The Limitations And Usefulness Of IQ

That Smarts

Everybody knows that some people are smarter than others, and that some are teachable and some not. Well, some people do deny these facts, believing that if everybody started from the same place, was taught in the same way, and had identical resources, all would arrive at the same destination, so that, down deep, everybody is equally smart.

This view—equality—cannot be supported. It has been disproved time and again by all experience, in every possible situation conditioned on every kind of contingency. Believing in equality in spite of the vast, vast mass of historical counter evidence proves only that intelligent people can believe stupid things—which everybody also knows. Indeed, is has often been observed that the more intelligent the man the more outlandish the fallacy.

Thus, though I will come to certain conclusions about IQ and intelligence counter to many prevailing views, I will not claim or advise, nor even hint, that equality is a possibility, or that it is in any way desirable.

My thesis is simple. That some are more intelligent than others, that some aspects of intelligence are driven by biology, and thus some are not. Intelligence to the extent it is biological can be imperfectly inherited; that because intelligence is in part immaterial, it can only partially and incompletely be measured, and with greater uncertainty than is recognized.

IQ is said to measure intelligence. Then, through the Deadly Sin of Reification, IQ becomes intelligence. The problem is not in the idea that some are smarter than others, for this is true (for some definition of “smart”). The problem is over-certainty.

IQ is a score on a test (a function of the number of questions, the number correct, and the like). Take a different test and the chance of a different score than that computed from the first test is not zero; i.e. the scores will differ. Take the same test, or one similar, at a different time, and again one may score differently. Is this a problem? No. Not if reification has not occurred. Different scores at different times are taken to be caused by any number of things: health, distraction, question content, and on and on. We thus need to understand what is causing IQ scores.

Let’s start small and in a subject with small controversy before moving to the Big Game. Even though what’s said is rather obvious in the introduction, stick with it, because it is important.

Decathlon

Let’s look at something similar to IQ, as far as testing goes, a thing which is not as controversial, and see what happens.

Suppose we are testing a long jumper. He jumps, we measure. He jumps again, we measure again. We have two measures, probably not the same, that give us some idea of the jumping ability of this man. The outcome is distance, which is easy to measure and unambiguous. That we value distance is a judgement and culturally peculiar: not all people in all places will take an interest. We might, instead, have decided to define jumping ability as distance plus height at maximum during the jump (combined via some function). Why not? In this case, the man jumping the longest distance might not be the man with the highest score.

On any test something caused this man to jump the distance he did, or the distance plus height. There will be many causes, and of four distinct parts: the form, material, efficient, and final cause. In each case the final cause is the reason for the jump: he wanted to do it, but he may want it more or less at different times. Will is involved. The formal cause is the rule of the jump, and will be the same across measures (we presume; though like in real events, the form does change). The material and efficient causes are subject to variation. The man may have eaten a potato before one jump and a slice of pizza the next, and at different times. The man’s muscles, digestion, nerves, and so forth contribute to the material and efficient causes: you need to be a physiologist to understand them all. The track conditions change, which might vary subtly or a lot, depending.

Anyway, cause is involved, as is obvious. The man didn’t land downfield for no reason, but because of reasons. In the interest of shorthand, we might call the conglomeration of all these causes (all four) strength. That’s because it’s a burden to recall each possible cause, because they are many and complex, but mostly because what really interests us is the outcome. We all understand that cause is involved, which we’ll call strength, but it’s not the main subject. Distance jumped is. There is thus little danger in saying something like, “The man who won is stronger than the others” when what we really meant, and what all understand it to mean in this context, was that “this man jumped the farthest”.

Now under other circumstances we might also mean, “Given this man’s previous performance, i.e. his great strength, there is a good chance he’ll perform well again.” This is straightforward. We mean, removing the reification, that “Given his past jumped distances, there is a good chance he’ll jump those distances again.”

This same conclusion holds if we group men together based on jumping ability. If it turned out upon inspection men of one race had a higher average distances than others (supposing we were interested in averages and not other metrics, which are also important), then it would make eminent sense to judge a man from this race as being likely to jump farther than a man from another race before making actual measurements, and knowing only the men’s races. After we see each individual’s performance, any knowledge about race is wiped away and of no use in making judgments of (only) these men. (We can define race by self-identification, and this will do fine.)

