Death Penalty and the Flynn Effect

Factual Questions
1

This is a question regarding the implications of the Flynn Effect on the death penalty ------- the Flynn Effect is the result we see when average IQ scores get higher each ‘generation.’ It appears established that we see higher IQ scores in the range of about 10 – 15 points for each subsequent generation – if we use the same IQ test every year. Of course, if this rapid increase goes back more than a few generations our ‘Founding Fathers’ were all ‘legally inculpable.’ Tell George III!!! Anyway ---- the idea behind the Flynn Effect is that in our society - general information is mostly for the taking, libraries are public, the overall culture / environment is more complex / enriched, nutrition is better, people are taller and living longer, and so -------just might have better physical brains and better environments in which to grow them.

Yet — as I understand - it is unconstitutional to employ the death penalty for any person who scores below 70 on an IQ test since this is one of the measures of legal culpability. The question is this ---- how does the legal system measure IQ for capital crimes committed years in the past? I ask because the legal culpability occurred at the time of the crime – yet the measured culpability — at least in the cases I’m asking about – takes places years later. How is this problem handled by the legal system?

2

My understanding of the IQ test, which is not extensive, is that if enough people write it, their scores will for a normal distribution. This can be standardized to form a median of 100 and a standard deviation of whatever.

I don’t know about the Flynn Effect. I doubt people are getting smarter by 15 IQ points each generation, and ieven f they were this would not affect the standardization – are you saying the effect is one that only positively skews the bell-curve, or just the standard deviation is increasing?

Guess I’m asking you for a cite for your claim. I don’t see why you couldn’t just get the person to do an IQ test now and see if that was over 70 – would an adult with a valid 60 IQ change much over their lifetime?

3

Here is a cite discussing your claim:

Here is the original article (pdf!)

http://www.apa.org/releases/flynneffect_article.pdf

No one seems to have discussed the legal consequences of the effect before 2003, so I’m sure it simply wasn’t considered. I think the biggest danger is defining MR based on IQ tests in the first place – surely funding should be based on other criteria… if you score between 71-80 on an IQ test, what would you score if you wrote four IQ tests? If you scored 65, what would you score if you wrote three other ones?

4

I guess what I still don’t understand is why the norming cycle has to be each 15-20 years, when it could be renormed any time a significant number of people write the exam.

If the standard deviation from the mean is changing, maybe MR could be defined, say, as four standard deviations below the mean in addition to the other psychological factors. It seems clear that if only 100 of 210 (or whatever) people with low IQs were thought to be MR than the IQ criterion isn’t very useful anyway.

5

Regarding the application of this IQ business to the application of the death penalty, the focus is on current cognitive ability because constitutional issues are raised when the person to be executed is too dimwitted to comprehend the nature of the penalty or the reasons for its application. Of course, the argument can prove too much, as you point out; if they were that impaired, how can they be legally culpable for their conduct at all? Legitimate question. Suffice it to say that our society, rightly in my view, is skittish about executing people it views as pathetic–the very old, the sick, the mentally disabled, whatever–regardless of whether those conditions existed at the time the crimes were committed.

In Texas, on the other hand, the only thing that excuses you from execution is if you’re already dead.

6

The simple answer is that no one has thought about the relevance of the Flynn effect to the death penalty before. If you were to ask a judge who was to decide whether to give the death penalty to a convicted murderer who otherwise qualified for it about the relevance of the Flynn effect to his decision, you would just get a funny look and be told, “Hey, all I have to do is ask a psychologist to test the person’s I.Q. and find out how high it is. It’s not my business to argue the usefulness or correctness of I.Q. testing.”

The chances that the average I.Q. has shifted significantly between the time of the murder and the time of the sentencing are so tiny that it can be pretty much ignored. In most cases, the sentencing happens about one or two years after the murder. The Flynn effect says that the average I.Q. will have only risen by perhaps half a point in that time. That’s a small enough amount that it’s not really even measurable.

