killing people. One woman, an otherwise very smart woman, got really upset. The conversation went roughly like this:

    Her: "I can't believe you're talking about killing people."
    Us: "They're not real people, they're hypothetical people."
    Her: "All people are real people."

    My mind boggled.

  6. A.C. said,

    August 14, 2008 @ 1:56 pm

    Just to change the discussion topic a little bit, about halfway through the example conversations, I understood, and can now explain why I often do not like fantasy fiction. Most of the stories are usually set in a medieval cultural context. Yet in that context, the writers (many are guilty of this) present characters that act and think like people who were raised in modern western civilization. It is this incongruity that frequently makes it difficult for me to like the story, or the genre. As the column illustrates, one big difference between a medieval culture where few are literate and modern western culture is, as pointed out, thinking in situational terms versus thinking in abstract classifications. If you have had to study writings by pre-modern authors (biblical or Greek or Latin authors), some of that difference comes through despite the literacy of those ancient authors. Modern fantasy authors trying to reproduce that kind of environment in their fiction could probably improve their stories if they read more ancient literature, abstracted the way people in those cultures tended to think and applied them to their characters to make them more believable.

  7. Matt McIntosh said,

    August 14, 2008 @ 2:47 pm

    East Asians actually have a somewhat lopsided cognitive profile: they actually do worse than Europeans on the "verbal" portions of the WISC, but their higher non-"verbal" scores make up for this in their total IQ score.

  8. dr pepper said,

    August 14, 2008 @ 4:12 pm

    @ A.C.

    As an avid reader of fantasy fiction, i think you misunderstand. Most fantasy worlds are not medieval, they are fantasy. And fantasy worlds have there own conventions, or rather sets of conventions. One of those conventions is the existence of people with what we would consider modern outlooks.

  9. Ann said,

    August 14, 2008 @ 4:34 pm

    Jonathan, there are no camels in Germany! Pay attention!

  10. Jonathan Mayhew said,

    August 14, 2008 @ 4:49 pm

    Does that sentence mean "There are no camels in Germany."


    "Suppose we agree to stipulate, as the first premise of a syllogism, that there are no camels in Germany…"


    "Imagine an fictional country, called 'Germany,' in which there are no camels."

    Those are very different language games we'd be playing. To understand a (probably dubious) declarative sentence in the second or third way requires some context. Are we playing a logical game or an empirical one? In the second case we'd have to figure out what the holding of German zoos were at the time of the experiment, etc…

  11. not feeling very nonymous said,

    August 14, 2008 @ 5:28 pm

    I am an avid reader of fantasy fiction as well, but this bugs me too. (In fact I think Diana Wynne Jones, an avid writer for fantasy fiction, discusses this in her Tough Guide to Fantasyland.) More exactly, I could better enjoy the inhabitants of a fantasy world thinking like early 21st-century middle-class Americans if there was a story-world reason for them to do so. Without that, it's world-building by lack of imagination. C. J. Cherryh is one of the few authors where I don't have to make these allowances, maybe because she was trained as a classicist.

  12. Skullturf Q. Beavispants said,

    August 14, 2008 @ 6:12 pm

    I have an anecdote that I believe may be somewhat relevant. Around grade 8, I was in a class that was being introduced to syllogisms. One argument we considered had two premises:

    Premise 1: If the whistle is blown, Snoopy will bark.
    Premise 2: Snoopy is not barking.

    The class was then asked if we can conclude that the whistle is not being blown. Interestingly, many students objected to this, saying things like "But somebody could be covering his ears!"

    I guess to try to summarize it succinctly, what you have is the students not really agreeing to play the same "language game" that the instructor intended. What was intended was to disregard one's common-sense real-world knowledge that no dog is going to bark without fail absolutely every time somebody blows the whistle, and to replace it with a strict literal interpretation of the first premise: If the whistle is blown, then Snoopy will bark.

    And of course, we students were from the same culture as the instructor and the textbook writer, so this anecdote says nothing about how different cultures reason differently. But it perhaps does say something about subjects' reluctance to play exactly the same language game as the authority figure.

  13. Stevehar said,

    August 14, 2008 @ 7:57 pm

    If I remember Northrop Frye's distinctions correctly in the Great Code, he also noticed 3 distinct, sucessive language games:
    -the saints, who had the original pre-biblical concrete experience
    -the scribes, who [like Luria] codified stories into a standard reference work [the bible]
    -the hermeneutical specialists [my term; aka textual literalists vs enlightened liberals]

    In the latter case, there remains nothing save dispute as to whether there really are any such things as Germanys and camels; and the decision making standard may be lost too [eye of a needle test].