We could say, “Given the (original) man’s demonstrated strength, I’ll bet he has high probability of also tossing a javelin a great way.” We may make many statements like that about other physical abilities. We reason those other abilities also require strength, in some way or another, and so we predict success in these other areas. These judgments, we’ll likely find on measurement, will bear out. Not always, not for every man, and not in every situation, but for enough men and situations that our confidence is justified in making predictions like this.

If we think about it, though, we realize that tossing a stick is not the same as jumping in terms of cause. The man may love jumping but sees stick tossing as a hated duty: his will is engaged differently, and variably, for the different activities. Arms are used in jumping, but not in the same way as in tossing. The rules of the contests are different. The stick itself is different. Strength applied conditionally to javelin throwing means something different with respect to cause than when applied to jumping.

Well, why not average, in some way, the observed measures (i.e. the distances) of jumping and tossing and call this average (however it is computed) strength quotient (SQ), perhaps normalized to an arbitrary scale (itself informed by making these observations over many men). If we’re not satisfied with the range of activities of what we’re calling SQ, we can add in, say, times for running a specified distance, which contains the idea of endurance, which the other sports didn’t have. We can add even more sports using similar reasoning.

There is a bit of circularity here, in the following sense. We add just those those sports we think are indicative of strength. Why not add, say, tiddly winks? Because although this can be defined as a sport (always a subjective cultural definition), we don’t believe it needs strength. To justify this circularity, we might claim (truthfully) that if we computed SQ sans tiddly winks contributions, we’d find little “correlation” between SQ and tiddly wink scores. (I use correlation in the plain-English sense, as things tending to move together, and not in its linear mathematical sense, unless otherwise specified.)

We can’t use correlation to see what performances to enter and which not, though, because of that circularity. Plus, if correlation was the sole criterion, the best correlation will be just jumping with itself. SQ would be jumping and nothing else because no other activity would correlate as highly with jumping. Correlations of jumping and tossing might be “high”, and so we include both, but any correlation cutoff we use to include or exclude activities relies on outside previous judgment. What kinds of judgments are these?

The (open) secret is that we put these things together because of our understanding of cause, it in all its aspects. We include those measures which we think have the most similar causes, in all four forms, that relate to athleticism (including will to perform). We consider blood platelets, muscle mass, twitching versus others muscle movements, nerve conduction, aerobic capacity, which is driven by inter alia lung capillary performance, which again relies via feedback on blood workings, and on and on. How do we know that these causes relate to athleticism? Well, by conjecture at first, informed by noticing correlations, then by testing cause explicitly at fundamental levels, and by repeating this back-and-forth until we reach a level of satisfaction specific to our task—which proves this level of satisfaction will vary based on task. In turn, knowing only a man’s SQ tells us nothing in particular about any of these causes.

There is something else we could do with these numbers besides just averaging (however defined), though, a thing that leads to confusion. But to explain that requires some in-depth statistics, so we’ll defer that explanation until later (see the discussion on principal components).

Now we could say “SQ exists”, but it would only be true for a man in the sense that he evinced a certain score on a jumping and throwing test. Nobody has SQ in a physical sense. People have strength. SQ is not strength, but a measure, in a loose and culturally biased but useful way, of strength—and only when strength has been defined circularly with respect to outcomes we identified. About other outcomes not in the mix, we are far less certain. Plus nobody has strength, either, except in a analogical way. What people do have are muscles, fat, lung capacity, will, and so forth.

To say SQ is strength, and to forget strength is only a proxy for cause, is to commit the Deadly Sin of Reification. The danger, given our cultural mania for measurement, is over-certainty, and over-certainty leads to sub-optimal and bad decisions.

Intelligence

So. What is intelligence? That is a hard subject, which is why many retreat to calling intelligence IQ, which is a test score. Let’s recall what strength was. It wasn’t a single thing. It was the culmination, or accumulation, of lots of different causes, which included will, human will. We defined its measure with respect to culturally interesting tasks, such as those found in Olympic sports. Yet these are not the only tasks where the same kinds of causes are in play (a judgement which also includes knowledge of the causes of strength). Instead of jumping and tossing, we could have used, say, battlefield endurance, or number slain during combat, or speed in shinnying up a tree to jostle loose coconuts, or facility in tracking animals across deserts, or many other activities. Maybe even the ability to sit still for long periods.

That last example belongs to Simon De Deo, who uses the analogy of ability to sit still as predictive of bird watching abilities. If sitting still is thought important enough, measures of sitting still becomes via reification bird watching ability, to the exclusion of all other things.

…it turns out that sitting still is not just predictive of bird watching performance, it’s also predictive of a whole host of other life outcomes. People who can’t sit still for five minutes have more problems with addiction, for example. Conversely, someone who can sit still for twenty minutes is often able to avoid addiction, or to break it if he falls victim. Very, very few people who can sit still for three hours die of alcoholism. Same with divorce, automobile accidents, and being good at chess. Bird watching ability is protective. This fits with how important bird watching is in the culture…

Racial differences in the sitting task appear. Extremely sophisticated linear regressions are done to control for SES, age, educational background of parents, etc., and they refuse to go away. People write books about how the lack of black bird-watchers is due to their genetic inability to do well on the sitting test.

…almost every great bird-watcher alive thinks the [sitting] test is absolutely crazy. Bird watching is not about sitting perfectly still for hours, they say! No great bird-watcher wants to brag about their sitting score. A famously egotistical bird-watcher who writes books about how awesome he is at bird watching, how he totally crushed this other bird-watcher, etc etc., is also really proud of the fact that he was, at best, at the bottom of the upper-quartile of sitting still. Birdbloggers clamor to reveal their crappy sitting scores…

…When we actually look at the sitting still performance of the elite bird-watcher population, they’re actually not so great. Yes, these people are good at sitting still, and some are really quite good. But not crazy good at it, even among the ultra elite. If you go by elite scores, in fact, it looks like literally a quarter of the population might meet the sitting still bar for being a great bird-watcher, even though the test sample was admitted to the birding academies partly on sitting scores…

We have this intuition that there are many different kinds of birders. Fine, the psychologists say. Make a test. The educators invent some tests, but in as much as they are predictive of bird-watching, they correlate with sitting score, and in as much as they aren’t, they don’t. Somehow, the other aspects of birding are resistant to isolated measurement in a test you take sitting down for a few hours. Grit doesn’t replicate.

What do people who teach bird-watching know about a person’s capacity to learn bird-watching? the psychologists say. Our best studies now show that we can isolate the ultimate essence of birding, the principal component of all the tests. It is a test conducted in a white room, with a chair of so-and-so-weight. All stimuli are excluded. It is totally silent. Nobody is present in the room. There are no windows.

Scoring an enormous BQ, i.e. the score on the sitting test, is best indicative of only one thing: how good somebody can score on the sitting test. Same for IQ tests: they are best indicative of how good somebody can score on IQ tests. Their predictive usefulness drops off the further the other activity is from formal IQ tests. Sitting still with a sharp Dixon Ticonderoga #2 in your hand and checking boxes to puzzlers is unlike most tasks found anywhere else in life, though.

This is why Nassim Nicholas Taleb was right to point out the relative unimportance of correlation of IQ scores with practical tasks. There is also the common statistical misunderstanding that correlation measured on samples will be as predictive to the same level on population. Correlations exaggerate predictive ability—a lot. And since much IQ research is “confirmed” using wee p-values, it is suspect or wrong at worst, and at best the certainty in the results is many times more than it should be.

Because people do not have a BQ, or an IQ, neither do nations have IQs, for instance. It is foolish to compare averages across nations when those nations do not have homogeneous populations. The lesson is that only homogeneous groups (of any characteristic) can be compared.

Finally, there is De Deo’s b, the principal component of all bird watching tests, which was the sophisticated statistical framework mentioned above. In IQ tests it is called g and is calculated in the same way: a weighted linear combination of individual test scores.

Since b and g are nothing but weighted linear combination of test scores, and the test scores have already been shown to be correlated, more or less strongly, with this or that, it must be that b and g will also be correlated. We have learned nothing new by calculating these letters or their correlations, however. People who do not have BQs do not have bs; people who do not have IQs do not have gs. To say that people have bs or gs is, as above, to confuse causes of scores with the names of the scores. It is reification.

Statistical analyses cannot create something which does not exist. Probability models are silent on cause.

What Is Intelligence?

