Gould, Mismeasure on Spearman

We began, as per usual, with questions:

What does SJG mean by g-loading and what is the difference between g and s?

What is and is not included in g for Spearman? 

SJG discusses “automatic reification” (p. 299).  Does this account entail that there are some valid or warranted forms of reification?  In other words, is SJG opposed to reification as such, or is he arguing against certain styles/types of reification?

Why is Spearman using political metaphors to contrast differing theories of intelligence (p. 287)?  He refers to “oligarchic,” “monarchic,” and “anarchic” components of intelligence theorizing/testing.

How do we fill in the missing assumptions in SJG’s critique of “reductionism” in Spearman (p. 292)?  What is at stake in a reductionist critique?

How do we understand SJG’s distinction between theory-guided science and “rough-and-ready” empiricist science (p. 293)?

 

Discussion ensued:

What does SJG mean by “automatic reification” (p. 299)?  Is it “automatic” insofar as it is a form of “reification” (or positing of a causal influence for a reified g) that reads off a causal posit from a mathematical factor?  SJG’s “complaint” is “with the practice of assuming that the mere existence of a factor, in itself, provides a license for causal speculation” (p. 298)?  For SJG, “no set of factors has any claim to exclusive concordance with the real world” (p. 299)?

What is SJG’s criterion for when reification is warranted rather than not?  His earlier critique of reification (282-285) refers to reifications “aris[ing] from the mathematics alone” but rather “only from additional knowledge of the physical nature of the measures themselves” (p. 280).

Does this make SJG a staunch realist?

So is SJG’s argument that there is no possible measure of a single thing that is “intelligence”?  Or is the argument that there is as-yet no viable measure of a single “intelligence”?

We need to separate SJG’s critique of “measurement” of intelligence versus a potential critique of the idea that there is a physical seat of intelligence.

Okay, here is what we think:

1.        (1) SJG holds that factor analysis of intelligence tests fails to license inference to the reification of intelligence (for the reasons stated above on p. 280, p. 292, etc.).

2.        (2) Given (1), SJG holds that there is no justification for an account of the causal processes productive of intelligence.

However here is an issue:

So in order to motivate (2), SJG ends up needing to adopt a kind of realist philosophy of science (based in physics).  But this quite possibly ends up being question-begging by assuming a realist standard.  So the standard for evidence/warrant becomes “reifiable” (in a sense that does not reduce/abstract real entities).

Why not adopt a more functionalist account?  Why not ignore the whole issue of reification and just don’t worry about whether or not that which is being measured (intelligence) is reifiable?

 

Gould Mismeasure pages 204-222

 Initial questions:

1. Can we speak about human pathology without reifying and thingifying humanity (e.g. without constructing an ideal conception of persons upon which pathologies are a deviation).

2. Was Lippmann’s debate with Terman formative for the genesis of what would be called neoliberalism?

3. How do we distinguish between within-group and between-group differences?

4. What do we do once Gould has destroyed the common ‘knowledge’ about IQ tests?

5. What kind of history is Gould writing? Is it genealogy? What kinds of objects is Gould surveying? Does he help us problematize the selection, use, and mobilization of data? 

6. What is and is not being tested for by Terman? 

Discussion:

What is the origin of a true pathology, and how is that differentiated from an arbitrary ideology which determines norms? How can we distinguish between the two? This question arises in response to page 218, in which Gould speaks of people with Down’s Syndrome as possessing a “true pathology.” But does this imply an understanding of a human norm?

Not necessarily, because the structure of his argument tells us that the importance of “Down Syndrome” in this scenario is not a supposed ‘lack of intelligence’ possessed by those with three chromosomes, but rather that, if they score low on an IQ test (and the belief is that this is related to their genetic trait of possessing an extra chromosome), we are not warranted to say that all people who score low on IQ tests do so because they are genetic. 

This paragraph can probably be salvaged, using less controversial examples of pathologies, like cancer. If a heavy smoker develops cancer, we cannot presume that all people with cancer have it on account of their behavior. This is the form of the error made by Terman. 

Regarding this obsession with IQ, the Foucaldean concern with knowing oneself, can we extrapolate for a second to the “ancestry” industry? It capitalizes upon our desire to know ourselves, and to understand our potential pathologies. On page 207, Gould examines how much money is to be made regarding these standardized tests. 

On 209 Gould writes about a mother who is excited because her boy is learning how to read, though Terman is skeptical of this excitement, as the child received a low IQ test, meaning that no matter how much energy he puts into reading, he will never be able to achieve what a reasonable parent might expect. The IQ test is meant to realize a disinterested, exacting measurement of human intelligence which is not subject to (say, motherly) intuition. 

Back to the question of the difference between within-group and between-group hereditability. Gould’s point is that he doesn’t know whether there exists any within-group heritability of traits (say, within a family), but even if you do find it, there would be no evidence to extrapolate that finding into a between-group situation, like a neighborhood. 

Let’s all remember, on a different note, that Lippmann has an early and late period. While his late period may have been neoliberal, his early period seems more radical (using the state to impose good). 

Let’s end off on the question of what we do once the IQ test is destroyed as faulty. How do we justify or standardize universal education without acknowledging intelligence as that which is fostered, and without a way of measuring the success of that fostering. 

Gould, Mismeasure (intro and concl.)

 

Stephen Jay Gould, The Mismeasure of Man

 

Reading from the “Intro” and “Conclusion” (Ch. 7) to SJG’s The Mismeasure of Man (1996 edition).

 

The group began, as per usual, with questions (grouped loosely as follows):

In discussing the two fallacies of intelligence testing (p. 56) SJG says his book is “about the abstraction of intelligence as a single entity”.  What in/about abstraction does SJG object to?  Or does he object to abstraction as such?

In discussing intelligence, SJG identifies “two components” – the IQ scale and factor analysis.  What do we take these to be?

 

Are there newer iterations of racism that SJG’s discussion doesn’t speak to?  What would his analysis say to these kinds of iterations?

If there are newer such iterations, what would the basis be for these newer iterations of racism?  And could Gould’s analyses be used to interrogate these iterations?

In discussing cultural evolution and biological evolution, SJG discusses cultural forms that speed up forms of evolution (ca. p. 355).  Are these relevant to these newer iterations?

 

In discussing the insights of biology for human behavior (which overall SJG finds well overstated), there is an argument that biology offers “fruitful analogies” for human behavior (p. 357), but the force (and sense) of these analogies is not quite clear.

 

In SJG’s arg. against biological determinism, there is a move away from the idea that behaviors are coded (or determined).  Where does this lead?  Does this motivate the importance of education?  If so, what kind of education does it prefer/motivate/justify?

One good locus for discussion for this would be discussion of “generating rules” and their role in determining behavior (ca. p. 360).

 

 

We then moved to discussion:

How would SJG draw the line between permissible and impermissible abstraction?

Are there two forms of abstraction involved here?  One focused on quantification of “intelligence” and the other focused on quantification across all persons (“each person”)?  So one simplifies multiple complex qualities for a person.  The other simplifies the complexity across persons.  Is the one of these “reification” and the other “ranking”?  Are these both forms of “abstraction” in Gould’s sense?  What makes these abstractions pernicious?  The argument on p. 56 seems to rely on the idea that abstractions are pernicious when they reify and rank.  Could one, then, for instance measure intelligence without reifying or ranking?

                What exactly are the problems with reifying and ranking?

                The problems with ranking are clear.

                Is the problem with reifying that it is reductive?  “This wondrously complex and multifaceted set of human capabilities” (p. 56).  SJG is concerned about being reductive about this. What is the philosophical posit here?  Is there a posit about that there is a complexity?

 

What do we understand by IQ scale and factor analysis?

                SJG emphasizes the “hereditarian” qualify of IQ scaling in the U.S..

                Factor analysis is what achieves the “reification” of intelligence as a single quality by way of mathematical analyses.

                Is SJG’s argument that IQ scales are what “ranks” and factor analysis is what reifies?  Or is it that SJG holds that both IQ scales and factor analysis each rank and reify?

 

Discussion of generating rules and their products.

                A “generating rule” is something like what codes “biology” to set limits on “behavior”.

                SJG distinguishes between:

                                “products of generating rules” (e.g., behaviors or phenotypes)

                                “generating rules” such as “genetic deep structures”

 

Does SJG’s argument have traction against newer forms of white supremacy or racism that would fully accept the cultural contingencies of the development of intelligence testing?

The idea of forms of racism that depend not on biological determinism but on supremacist historiography.

This is related to why SJG’s implicit (explicit?) scientific realism is not going to actually take on the political/moral issues.