    David Brooks seems to have been tap dancing too often in the 3rd game and may wish for a lawn chair and some light summer reading like the fantasy A Canticle for Leibowitz and Alasdair MacIntyre's After Virtue before he writes his endorsements for the Fall campaign.

  14. Robert said,

    August 14, 2008 @ 8:40 pm

    There's a 'riddle' that starts "If a rooster laid an egg" or something like that. The "correct" answer is supposed to be the peasant answer, to object to the hypothesis that a rooster lays an egg. The standard "wrong" answer is to silently mentally convert rooster to hen. Whereas if I used (non-paraconsistent) logic, I would conclude that from ¬A I may conclude A -> B for arbitrary propositions A, B.

  15. Nick Lamb said,

    August 15, 2008 @ 6:28 am

    Refusing to follow the authority figure's line of thought does Lewis Carroll's tortoise very well up against Achilles. By refusing to accept a rule of logical inference the Tortoise prevents Achilles from making any progress in his argument. Refusing to accept the "White bear" hypothesis seems just as smart.

    How about testing people on two group splits, so they're given four objects and told that they should find two different ways to group the objects and explain their grouping.

    I'd expect that a modern, literate, Western teenager would be able to split "Axe, Cup, Water, Wood" into "Axe, Wood" and "Cup, Water" as well as "Axe, Cup", "Wood, Water" and explain why.

    If I'm right then the apparent IQ rise does represent an increase in some potential, (even if not necessarily a very interesting one) rather than just a shift in emphasis. The ability to classify things by how they're related isn't gone, it's just less important than abstract groupings now.

    Of course it could be that illiterate 60 year old villagers who've never seen a white bear can do this same trick, and the only difference is that they'd offer the categories in the opposite order – in which case the problem was that IQ tests were poorly designed all along.

  16. dr pepper said,

    August 15, 2008 @ 9:13 am

    Actually, i should add that i have one peeve about fantasy fiction, whether written by a classicist or not, and indeed a lot of straight historical fiction, namely the use of the word "fire" to refer to shooting with bows in a culture that has no firearms.

  17. majolo said,

    August 15, 2008 @ 1:43 pm

    This reminds me of the old riddle:
    Q. If you call a tail a leg, how many legs does a dog have?
    A. Four. Calling a tail a leg doesn't make it one.
    (And am I the only one wondering why someone would call a square an apricot in the Ong quote?)

  18. Bill Benzon said,

    August 15, 2008 @ 2:04 pm

    Let me offer a piece of evidence that seem relevant to this discussion. In his 1992 review of the literature on folk taxonomy (Ethnobiological Classification: Principles of Categorization of Plants and Animals in Traditional Societies, Princeton) Brent Berlin observes that pre-literate peoples generally lack terms corresponding to plant, animal, and living thing (pp. 190-195). There is, however, no reason whatever to believe that these peoples are confused, either perceptually or behaviorally, about the differences between plants and animals. Further, Berlin gives examples of morphosyntactic markers that occur with the names of plants and animals and thus, in effect, assign them to semantic classes.

  19. Timothy Perper said,

    August 18, 2008 @ 4:44 am

    Many years ago — decades in fact — when I was in grad school studying biology, some of us were teaching assistants and had to teach introductory biology. Some genius (unspecified, but a faculty member) had concocted an exercise in "taxonomy" in which the students were given an odd collection of objects, like a wooden ball, a mallet, a stuffed duck, a box of nails, all sorts of things and were told to "sort them into categories." The idea was to elicit from the students certain "natural categories" (e.g., wooden ball, wooden mallet are part of the "natural" category "wooden objects"). We TAs were then supposed to explain that they had discovered the bases of biological taxonomy as a science.

    The results were genuinely wonderful. NO ONE, never, sorted this motley collection the same way (I'm sorry I've forgotten all the objects). The students weren't playing power games with us, and they weren't trying to be stupid peasants (for whom the Russian word is "bolvan" = ox). They were discovering a deeper truth: categories can be created in an immense variety of ways, all supportable by the evidence and by reason and all of them quite "correct." Expert: "There are no camels in German cities." Peasant: "You've been to all the cities in Germany?" One of our students would have said, "But I've been in Germany and we saw camels in the Dresden zoo!"

