Images of Empiricism: Essays on Science and Stances, with a Reply from Bas C. Van Fraassen
An Alternative to Realism
Scientific realism is the view that the aim of science is to produce literally true theories about the world. Although realism is probably the commonest philosophy among scientists, and perhaps even dominates among philosophers of science, alternative views abound, even after a century of serious inquiry about realism in science. I take it as a premise that if any discussion about the relationship of science to religion in our society is to be taken seriously, it cannot presuppose about science those things that are most uncertain—among these is whether or not the aim is to provide theories that are literally true.
But this so often is presupposed, and not only in general discussions about science, but especially when it comes to the relationship between science and religion. Indeed, there is virtually no discussion of what this conversation would look like if it entertained reasonable contrary proposals to scientific realism. A rare counter example is Alister McGrath’s three-volume Scientific Theology.1 But even here, the goal is only to advance a realist position. It does not seriously entertain the idea that perhaps we should not be realists. It should come as no surprise that McGrath is a realist, of course. He draws heavily on the twentieth-century Scottish theologian Thomas Torrance, whose Theological Science,2 like McGrath’s Scientific Theology, sought to establish objective scientific grounds for theological reasoning. But, what if their presumptions about science, both concerning its method and the truth of its content, turn out to be flawed? What if objectivity as construed traditionally turns out not to be the grand explanation for science’s great success? Will the project of an objective theology then languish? And what theology was lost for the pursuit of objectivity? What reasons, then, will Christians have for being realists? Do Christians have such reasons now?
In seeking answers to these questions, we would do well to consider the alternatives. Actually, there are a number of anti-realisms and qualified realisms with respect to science, ways that devolve neither into anything-goes relativism nor idealism. One that has captured the attention of both religious philosophers and philosophers of science is the constructive empiricism of Bas van Fraassen. The position was first introduced fully in his book The Scientific Image.3 The position holds that science aims to give us theories of the world that are adequate empirically and that acceptance of a theory requires belief only that it is adequate empirically. In even this very brief synopsis of constructive empiricism, already we see the two concepts—empirical adequacy and theory acceptance—that are the defining features of the position.
In 1980, not everyone was ready to accept the theory of constructive empiricism. In fact, hardly anyone did. And a large secondary literature has grown up around the theory.4 Thus, in the decades that followed, the theory underwent some modest revisions, developed both in books like van Fraassen’s Laws and Symmetry5 and in numerous contributions to technical philosophy journals. An important turning point in the theory was then reached in van Fraassen’s next book The Empirical Stance.6 The argument introduced in this book is important for a number of reasons. For philosophers, as will be seen shortly, the argument was of a very general kind with very general implications for epistemological methodology and, indeed, for the resolution of philosophical disputes of all different kinds. For readers of this journal, it will also be of interest for its content. The Empirical Stance is the published version of van Fraassen’s Terry Lectures series at Yale University in 1999. In the book, therefore, he takes up not only questions of philosophical method, but also a question at the heart of the problem of science and religion. What, asks van Fraassen, is science after all—and what is it to be secular?
To arrive at the answer, van Fraassen first advances a very basic thesis whichis central to all of the following arguments in The Empirical Stance. To motivate the thesis, Van Fraassen begins by reviewing the history of empiricism (to which his own version is intellectual heir) and seeking to identify whatever characteristics there are of a view that render it empiricist. That is, through historical analysis, van Fraassen aims to identify foundations of empiricism. This is problematic of course, exactly because empiricism is notoriously anti-foundationalist. We are thus led to the seemingly innocuous assertion that van Fraassen calls “Principle Zero”:“For each philosophical position X there exists a statement X+ such that to have (or take) position X is to believe (or decide to believe) that X+” (41). Principle Zero is about dogmatism in philosophy, and though not obviously correct, it is attractive. What might a philosophical position be, if it cannot be summarized, at least in some form, in some statement? As it turns out, it is the main project of The Empirical Stance to demolish the common enough presupposition articulated as Principle Zero. But first, more on empiricism. Van Fraassen wants to explore what, for empiricism, could be that statement X+. Whatever it is, it must be strong enough tosustain the empiricist argument against metaphysics and satisfy the empiricist conditions for a good argument simultaneously. With respect to this latter criterion, van Fraassen argues that it cannot be a tautology, and must therefore be a factual claim. We turn, then, to what van Fraassen calls the “third characteristic of empiricism,” asserted in The Empirical Stance, but actually developed and defended insome previous articles: “As in science, so in philosophy: disagreement with any admissible factual hypothesis is admissible” (43). There is a problem now, however. Principle Zero and the third characteristic of empiricism cannot both be correct because the statement X+ that characterized the empiricist view (whatever that statement is) must be both definitive of the empiricist view—in which case it cannot be contradicted (by an adherent)—and, as factual a hypothesis, its contrary must be admissible. The compelling conclusion is that either empiricism is not a proper philosophical view (meaning, that empiricism is inconsistent with Principle Zero) or Principle Zero is wrong. But Principle Zero is so broad that if it is wrong, it is hard to see what else might constitute a philosophical view. At any rate, it will not be merely propositional. Well, actually, this is precisely what van Fraassen thinks. A philosophical position need not consist in holding a doctrine or dogma, because it might consist in a stance.
The remainder of The Empirical Stance is devoted to exploring this idea, first characterizing what a stance is (a cluster of attitudes, commitments, and methods that may include beliefs but is not reducible to them), then walking through some implications: materialism is a stance misunderstood as a doctrine; scientific revolution as a philosophical problem solved by a voluntarist epistemology; the roles of experience, ambiguity, and vagueness in the pursuit of knowledge; and finally, the persistent tension between religious and scientific belief.
The potential consequences of the thesis that a philosophical position may bea stance are manifold—as the secondary literature on this book attests. In The Empirical Stance, van Fraassen takes up one in particular. In the final chapter of this work, van Fraassen turns his attention to the tension between religious beliefs and science. Specifically, he asks: what is science and what is its relationship to the secular? For most people, the natural answer to the question about science is that it has something to do with objectivity. But is scientific practice in fact objective? Certainly not. If we have learned anything since Thomas Kuhn’s The Structure ofScientific Revolutions,7 it is that science is not objectively objective. What role, then, does objectivity place in science? Van Fraassen’s answer is that scientific inquiry is objectifying inquiry. That is, the objective switches from description to action, from adjective to verb. What is it to objectify? Van Fraassen identified three relevant senses. The first, which he calls “objective distancing,” is the conceptual or actual procedure of “taking ourselves out of the picture” and treating the object of analysis as a thing (157). The second, which he calls “objective neutralization,” is the removal of our interests (158). This is the construction of an ostensible value-neutral discourse with respect to nature. The final and most important sense concerns science’s very special mode of asking questions about nature, particularly the demarcation, before any data collection or analysis is performed, of the domain of investigation, the form questions might take, and the ways in which the response of nature will be taken as speaking to these questions. It is this final sense of objectification, called “objectifying inquiry,” that is most subtle and most central to van Fraassen’s understanding of science.
Although he never puts it quite this plainly, I read van Fraassen as saying thatall and only scientific inquiry is objectifying inquiry, at least when it is pursued successfully. And, we are enabled now to address our second question about what it means to be secular. Van Fraassen’s answer is that just as objectifying inquiry is characteristic of science, so the objectifying attitude that accompanies it is characteristic of secularity. An objectifying attitude is one that has in it no place for the mystical, no place for an encounter with the divine. Can we have objectifying inquiry without the objectifying attitude? Can we have both science and religion? Van Fraassen leaves us with a metaphor:
I see objectifying inquiry as the sine qua non of the development of modern science and its incredible, breathtaking achievements in our increasing knowledge of nature. At the same time, while this objectifying inquiry has brought us untold riches, what does it profit us to gain the whole world and lose our own soul? Riches come with a temptation, a tempting fallacy, namely, to have us view them as all there is to be had, when they are so much. This is true of all riches, and it is true of the riches of objective knowledge. Poor are the rich who succumb to this fallacy (195).
One is reminded of the saying of Jesus that it is easier for a camel to go through the eye of a needle than for the rich to enter the kingdom of God,8 and what also follows, that though the path may be fraught with difficulty, for him who is willing to forsake riches if need be, there is nonetheless the promise of rest, for “things which are impossible with men are possible with God.”9
Empiricism, Science, and Religion
Now a new book of criticism, Images of Empiricism, has been published.10 In this book, fourteen authors take up not only the methods and questions raised by The Empirical Stance, but also many from the whole corpus of van Fraassen’s work on constructive empiricism, which in light of The Empirical Stance, we see now as having implications both for the satisfaction of philosophical questions in general and the conduct and relationship of religious inquiry to science in particular. This book is divided into two main parts. After a general introduction by Bradley Monton, the first half takes up the questions and problems of constructive empiricism in general. Essays by Maarten van Dyck, Nancy Cartwright, James Ladyman,Alexander Bird, Philip Percival, Peter Lipton, and Stathis Psillos take positions ranging from nearly full agreement to outright opposition.
Of these chapters, the most interesting to readers of this journal may be the essay “Accepting Contradictions” by Peter Lipton. Lipton identifies van Fraassen’s notion of acceptance as the very heart of constructive empiricism. Recall that, as explicated by van Fraassen, acceptance is the attitude toward scientific theories that holds that a theory’s observable consequences are true, but requires no such belief about the unobservable parts. Theories, then, are adequate empirically when the observables they entail are matched exactly to the data. Importantly, there are no evidentiary grounds for choosing among empirically adequate theories.
Lipton’s guiding principle might as well be “as in philosophy of science, so in philosophy generally.” The project he undertakes in this chapter is to generalize van Fraassen’s notion of acceptance in such a way as to work in all areas of epistemic life: for example, to take constructive empiricism (at least the part that says one can accept a theory without believing it) beyond the philosophy of science. In Lipton’s generalized version, acceptance means to take toward some set of propositions an attitude wherein belief is accorded only to a privileged subset. Constructive empiricism is the view that this subset is comprised of the observable consequences, but it is easy to imagine alternatives. Of what use is acceptance, thus generalized? Lipton identifies at least one virtue, managing contradictions, and it is to the development of this idea that he devotes the remainder of his essay.
To get started, Lipton notes that it must be possible for an individual to have contradictory beliefs, since we all have them. However, when we become aware of the contradiction, something changes in our attitude, for we cannot believe something we know to be false, as the conjunction of two contradictory propositions is. How to manage these contradictions is precisely the problem countenanced by myriad scientists, philosophers, priests, religious laypersons and, indeed, epistemic agents everywhere. Indeed, it is on contradictory claims of facts and doctrines that so much of the persistent tension between religion and science centers. But, argues Lipton, now that we are in possession of epistemic notions short of belief, such as acceptance, we have options. We can, for instance, accept contradictions, since the subset of propositions marked for belief can be selected such that it itself does not entail contradictions. Of course, we do not have to accept contradictions. Lipton points out that we still have available to us the old option of contracting the content of the propositions entertained. So, what we have is in fact a more complex set of possible responses, for we may contract content or reduce the scope of belief (we may back off from full-blown belief to a position of acceptance), and we are not restricted by either philosophy or rationality in how we choose to execute these options. We may pick and choose.
Now, let us not allow either our philosophy of science or philosophy of religion to get too watery. Certainly we should be wary of any maneuver that appears to remove the force of all such contradictions with one fell swoop. I, for one, think we should be wary here. Some content cannot be contracted, at least not without forsaking orthodoxy. (Is it not practically the definition of orthodoxy to be dogmatic, even doctrinaire?) Perhaps there can be philosophy without dogma. But I doubt that either religion or science can be reduced merely to stances and survive intact. Still, the notion of acceptance as proposed by van Fraassen and amplified by Lipton might do a great deal of work. The problem of managing contradictions arises not only in debates between fundamentalists of different camps (such as creationism and evolution), not only between science and religion, but also in other arenas, including within science itself. Examples highlighted by Lipton are the contradiction between the scientific world picture (in which the atoms that makeup a table are comprised mostly of empty space) and the picture presented by experience (in which tables are solid); the contradiction between ideal models in science, needed for pragmatic reasons, and the equations we would solve if only we knew how; and the inevitable inconsistency between predictions and data. This is a stimulating chapter, but there is much in the details that has yet to be worked out.