Enough of that. What exactly is intelligence? To discover that we need to turn ourselves into mental physiologists. We must not be only physicians, but metaphysicians, too.

We cannot define intelligence circularly and say it is ability to score well on certain tests, and then say this person must have high or low intelligence because he scored well or poorly on these tests. It must be clear (you have to perceive and apprehend this point) that this is not a proof the tests measure what they say they are measuring. In order to properly measure intelligence, if it is even possible to measure, we have to have a non-circular definition in hand, which outlines the power or powers of intelligence.

Here are the functions of intelligence: intellect, will, memory, sensory imagination, all of which contain a notions of capacity, endurance, speed. None of these powers can be substituted for the other. Each is a facet of intelligence in the same way the various physiological causes were said to be strength. No individual function is intelligence alone. All operating in concert produce the range of what we mean by intelligence.

The intellect itself has three different powers: (paraphrasing the link) perception, apprehension, cognition. The will also has three powers: motive, appetitive, conative. And the passions two: sensitive and emotive. The powers of sensation and perception, memory and imagination, which are also possessed by lower animals are completely different in nature than those powers of the intellect and will. Only man has intellect and will; and these powers are vastly more than mere computation, which has been proved in various ways (e.g. Searle’s Chinese Room or Chinese Gym experiments).

The senses take in information from varying sources, in time and place, including memory, and serve it up to the common (think of this as “communal”) sense, where it forms the imagination, which in turn is fed to the intellect, which talks to the will, which talks back to the intellect, all of which talk back to the passions. There is no linearity to this process, which is comprised of various feedbacks, interactions, and even time components, given change occurs to each of these elements due to outside causes. Intelligence, like strength, therefore cannot be fixed and unchangeable. Hence it is a double mistake to say “IQ is constant.”

Dogs do not have intellects and wills; they are not rational animals as we are, where rational by definition means having intellect and will. Dogs do have sensation, memory, and imagination. Everybody knows that some dog breeds are more intelligent than others, but we now see (“seeing” in this metaphorical way is a higher intellectual power) that what we mean by intelligence in dogs is not the same of what we mean as intelligence in men.

Sensation and so on are biological functions, and therefore differences in intelligence in dogs can be put down to genetic differences, or rather genes-plus-environment and their interplay of continuing and continuous interactions (in a way that is surely not understood completely). Therefore, if we test separately for each power of intelligence (memory etc.) in dogs, or in their combination (in a way we might not be able to separate) via activities we deem interesting and important (such as guiding sheep, fetching, or even fighting), we might be able to tease out the genetic and environmental contributions to animal intelligence.

This genetics (to give it a short name) testing, to the extent it works with any certainty in animals, works in man, too. But only for the lower powers. It does not work for the higher powers of intellect and will. It does not work because the intellect and will are not material, and therefore are not subject to direct genetic influence. In spite of what you might have heard, medical science has in no way proved the intellect and will are part of the brain.

Here is an analogy of medical testing that makes such claims. Scientists have noticed that when a certain body of water sparkles in the morning, the air temperature increases. An fMRI (presumably the acronym is from the French) measured the strength of the sparkle, and the correlation between the sparkle and temperature was confirmed (p<0.0000001). Therefore, it is judged, correlation becoming causation when ps are wee, there is something in the body of water that is causes the temperature increase.

Right?

The analogy is better than you might first think. The sun causes the heat; the sparkle on cloudless days and its lack on cloudy days is incidental. The water does mitigate the temperature change, but its causal effect pales next to the driving force.

Same thing with medical measurements of the brain. An fMRI sparkling is not proof that the brain is causing the intellect and will to operate. We know by other arguments the causation goes the other way around. This is proved elsewhere (here, here, here).

What all this means it that it is impossible—not just unlikely or difficult, but impossible—to draw a causal link between genes and the higher powers of intelligence. Genes thus have no direct influence on the higher powers. And thus evolution, whatever that might be or however you want to define its mechanisms, has no influence on the most important part of human intelligence. That is so important that I want you to re-read it: evolution and genes have no influence on the most important part of human intelligence.

Genes and their environmental interactions can and do have indirect influence. A man who is mentally retarded has intellect and will, but his biological powers of intelligence are less than those of a non-retarded man, and these biological differences are caused by genes (always understand this language is a dangerous shorthand meant to include environment). The biological functions of intelligence serve, as it were, the higher functions, which in turn inform the biological functions. Damage the biological functions, via genetic accident, injury, or old age, or examine them in their immature form, as in infants, and we see changes in intelligence which are naturally ascribed to biology. Intelligence therefore is not and cannot be constant in a person’s life, at least regarding its biological components. (This would seem to be almost trivially true, but there are a few people who claim “IQ” is fixed, which is now seen to be false.)

What about the non-biological components of intelligence? They are how we touch the infinite, if you will allow me the poetry, which in this case is literal. When we grasp (another metaphor) a universal, we comprehend something infinite (think of how you know there are an infinite number of numbers). Any operation of intuition and induction, the highest forms of intellect, require a mechanism (a cause) of making contact with the infinite (this was first proved, as many things were, by Aristotle). What is this cause?

I don’t know how it works, precisely. I do know it has to be a mighty cause, because the infinite is not small. One such mighty cause is the will of God. Well, you can dismiss this metaphysics about God and the infinite, but you are left trying to explain how our already proved non-computational, non-material intellects and wills work. You can say, like Roger Penrose does, that it has something to do with quantum mechanics because nobody understands cause in QM, thus its mysteriousness makes for a great rug to sweep cause under. You can evoke “emergence“, but that’s to invoke magic since nobody knows how that works, or if it even does, or even if it makes sense. Neither of these is an explanation; they are only hopes of an explanation. My explanation is hope, too, but based on a solid foundation. Which proves you can’t escape faith in discussions of intelligence. Anybody that pretends otherwise is bluffing.

How separate are the intellect and will, which Christians say are the form of the rational soul and which therefore because of its non-material nature survives after the death of the body, from the biological functions? Can a retarded man, who has provably inefficient (let us say) biology access his higher intellectual functions? Yes. Not as well, given the inefficiency, as a non-retarded man. The retardation affects the intellect, giving it less material to work with, so to speak. But that is far from a proof that the retarded man cannot experience revelation or have an insight via intuition not accessible to the non-retarded man. God could grant any man wisdom regardless of his biological apparatus. Too, the retarded are very often extremely happy people—but you don’t find any of them performing functions which we classify (possibly circularly) as intellectually challenging.

I am thus not claiming to know how intelligence works in its entirety. That is the key point. Because we don’t know how the intellect works, in all its components, because we don’t understand its causes, we cannot specify its limits or say how to measure it with precision. What is its lower and upper limit? How far above the animals are we, precisely? Can the intellect of one man be poorer than a second man’s, but the first possessing superior biological function (such as accessing memory and producing phantasms), such that first man is said to be more intelligent? Maybe. How different can intellects and wills be? Nobody knows. In any task requiring intelligence, how much does each power contribute? Probably depends on the task. The intellect and will not being material clearly cannot be damaged as biological functions can. The higher powers can be damaged in ways more horrible, as when falsities are embraced as true and immoralities as good. The intellect and the will can be stained (if you like), by loading it with sin (to coin a word) and falsehoods. As the old saying goes, sin makes you stupid. Thus as sin increases, the higher powers of intelligence decrease. (Perhaps an adequate description of the West.)

Even given all these (and more) unanswered questions, we can measure effects which we say are related to the intellect, and we can give these measures some pseudo-quantification, but they are at best groping efforts (on a scale of -47.2456 to 18/e in units of sqrt(pi/1.2), how much do you agree with that opinion?). We can’t escape the circularity until we know causes—which we may never know for the higher functions.

This is not a claim that certain measures of performance are not useful, for of course they can be, and are, though the uncertainty in them as measures of intelligence is much larger than many think, as I show next. There is nothing to be gained by fooling yourself into believing you have captured all there is, or even the most of what is, of intelligence if you can’t even define what intelligence is. We do not have this same limitation for strength, where cause is understood (to a great degree). To say, then, as everybody does say, that people “have” IQ, or that IQ is “real”, as if a test score is intelligence, is a terrible mistake caused by laziness, ignorance, and hubris.

Testing Tests

What’s the right answer to this pictorial puzzle? (Source.)

Obviously, 5. Yes?

Well, it’s obvious to the residents of Crete, growing up as they do with architectural alternating patterns similar to this one, used as inlays to wainscoting and various other decorations. Those coming from Cambodia might argue and say 4 is a better fit, as it better resembles common direction markers and light switches, only 4 is fancier.

Americans and others who grew up with puzzles like this will wave all these aside and say 1. They’ve seen these kind of patterns before in various testing situations, and this answer best accords with their experience of the test-maker’s mind.

If you’re like most, you’re so used to seeing these kinds of teasers you can’t see that they must have context, which can only be provided by experience. And experience takes place with cultures.

Ken Richardson, in the paper “What IQ Tests Test” would agree. Every person who thinks they have IQ tests pegged must read this work.

For instance, Richardson would agree that the puzzle above is similar to Raven’s Progressive Matrices test, which Richardson says many consider a “pure”, i.e. unbiased, measure of intelligence. This isn’t so:

…[A]nalysis suggests that the cognitive processing demanded by Raven’s items simply reflects knowledge structures most common in one particular culture. Thus, many middle-class cultural tools are based on the manipulation of symbols (e.g. words, numbers) in two-dimensional array on paper. These include record sheets, tables with rows and columns of totals and subtotals, spreadsheets, timetables, and so on, as well as textual material. These nearly all require the reading of symbols from top left to bottom right, additions, subtractions and substitutions of numbers or other symbols across columns and down rows, and the deduction of new information from them. As the analyses of Carpenter et al. (1990) show, these are precisely the kinds of manipulations (or ‘rules’) built in to Raven’s items.

So what does a Raven’s matrix test? The ability to solve puzzles embedded in a particular cultural context, even though the symbols in the test are somewhat out of daily experience. They are seen, however, often in IQ tests. Given the ubiquity of Descartian graphs and numerical examples which surrounds us might not seem like cultural “bias”; to us they are more like natural facts. They aren’t, though. They must be learned.

Some will learn it better than others, however. It is not that intelligence does not matter in these tests. Of course it does matter, even though it is not clear which power or which powers of intelligence are being tested, precisely. A literate Chinese man living five centuries ago, used to writing in a more symbolic language, and in opposite directions to English, might fail the Raven’s matrix test abysmally, even though he has managed to memorize a substantial body of poetry and can exposit on the beauty and importance of calligraphy in a way impossible to understand unless one was saturated in his culture.

Culture matters. Richardson says, “Another tacit assumption of the computational model underlying IQ is that testees operate in a social and affective vacuum. However, humans also have complex values, beliefs, attitudes, motives, self-concepts and feelings, which make them more or less well prepared for specific testing situations and engagement with them.” Not only that, but preparedness matters, which also involves will. Richardson highlights many other such culturally tainted aspects of IQ tests, many quite surprising (don’t be lazy: read the paper).

Now, the question works both ways. Yes, these puzzles are culturally important. But that does not imply that the puzzles are somehow frivolous or of no use in aspects of the culture which are important for independent reasons. Somebody who cannot do these puzzles at all might make a lousy theologian or geneticist, for example. But then those who cannot do them at all might just be disengaged with the culture, or they might be of low intelligence, or both.

Is culture therefore so important that IQ tests are of no use? No, of course not. They are a good measure of ability to answer questions and perform tasks that are like those used in IQ tests. If these tasks are found in the culture, and they sometimes are in ours, then the tests can be predictive of success on those tasks—even if IQ tests are a not a good quantification of intelligence. Such a quantification, we saw, requires defining intelligence in detail, which we haven’t done or can’t do. (Many do work on these details, of course, but if any have acknowledged the essential difference between intellect and will and biological functions, I have never seen them. It is also true some aspects of biology and intelligence are more or less understood, such as how memory might work, but there is no precise model of how all aspects tie into intelligence, such that it can be well quantified.)

Assuming we can well define at least some aspects of intelligence, can we separate culture from intelligence in tests designed to measure this well-defined aspect of intelligence? No. Not if the test is task-related (like puzzle-solving is); not from inside a culture. There is no way to design a test in a culture to eliminate that culture’s influence. What we might be able to do is to have each culture design their own test, and have everybody take the resulting battery of tests. But since those other tests would be in other languages, and translation is by definition to impart a cultural bias, it’s not clear how to compare results. Still, if we can get something like this to approximately work, it would be of interest to compare those who score well in some cultures and not others. That variance would give an idea of cultural influence. Still, since it is all task-related, we would never be able to remove the suspicion that bias has crept in.