    Yes, if we teach certain things to students and test them on it, they'll learn to gve the right answers. "No, there are no camels in Dresden." But they won't believe us. In fact, they think we're crazy. Just like the peasants who shook their shaggy heads — they're Russian peasants with long beards, as in Tolstoy — and said that we are bereft of the good sense that God gave us all… and, you know, I'm beginning to agree with the peasants. As Bill Benzon just implied, no, these peasants aren't stupid.

    Tim Perper

  20. Mark Liberman said,

    August 18, 2008 @ 5:23 am

    @Timothy Perper: This sort of thing remains a problem for the annotation of biomedical text with "entity type" as prescribed by an ontology: see "Ontologies and arguments", 11/11/2003; "Their GO-mark'd love", 10/23/2004; "The post-modern web", 4/20/2007.

  21. Timothy Perper said,

    August 18, 2008 @ 12:03 pm

    Neat — thanks for the references. I went on in graduate school to get a doctorate in genetics, and I think I can assure you that it's worse than you think in genetic bioinformatics. All the really important stuff lurks in complex n-th order interactions among sentences (whatever a "sentence" or part thereof may be) but those interactions are lost when the stuff is boiled down by either human abstracting or machine summarization. For example, Watson and Crick 1953, wherein they described the structure of DNA: the entire, world-shaking crux of that paper is in the last sentence, which implies but does not state the basic principle of modern genetics. That's the sentence that starts "It has not escaped our notice…" and — trust me — it didn't escape anyone else's notice either. Not after THAT sentence it didn't.

    Something else occurred to me about those peasants (aside, I mean, from the possibility that Kirghiz and Uzbeki peasants aren't shaggy like Tolstoy's peasants). The questioner asserts a premise — "There are no camels in Germany" — and the peasants, wise folk that they are, do NOT assume that this statement is true. It might be; it might not be; I don't know, I've never been there. It's like saying that Vanya, you know, Vanya the Idiot, the crazy one, sees demons in the forests… Maybe he did (it's hard to say what Vanya actually sees) and maybe he didn't. Either way, speaking for myself, *I* have never seen demons in the forest. So maybe — or maybe not — there are green demons in the forest. Can't rightly say…

    Another answer, which illustrates the same principle, is to reply solemnly, "If God wills it, there are camels in Germany." In brief, the peasants saw right through the counterfactual and avoided the trap posed by all counterfactuals. Clever old peasants, I'd say.

    Tim Perper

  22. Timothy Perper said,

    August 18, 2008 @ 12:12 pm

    Oops — I forgot something. Sorry. It's no answer to say that these peasants lack "abstract" logical abilities based on saying "Let us ASSUME that there are no camels in Germany…" and then reasoning (abstractly) based on such a premise. That is **exactlly** what they *are* doing. They are refusing to make such an assumption, just as we, today, might refuse to accept Vanya's word for it that he has seen with his own eyes the demons eating chickens in the forest. They see the trap, and know very well that you too might be as crazy as Vanya. And if you'd like a real example, you can read Tolstoy's "Ivan Durak" — "Ivan the Idiot" — about a man who surely is an idiot, a fool, a ninny. Just look at him — he refuses to lie, he works with his hands, he earns his bread by his own labor… what an idiot!

    Tim Perper

  23. Bill Benzon said,

    August 18, 2008 @ 2:47 pm

    Just to clarify where I stand on the matter of those Russian peasants (I've read Luria's book cover to cover, though not recently), I'm inclined to go with something like Luria's interpretation, that the peasants lacked techniques for certain kinds of abstract reasoning, which is not at all the same as being stupid. That is to say, they weren't playing power games, but they didn't understand what was being asked of them either.

    Here's another example from Luria (pp. 81-82):

    What do a chicken and a dog have in common?

    "They're not alike. A chicken has two legs, a dog has four. A chicken has wings but a dog doesn't. A dog has ears and a chicken's are small.

    You've told me what is different about them. How are they alike?

    "They're not alike at all."

    Is there one word you could use for them both?

    "No, of course not."

    What word fits both a chicken and a dog?

    "I don't know."

    Would the word "animal" fit?


    Here the subject knows that chickens and dogs are animals, but nonetheless has to be prompted to explicitly state that. I'm quite willing to say that that – and scads of other examples in the book – is an example of difficulty with abstract thinking. I'm also inclined to think that such a conclusion is mostly a fancy lable and a place-holder for a real explanation. And the real explanation will likely require some detailed knowledge of the underlying cognitive mechanisms.

    The same goes for that peculiar business of pre-literate cultures lacking terms comparable to plant and animal while recognizing such categories by other means. I take that as a clue to the operational characteristics of the underlying mechanisms, but I don't know of anyone who's offered an explicit computational model that behaves in that way. That is, whatever abstract reasoning is, it isn't some mental module that oozes inferences whenever it's tickled. It's a pile of techniques, some of which may be universal, while others are culturally specific.

    Getting back to Shalizi's disagreement with Luria, I suspect that Shalizi misunderestimates the difficulty of syllogistic reasoning. No doubt such reasoning is trivially easy for him and just about everyone with whom he interacts. But that in itself doesn't make syllogistic techniques universal.

    Counting out a dozen marbles is also trivially easy, but that doesn't make it a universal capacity. We know it isn't. I would be nice to have an explicit computational model in which counting "one, two, lots and lots" is easy and natural, while counting out a dozen of anything is not. Now, how do you train the model so that it can count out a dozen?

  24. Timothy Perper said,

    August 19, 2008 @ 1:35 am

    Let me give another example, a real one that happened to me last week. Last Friday I went out to my corner tavern to meet friends, have a few drinks, meet my wife, go home and have dinner. We usually all sit in one particular corner table. Well, when the place opened at about 4 PM, I walked up and saw they had a new waitress at the door outside. I said hello, and asked her if my table was empty.

    She gazed at me a moment. “Which one is your table?” she asked. “They’re all empty.”

    “Idiot!” we might say to ourselves. “You are unable to Reason Abstractly! If ALL the tables are empty, then no matter which one is his, it too is empty. Therefore the only rational answer is ‘Yes, it is.’”

    You all know the cliché — women can’t reason abstractly, and here’s an example. It’s **obvious**.

    Or is it?

    Imagine that she had said that my table was empty. My response MIGHT have been: “Oh yeah? You don’t know which one is my table, what are you talking about? You’re brand new here, and you don’t know anything about this place or me!”

    Now look at what she actually said. She was being polite, and forestalling possible hostility from some crazy customer whom she doesn’t know. So she asked me **politely** which one was mine. Then she said, reassuringly, that all the tables were empty — an invitation to sit wherever I wanted, including the one that is “mine,” which she has now just told me was also empty (like all of them).

    In plain English, she covered all her bases politely and courteously. She was behaving **in context** and her answer implies that she had (without effort) thought out all the possibilities. She was reasoning abstractly in context.

    “Thanks!” I said.

    We must avoid the elitist trap of assuming that the subaltern (peasants, women, Blacks, cabdrivers, laborers, and so on endlessly) are incapable of the highly trained, elite thought patterns that WE, members of a self-proclaimedly superior group (men with higher educations), want to claim as **our own unique prerogative and ability.** Well, ‘taint so…

    Another example, which I heard long long ago. You may know a different version, but this is how I heard it — which seems appropriate given that it's Russian. Seems like in the early days of the Russian Revolution, when they were electrifying Russia, a group of engineers wanted to put a huge multi-ton electrical generator into a hole they had dug for it (in the dead of winter, no less). So here’s this 18 foot deep hole and there’s a multi-ton generator. Can’t pick it up — no cranes; can’t roll it in — that’d smash it; what to do?

    The peasants are watching. “Huh,” says one —

    Can you guess? Don’t peek now.

    “Fill the hole with water and let it freeze overnight. Then shove the generator on top and let the ice melt tomorrow while you drain the water out.”

    Now, who are we saying can’t reason abstractly? Us or them?

    Don’t be too quick to say that this peasant was just being “shrewd” whereas we “reason abstractly.” That’s because if you do, I’m going to ask next what the difference is between shrewdness and abstract reasoning.

  25. John Cowan said,

    October 23, 2008 @ 2:15 pm

    Nick Lamb: The Tortoise doesn't so much reject the rule of inference as turn it into a theorem of the system for which she (rightly) demands proof. On the one hand, this shows that logical rules of inference can't be logically demonstrated (I'm a natural deductionist, so I don't have to worry about axioms); on the other, if you don't accept them, whatever do you accept? If I hear you denying "A or ~A", I probably conclude that there is a problem in translation.

  26. Space Kossack said,

    July 20, 2009 @ 12:58 pm

    > “Fill the hole with water and let it freeze overnight. Then shove the
    > generator on top and let the ice melt tomorrow while you drain the water
    > out.”

    A hole 18 feet deep in the ground. In almost any climate that is a large depth of water to freeze and unfreeze so quickly during the course of 24 hours from only the daily temperature changes. Maybe this is special hole in country of Space Germany on planet Mercury, where there are no Kazakhs.

RSS feed for comments on this post · TrackBack URI

Leave a Comment