IT CERTAINLY LOOKS LIKE A PIPE: FOUCAULT AND
THE CALIFORNIA ACHIEVEMENT TEST

Murray Ross
University of British Columbia

In I’m not Lying, This is Not a Pipe James Palermo ingeniously reworks a familiar theme, namely that culturally biased standardized tests threaten to limit the educational opportunities of minority children. Palermo argues that Foucault’s meditation on Magritte’s This is Not a Pipe reveals “the ambiguous, contradictory and non-representational relationship between words and things.” Once this relationship is understood teachers can begin to develop the critical aesthetic consciousness necessary to unmask the illegitimate practices of normalization, subjectification and exclusion associated with the CAT and tests like it.

The most intriguing and original elements of Palermo’s argument center on the contradictory “message” of This is Not a Pipe. Given the conventions of representational art, images are meant to represent, while language is meant to fix the reference of the image. That Magritte breaks with these conventions is seen by Palermo to be of great significance. He likens the painting’s contradictory message to a lie “which repeat[s] within us the agonizing experience felt when we think the friend conversing with us is lying.” I do not think it is unfair to suggest Palermo is exaggerating here. There are many speech acts besides lies that convey mixed messages. Puns, for example, or jokes are frequently based on an inversion or violation of the conventions and background assumptions that make our various language games intelligible to us. The people I asked take the painting to be a quirky and whimsical work; most felt the contradiction between inscription and image to be too transparent to qualify as a lie, or even an attempt to lie. I think Palermo is too quick to speak of lies, both with respect to the painting and the CAT. More of this later.

The heart of Palermo’s discussion of This is Not a Pipe centers on the distinction between resemblance and similitude. Unfortunately, Palermo is not very clear on just what the distinction is between these two, and the excerpts from Foucault’s discussion are too brief to clarify sufficiently what special sense is being attached to these terms. I am not sure, for example, what sort of hierarchy is being alluded to, or how representation rules over resemblance. I suspect that the meaning Foucault gives to resemblance presupposes a particular fixed ontology where words somehow correspond to things in the world, while similitude refers to relations between words or statements, and allows for a variety of interpretations. According to a fixed correspondence relation between words and things, there are right and wrong ways of speaking, in particular, ways of speaking that fail to match up with reality in the proper way. In a relation of similitude, statements are true not by virtue of some correspondence with things in the world, but relative to a language. I might well be wrong about this for I found the discussion obscure. The upshot appears to be that the CAT assumes a fixed correspondence between words and things which inevitably produces among test-takers an arbitrarily designated category of losers, individuals who are judged sub-normal or abnormal in virtue of the different language they use. Presumably it is the arbitrary nature of this category that makes the designation a lie. Palermo draws several conclusions from Foucault’s discussion:

1. Plastic imagery and discourse can be shown as incommensurable sign systems.
2. Plastic imagery need not represent; it can instead be an expression of similitude.

These are interesting and provocative claims, but their meaning too is less than perfectly clear. The few sketchy comments on similitude and resemblance do little to convince me that the concept of incommensurability is appropriate in this context. For one thing, it is not clear whether Palermo is saying imagery and language are incommensurable sign systems or merely that they can be. Since incommensurability is usually taken to be an all or nothing affair I find these conclusions puzzling. It is easy to accept the second claim that artistic images need not have a narrowly fixed reference, but we do not need Foucault to tell us this. If the first claim amounts to anything more than “the meaning of paintings is harder to pin down than the meaning of words,” I would have liked a bit more argument. If the plastic images and language of the CAT are literally incommensurable sign systems, then it must be that everyone will do poorly on the CAT, including the designers of it, and that is clearly not what Palermo is saying. If Palermo’s remarks concerning incommensurability amount to nothing more than the claim that African-American children will have difficulty interpreting the test, then I do not see how Foucault’s distinction between similitude and resemblance illuminates the problem of test bias more vividly than the standard sociological critique.

A word or two about disciplinary power seems in order. In Discipline and Punish Foucault remarks that

the workshop, the school, the army were subject to a whole micropenality of time (latenesses, absences, interruptions of tasks), of activity (inattention, negligence, lack of zeal), of behavior (impoliteness, disobedience), of speech (idle chatter, insolence).¹

The issue, of course, is whether such standards of conduct, activity, and speech are essentially implicated in practices of domination and repression. Must norms of academic competence (or readiness) inevitably be part of a normalizing practice which by its nature mistreats schoolchildren or in some way brings them harm? It is a one-sided analysis, Foucault says, to describe power only in its negative functions of exclusion, repression, and censorship, for power also produces abilities. Disciplinary power increases the force or productive capacity of the body at the same time it diminishes the individual’s power to resist.² The strategies of hierarchical observation, normalizing judgment and examination yielded in schools a remarkable increase in achievement, most often by restricting the student’s power to resist. Is this evil? Foucault himself admits the answer will vary case by case.³ Despite Foucault’s disavowals in interviews, Foucault’s genealogies are often seen as protests against the disciplinary effects of hierarchical judgment or examination, and nothing more. Freedom is a good in itself against which the discipline of schools, asylums, and leprosariums is an offense. Foucault makes plain, however, that he does not believe all educational norms must be, in some way, arbitrary, or based on false premises, or compromised by the will to power. Nonetheless, it is apparent that some educationists ignore in Foucault’s writing this tension between the potentially oppressive character of normalizing discourse and the unproblematic nature of our trying to make accurate judgments about children’s intellectual growth. Foucault once said in an interview:

Let us also take something that has been the object of criticism, often justified: the pedagogical institution. I don’t see where evil is in the practice of someone who, in a given game of truth, knowing more than another, tells him what he must do, teaches him, transmits knowledge to him, communicates skills to him. The problem is rather to know how you are to avoid in these practices — where power cannot not play and where it is not evil in itself — the effects of domination which will make a child subject to the arbitrary and useless authority of a teacher.⁴

Schools are places, says Foucault, “where power cannot not play and where it is not evil in itself.” Our task is to figure out how to avoid subjecting children to arbitrary and useless authorities, not all authorities.

Is the CAT implicated in a lie, part of a normalizing discourse that fabricates, rather than merely identifies, an inability to switch codes, as Palermo insists? Palermo accepts that schools should teach Standard English to African American children because these children need to be able to translate from the dominant code to their own. One need not think Standard English is superior to recognize the disadvantage that falls to those children unfamiliar with it. Thus, an inability to switch codes is a real inability, not an imagined one. I am, therefore, at a loss to understand how Palermo can say “the language of discursive practices creates the reality it describes.”

It has been observed that there is an American Foucault and a French Foucault.⁵ Rorty says the American version “can be read, with only a little strain, as an up-to-date version of John Dewey…with most of the Nietzscheanism drained away.”⁶ The American Foucault is equally concerned with injustice, but has not hobbled his critique with the same ambiguities and paradoxes that beset the Nietzschean Foucault. The American Foucault is more coherent, though less dramatic. I think Palermo’s paper is written in the spirit of the French Foucault, and is as a result more dramatic than coherent. The story he tells has been told before, but less dramatically, in the sociological critique. It is an important story to tell again, and some of the Nietzscheanism, at least, is salutary. Philosophers such as Rorty, Taylor and Walzer prefer the American Foucault, but admit the French Foucault is correct enough to be disturbing.

---

1. Michel Foucault, Discipline and Punish: The Birth of the Prison (New York: Vintage Books, 1979), 178.

2. Foucault, Discipline and Punish.

3. Michel Foucault, “The Ethic of Care for the Self as a Practice of Freedom: An Interview with Michel Foucault on January 20, 1984,” in The Final Foucault, ed. James Bernauer and David Rasmussen (Cambridge, Mass.: MIT Press), 1988.

4. Ibid., 18.

5. Vincent Descombes, review of Foucault: A Critical Reader, ed. David Hoy London Review of Books 5 (March 1987): 3.

6. Richard Rorty, “Moral Identity and Private Autonomy,” in Michel Foucault: Philosopher, trans. and ed. Timothy J. Armstrong (New York: Routledge, 1992), 328.