There might be hope in some base biological measurements which would remove all cultural bias, but things like fMRI right now are not far removed from tea-leaf reading (with worse statistics). These would only capture the lower and not higher, non-material functions, and only capture them crudely. Think of Leonardo da Vinci being handed four cheap tomato-sauced covered crayons at a “family-style” restaurant and asked to reproduce the Mona Lisa. We’re nowhere near close to unambiguous exact causal biological definitions of lower intelligence—which is key.

Race To The Finish

Let’s get to the elephant. It is true that blacks in the States score on IQ tests on average lower than (east) Asians. Of course, blacks do not “have” an IQ; neither do Asians. Nobody who hasn’t taken an IQ test has an IQ. IQ is a score on a test: it is not intelligence, or it is only dimly a view of some aspects of intelligence. Everybody has intelligence, which we saw was not simple. So, given these reminders about the danger of the Deadly Sin of Reification, are blacks on average less intelligent than Asians?

We have a hard question here, because we have already admitted we don’t know how to disentangle cultural influence from whatever aspect of intelligence IQ tests are testing—and we’re not sure what aspects of intelligence precisely IQ tests are testing. Whatever conclusion we come to, we’re going to be left with some healthy uncertainty. That is, we ought to be.

Blacks and Asians in the States do share a portion of culture. This sharing is not complete. If the sharing were complete, and I mean 100%, then we could say that, yes, blacks on average are less intelligent than Asians, given the test scores we’ve seen, but only on those aspects IQ tests measure. How much different? Well, who knows. We can’t go by the IQ scores alone, as we now know. Intelligence involves will, and wills greatly differ; it involves biology, and that differs a lot; it involves intellection, and that maybe differs (recall we don’t know how to tease out the biology from the higher powers). Since we don’t know precisely which aspects of intelligence IQ tests are testing, we have to be vague in our judgement. Simply lapsing to the hard quantification of test scores inevitably leads to claiming the unquantifiable was quantified, hence causing massive over-certainty, hence allowing entreé to the Reification Demon.

IQ tests imperfectly measures intelligence because these tests are in part cultural creations and because they only measure a small part of intelligence in that cultural context. Thus differences in IQ scores can also be predictive of cultural differences. Perhaps not to the same extent as intelligence, but surely to some level.

As far as blacks and Asians, or any races, go, even if the observed differences in distributions of test scores are entirely cultural and not one whit biological or spiritual, the differences are still the differences. They don’t disappear because the different cultures caused the differences in scores (material cause, here). The hope, then, is that differences can be eliminated if the cultures can be made the same. As the kids say, good luck with that.

It is also true that some aspects of intelligence are biological, and as such can be inherited partially. Statistical correlations of this gene expression with that test score and the like exaggerate inheritances, because correlation exaggerates. All classical statistics analyses exaggerate and generate over-certainty. Still, if people of like intelligences mate, then it must be the case that their offspring would tend to be of the same intelligence class—to the extent intelligence is biological and heritable. The real correlations are lower than thought, but they aren’t zero, either. Which is to say the causes on intelligence inheritance are not as strong as thought. But they’re not absent, either.

Races (self-identified will still do) tend to inter-marry. If there are biological racial differences in those parts which cause the lowers powers of intelligence, then attempts to erase the differences are doomed—besides forcing intermarriages. All evidence points in the direction of real differences, but, as said, this evidence is not as robust as thought. It is not zero, either. Forbidding speech on the matter can only cause many to assume the differences are larger than they really are.

It is also true, and has been widely noticed, that those who most would say are of high intelligence, however inaccurately defined, tend not to breed, they being prone to chasing other delights beside families. A fellow calling himself Spandrell calls these distractions “IQ shredders“, a reification if there ever was one, though his point is taken. These delights must lower intelligence on average in any future population, even if the lowering is smaller previously presumed.

This will happen unless those of lower intelligence who are breeding have children who have improved functions in those areas responsible for the lower powers of intelligence. It can happen. This kind of rugged mating happens in other animals. Meaning surprises might happen. Smart money is still on the side of worsening, though.