Incidentally, an I.Q. of 70 is more like two standard deviations below the average, not four standard deviations.

7

The OP is conflating two issues: a rise in I.Q. scores, and an actual increase in average intelligence.

James R. Flynn, for whom the “Flynn Effect” is named, is quite outspoken in saying he does not believe that the population is becoming markedly smarter on average, only that the average scores rise over time. These can be two entirely separate issues, as one deals with the thing measured, and the other deals with the thing doing the measuring.

An I.Q. test is supposed to measure intelligence. Even if we grant that it does this in some fair and meaningful way, and that is subject to enormous argument, it should be acknowledged that it can only do so in general, indirect terms. Intelligence is an abstraction. It is not a concrete thing that can be measured absolutely, in the same way, say, that the length of a piece of metal or the weight of a piece of produce can be measured.

But, to simplify matters, suppose for a moment that I.Q. tests really were as precise as a weight scale or a calipers.
Anyone who has worked for long in a machine shop or similar facility knows that measuring devices have to be recalibrated from time to time to assure their accuracy.

Imagine you have a dial calipers and a block of steel which you know for a fact is an exact one-inch cube. The first day that you use your calipers you measure the block and it measures out at exactly one inch on all sides. The second day that you measure the block it measures at exactly one inch on all sides. And so on, for days on end.

Gradually, though, something seems to be gradually going out of whack. After months of using the calipers without ever testing it or servicing it, it now tells you that the cube measures 1.001 inch on each side. And then, months later, the measurement has crept up, so that now the cube measures 1.002 inches on each side.

If a gradual increase in the average I.Q. test score using the same test over a period of years means that people are, on average, getting smarter, then, by the same logic, a steel cube which gradually measures larger and larger using the same untested calipers must be growing.

Here’s another example: suppose you have a produce scale, the sort of device one sees in grocery stores all the time, where there is a basket dangling underneath in which people can stack carrots, potatoes, etc. Suppose that people throw their produce into the basket roughly all the time, so that the basket sags and bounces back, and the needle bounces wildly before settling down. Gradually things start weighing more and more on average when placed in the scale. In fact, even the surrounding air weighs more: at the start the scale showed a weight of zero whenever the basket was empty, but now it shows a half ounce when at rest.

Using the logic of the OP, we can only conclude that either the dangling basket, or the air that sets on it, is getting heavier, along with everything that is measured on the scale.

The inference here is that I.Q. tests, calipers and scales don’t simply attempt to measure things. Rather, they define them infallibly. The I.Q. test, the calipers and the scale each take their turn at being God.

If there is no distinction between the results of an attempt at measurement and the underlying reality that one is attempting to measure, then the universe is a whole lot cooler and more accomodating that I, for one, ever realized. Like millions of people, I’ve thought about losing weight in the new year. Instead of dieting or exercise, all I have to do is set the needle back on my bathroom scale so that it shows less than zero before I step on it.

Half of the questions on an I.Q. test measure general knowledge. Before an I.Q. test is put into general use, it is administered to a grat many test subjects. In this way, the testers are able to determine how many correct answers the ideal average person at the center of a bell curve is “supposed” to know. Neither the test nor the people who administer it, though, have any way of controlling whether the answers to some of those questions will become more generally well-known over time.

Suppose they do.

Does this mean that the population is becoming smarter on average? Only if knowing the answers to those particular questions, and only those questions, in and of themselves are an absolute definition of intelligence. Then, by definition, the opulation at large is getting smarter, and simply because the answers to those particular questions are gradually becoming common knowledge. If so, then a person who gets the answer key to the I.Q. test and memorizes it is, by definition, incredibly smart, even if he doesn’t know another thing in the world.

Alternatively, one could argue that if test scores are rising on average, it is only because the I.Q. test being used is losing its accuracy, a bit like a dial calipers or a weight scale with a bent needle.

8

But they’re working on it. :slight_smile: