The Disunity of Science: Boundaries, Contexts and Power

The Disunity of Science: Boundaries, Contexts and Power

by Peter Galison

View All Available Formats & Editions

Is science unified or disunified? Over the last century, the question has raised the interest (and hackles) of scientists, philosophers, historians, and sociologists of science, for at stake is how science and society fit together. Recent years have seen a turn largely against the rhetoric of unity, ranging from the please of condensed matter physicists for

…  See more details below


Is science unified or disunified? Over the last century, the question has raised the interest (and hackles) of scientists, philosophers, historians, and sociologists of science, for at stake is how science and society fit together. Recent years have seen a turn largely against the rhetoric of unity, ranging from the please of condensed matter physicists for disciplinary autonomy all the way to discussions in the humanities and social sciences that involve local history, feminism, multiculturalism, postmodernism, scientific relativism and realism, and social constructivism. Many of these varied aspects of the debate over the disunity of science are reflected in this volume, which brings together a number of scholars studying science who otherwise have had little to say to each other: feminist theorists, philosophers of science, sociologists of science.

How does the context of discover shape knowledge? What are the philosophical consequences of a disunified science? Does, for example, an antirealism, a realism, or an arealism become defensible within a picture of local scientific knowledge? What politics lies behind and follows from a picture of the world of science more like a quilt than a pyramid? Who gains and loses if representation of science has standards that vary from place to place, field to field, and practitioner to practitioner.

Read More

Editorial Reviews

From the Publisher
“This is a very important work, with contributions by many of the most prominent scholars in science studies....It actually delivers on its promise to renew discussion and develop fresh ideas about the allegation that the sciences are no longer (or never were) unified by a single theoretical view of nature or a methodological foundation.” —Michael Lynch, Brunel University
Comprises 16 papers addressing the varied aspects of the debate concerning the unity or disunity of science. Contributors, including scholars of feminist theory and philosophers, historians, and sociologists of science, address topics including how the context of discovery shapes knowledge, whether antirealism, realism, or arealism becomes defensible within a picture of local scientific knowledge, and what politics lie behind and follow from particular views of the world of science. Annotation c. Book News, Inc., Portland, OR (

Product Details

Stanford University Press
Publication date:
Writing Science Series
Edition description:
Product dimensions:
(w) x (h) x 1.50(d)

Read an Excerpt


Boundaries, Contexts, and Power

By Peter Galison, David J. Stump

Stanford University Press

Copyright © 1996 Board of Trustees of the Leland Stanford Junior University
All rights reserved.
ISBN: 978-0-8047-2562-0



The Disunities of the Sciences

* * *

What's in a name? Often, an ideology. Take the present names, in English, of the fields of specialization represented in the present collection of essays. We have the philosophy of science, the history of science, social studies of science, the history and philosophy of science—as if there were one thing, science, for there to be a philosophy or a history of social studies "of." Philosophers used to speak of the sciences, not science. In this paper I am concerned with different kinds of unity and disunity, not with different kinds of science, but it is well to begin by thinking about how an ideology, the unity of science, has affected even the names of what many of us sometimes do.


Specialist philosophy or history of the sciences descends from two landmark bodies of work written in the 1830's. In one case the very titles make the plurality plain: William Whewell's The History of the Inductive Sciences (1838; three volumes) and The Philosophy of the Inductive Sciences, Founded upon Their History (1840; two volumes). Diversity is also firmly asserted in the other case, Auguste Comte's Cours de philosophie positive (1830–43; originally six volumes). His vast classificatory system of the departments of knowledge was an affirmation of difference. Comte fought long and in vain to establish the first professorial chair in the field—of histoire des sciences. The founding fathers of our fields of specialization differ greatly, the one adumbrating the rationalist vein still apparent, and the other prescribing the more common empiricist analysis. The role model for the rationalist Whewell was Francis Bacon, and for the empiricist Comte it was Laplace; so much for philosophical, as opposed to national, -isms. The two men agreed in this: both thought of themselves as philosopher-historians of the sciences, not of science.

Whewell, ever thorough, addressed the issue of science/sciences in the opening two paragraphs of his Philosophy. He did think that there might be something worthy of the name philosophy of science, but it was not something that he or anyone else could propound: "The Philosophy of Science, if the phrase were to be understood in the comprehensive sense which most naturally offers itself to our thoughts, would imply nothing less than a complete insight into the essence and conditions of all real knowledge, and an exposition of the best methods for the discovery of new truths." As optimistic, encyclopedic, and influential as he was, he did not reject such a philosophy out of hand, but he did consider it impracticable. We should concern ourselves not with some "real knowledge" of which human beings could dream, but, as he put it, with the doctrines of solid and acknowledged certainty that do exist among us—the several sciences. He did think, what many now reject, that the very idea of a philosophy of science (in his ample understanding of the words) makes sense; but any who aspire to it "may best hope to make some progress towards the Philosophy of Science by employing [them]selves upon THE Philosophy of the Science."

Comte's talk of the sciences, in the plural, was also based not on abstract principles but on what we are able to do. He spoke to this in the first lesson of the forty that constituted his Cours: "One cannot reduce all the sciences to a unity." Comte is not to be understood maliciously, as somehow self-refuting, saying something about all science. He wanted only to avoid being misunderstood. Since he wished to give a systematic presentation of the departments of knowledge, he feared that his course would be seen as one more of those attempts at universal explanation "that are daily hatched by those to whom scientific methods and results are entirely foreign." At that juncture he wanted to dismiss one model of unity, namely derivation of all laws from one fundamental law of nature—an "eminently chimerical" project: "I believe that the powers of the human mind are too weak, that the universe is far too complicated for such a scientific perfection to be within our powers, and I think, moreover, that one usually forms a very exaggerated idea of the benefits that would necessarily accrue, were it to be possible." Despite our admirably clear forefathers, we cannot undo the terminology that has since been adopted. "Philosophy of science" is agreeably shorter than "philosophy of the sciences." Our journals are called Philosophy of Science and the like. And why not? The stage was set long before, with, for example, the British Association for the Advancement of Science—"science" seen as a special-interest group in 1833. The American Association publishes Science. But because of my interest in diversity, I shall return to Whewell, Comte, and our roots, and write about the sciences.


Otto Neurath said that "the unity of science movement ... includes scientists and persons interested in science who are conscious of the importance of a universal scientific attitude." The movement was closely associated with logical positivism, but the unity of science had been an important slogan, for diverse reasons, for Helmholtz, Mach, Karl Pearson, and many others. It breaks into several parts: unity has a fairly clear meaning; unity is a good thing; the sciences are a very good thing; and the sciences are or should form a unity. The first is a little-discussed point of logic or language. The second and third are judgments of value. The fourth is hortatory, an injunction about the status or aims of the sciences: the sciences do or should form a unity.

The fourth item is usually the point of contention. I shall be concerned with its presuppositions rather than its content. The very title of this volume expresses the vogue for doubting unity of science doctrines, but the essays in it tend to emphasize the "science" side of the unity of science; to balance this I shall be concerned with the "unity" side. There are a lot of different possible types of unity. I shall seldom argue that the sciences are or fail to be unified in this or that respect. I shall mention some pros and some cons, not so much to take sides as to emphasize that there are disputes about different types of unity or disunity. Hence I do to some extent side with the disunifiers, because the hidden strengths of the unity of science movement lie in the implicit assumption that there is either a single unified way in which the sciences are or should be unified, or else that there is a simple hierarchy of increasingly strong theses about the unity of science.

Unity Is Not a Predicate

I shall not attempt a logical analysis of the concept of unity or a linguistic analysis of uses of the word "one"—although both would be in order. I shall not try to describe the complex relationships between those different words, "one," "unity," and their cognates. I do take for granted a familiar lesson from the philosophy of Kant and of Frege. Kant taught that existence is not a predicate. Frege added that being one or more in number is very much like existence—not a predicate, or at any rate not a predicate of things. Existence and number are, in their primary usage, concepts that apply to concepts. In Frege's jargon, to say that God exists is to say that something falls under the concept "God"; to say that God is one is to say that exactly one entity falls under the concept "God" (or, to eliminate the "one" in favor of the nonrelation of identity, if x and y fall under the concept, x = y). Existence and unity, both said by some medieval schoolmen to be perfections, are not (first-order) perfections at all, because neither is a predicate. Unity is thus like existence; as Kant said, I do not add something to the golden dollars of the merchant in saying, "and then they exist." I do not add to the properties of an apple, after saying that it is crisp and tasty, by saying that it is "one." Hence, insofar as unity connotes singleness, it cannot be a virtue or perfection. It could be a good thing that there be only one editor of this volume, but that would not be a property of the editor.

Singleness and Harmony

Logic-chopping makes us forget the emotive power of the ideas of unity and existence that make people think of them as perfections alongside omniscience and omnipotence. The unity of science was a rallying call in part because unity, in a certain framework, was a good thing. Unity has, in our tradition, been a virtue associated with many powerful ideas: the God of monotheists, the nation, the state, a people, the self. Men and women have been dying for (and against) Unions pretty much since the invention of nationalisms. Kant's transcendental unity of apperception was a necessary condition for human knowledge. The integration of a personality may be the highest aim of many psychiatrists. Hence the unity of science is one among a passionate crowd. A larger discourse on unity would make us better understand its appeal.

It is easy nowadays to be flippant about unity. It is also possible to be angry. Some of the current rage against reason is directed at an ideology of science that says there is one ultimate reality, one ultimate truth, one road to the truth (the scientific method), one sound mode of reasoning, one rational way of speaking. Because unity now rings in our ears as hegemonic, patriarchal, imperial, it is important not to dismiss the old virtues and values of unity. Oppressed people in the past, today, and in all the foreseeable future require those very onenesses about which we find it so easy to be lazy. People resisting despotism and its lies need ideals of one truth, one reason, one reality, and on occasion, one science. To be able to be critical of the unities is a luxury, and let us never forget it.

Unity has been an immensely powerful political tool, sometimes for what I find good, sometimes for what I find evil. To counter my logical observation that unity is not a predicate, I should notice a linguistic point about the concept of unity. With the logical point in mind, but forgetting the linguistic point, it would be hard to guess why unity could be deemed a virtue. It can be a virtue because two distinguishable although interconnected ideas are at work: not even unity is at one with itself.

The root word is unum, one, and for sure unity connotes singleness, oneness. Now being a singleton, that is, being the only instance of a concept, is never in itself a virtue. But there is a virtue in the offing. We can speak in context of unity's being desirable: the unity of certain concerts that we have heard, novels that we have read. A speech, a political platform, may or may not have unity. So may a character, a soul. This unity has something to do with the integration or harmony of the parts, a harmony that exists or does not exist after the item has already been individuated as one thing, one concert (starting at 8:00 and ending at 1o:30, with an intermission).

Let us call these two aspects of unity singleness and harmonious integration. The unity of the God of Israel or of Islam is singleness. This unity is not a property of the God and hence in itself not a perfection. Many generous souls have urged that the gods of all faiths are identical, but this is to say that we have different names and rites for the same thing, not that the same thing has the property of being identical to itself. The unity of the self, in contrast, is largely a matter of harmonious integration. That can be a real property of a person, a property that the person works hard to acquire. All traditional Western psychology regards such harmony as desirable. This may be connected with our notion of the soul. Many other peoples actively cultivate alternative personalities; individuals who do so successfully have special status, for example as shaman. Integration, I think, is something to be valued (or not) within a culture. But at least it makes sense to value it.

There are deep conceptual questions about the relation between the idea of singleness and the idea of harmony. In the unities that Aristotle introduced into the criticism of drama, the events are supposed to take place in one compressed period of time, and to be integrated. Everything happens (perhaps) in a single day. Is the integrated-harmony side of unity part of an archaic theory about what makes for singularity? For example, does it represent the idea that if there is a single entity, it will have, in some sense of "cause" (perhaps one of Aristotle's causes), a single cause, and thus betray a set of harmonious effects? Or, in contrast, is the idea of oneness to be thought of as derived from an experience of harmony? Is what makes us count something an entity, a singleton, something like the harmonious relationship of its parts? (In Leibniz's monadology, were it not for the preestablished harmony there would be infinitely many actual worlds—the monads—rather than one actual world.) If harmony is what makes for identity, then the logicians with their second-order-concept notion of unity have missed something of fundamental metaphysical and conceptual importance.

I shall not further discuss these hard questions. But the distinction between singleness and harmony matters to the very idea of the unity of science. Some of the unities of science that I shall mention have to do with singleness, and some with harmony. Not even unity is united.

Three Unities of Science

The unity of science denotes at least three distinct families of theses, each of which can be subdivided or organized in numerous ways. The first family is metaphysical, a collection of ideas about what there is. I see it as starting with a certain metaphysical sentiment, which is then followed by a number of what are, by comparison, relatively plain theses. The theses do not follow from the sentiment, and none entails the others.

The second family is a collection of practical precepts about the sciences. They correspond, in a rough and ready way, to the metaphysical theses. They have to do with method and the aims of the sciences. Each draws on a different insight about what scientists are up to. One and only one of these has deeply moved working scientists, chiefly physicists, namely the precept that one should try to find connections between important phenomena that have hitherto seemed independent. Integration and harmony are what seem to attract the scientist.

The third family forms a set of theses about scientific reasoning, and includes both logic and methodology. It seems to be almost completely independent of the preceding two. Thus the logical or methodological theses could be correct, and even the core metaphysical sentiment and all that flows from it, theoretical and practical, could be wrong. This is the strongest instance of the disunity among unity ideas, but at every possible juncture we find disunity. The lowest denominator of disunity is found at the level of methodology, where a wide spectrum of analytical philosophers of science have asserted that there is one scientific method applying across the board in the natural and human sciences (they are all on the same side in the Positivismusstveit)—and then produce seemingly incompatible methods, the well-known methodology disputes of Carnap, Popper, Lakatos, and the rest. But this is the least instructive kind of disunity, because it is at the level of philosbphical doctrine rather than at the level of scientific activity.

Many readers will want to reorganize my kinds of unity. My purpose is not to structure the distinctions but rather to display some of them in a handy way. I am cautious about every one of the unities that I shall mention. My interest is less in skepticism than in the fact that the grounds for skepticism differ from unity to unity. There are a lot of different types of unity, each of which may be called in question for its own specific set of reasons.

A Metaphysical Sentiment

The unity of science is rooted in an overarching metaphysical thought that expresses not a thesis but a sentiment. Since it is not exactly a doctrine, it lacks straightforward expression. On the one hand I could try, "There is one world, one reality, one truth." On the other: "There is one world susceptible of scientific investigation, one reality amenable to scientific description, one totality of truths equally open to all scientific inquirers who may share their techniques and experiences." I shorten the latter to, "There is one scientific world, reality, truth." Inserting the word "scientific" seems to beg some questions—Wasn't science supposed to be what got at the world, at reality, at truth? So should we not eliminate the modifier "scientific," thus going back to the first unequivocal statement?


Excerpted from THE DISUNITY OF SCIENCE by Peter Galison, David J. Stump. Copyright © 1996 Board of Trustees of the Leland Stanford Junior University. Excerpted by permission of Stanford University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Read More

Meet the Author

Peter Galison is Mallinckrodt Professor of the History of Science and of Physics at Harvard University. He is the editor, with Bruce Hevly, of Big Science: The Growth of Large-Scale Research. David J. Stump is Assistant Professor of Philosophy at the University of San Francisco.

Customer Reviews

Average Review:

Write a Review

and post it to your social network


Most Helpful Customer Reviews

See all customer reviews >