The second half of Images of Empiricism takes up The Empirical Stance in particular. Contributions are by Ernan McMullin, Anjan Chakravarty, Chad Mohler, Michael Bitbol, Anja Jauernig, and Dien Ho. Of these, I found the essays by Jauernig and Ho to be the most stimulating.
Of all chapters in this book, Jauernig’s is one of the most squarely opposed. Her main thesis is that empiricism, indeed any philosophical position, cannot be a stance. Assuming (as Jauernig argues we should) that a necessary condition forsomething to count as a philosophical position is that it is intended to make the world and the human condition understandable in an explicit, intellectual way, she argues that a stance cannot be a philosophical position because values, commitments, and goals (the elements of a stance) cannot be intended to make the world and the human condition more intelligible “in the relevant sense” (310-311). This argument is not very compelling (to me, anyway), both because the premise is doubtful and because so much hinges on what precisely is meant by the “relevant sense.” When Jauernig takes a more defensive position with respect to metaphysics, however, the arguments become more pointed, more interesting. “So far,” writes Jauernig near the end of her essay, “we haven’t talked about the question of what philosophy’s distinctive contribution could be to the project of making the world and the human condition more intelligible to ourselves in an explicit, intellectual way” (313). And to this question Jauernig has an answer, one which has a role for metaphysics: “to provide theories that complement scientific theories about the world… to arrive at a more complete story” (315). That is, if science leaves off at unobservables (as van Fraassen, among others, thinks it should), then metaphysics can take over. What does this complementing of scientific theories consist of? Precisely the same kind of engagement envisioned by constructive empiricism. “These metaphysical theories are not presented as true but as metaphysically adequate, that is, as possible and explanatorily or intellectually satisfying” (315). After all, wonders Jauernig,
isn’t it plausible to say that understanding the world and the human condition involves knowing the possible ways the world could be with respect to those aspects or domains that cannot in principle be empirically investigated, and which are, thus, inaccessible to the empirical sciences? Couldn’t that be one of the distinctive contributions that philosophy has to offer? (316)
Plausible indeed. In fact, more than plausible, for if not to philosophy, where else is science to turn for an evaluation of the possible?
As I said, this defensive turn in the argument is a good deal more compelling. But there is more. Switching to the potential mood, Jauernig anticipates an objection. Is there anything at stake? What could be gained from metaphysical speculation? Taking her cue from the voluntarism already well in play (for Van Fraassen’s philosophy itself is voluntarist), Jauernig remarks that the answers to these two questions may not be universal. For here, surely, a satisfactory answer will depend on the values and attitudes of the questioner. But one answer, at least, is at hand, one which Jauernig herself finds to be satisfying. What we gain is the “satisfaction of our natural desire for a complete description of how things might be” and this, surely, is a deep part of understanding the world and the human condition (316). Indeed, even if one does not agree with Jauernig that this is the interesting question, surely it should be admitted as one valid possibility among many. Jaurneig’s conclusion, then, is summarized in the motto “let a thousand (metaphysical) flowers bloom!” because “the more metaphysical systems there are, the more of conceptual space is covered, and the more we understand” (316).
In the end, van Fraassen is afforded the opportunity to reply to his critics and amplify the ideas of his adherents. Jauernig’s arguments are taken up at length. As usual, his reply is marked by both grace and confidence. However, rather than reiterate his answers to Jauernig, I would like to turn briefly to one final essay, “Farewell to Empiricism” by Dien Ho.
Of all the chapters in Images of Empiricism, Ho’s was the only one which I found consistently provocative. The thesis of Ho’s chapter is that, in the light of van Fraassen’s argument in The Empirical Stance, philosophy is now at a crossroads. Van Fraassen, Ho believes, has correctly brought to light a very severe problem with the “philosophical orthodoxy.” As summarized by van Fraassen, the problem is to find some justified foundationalist doctrine that epitomizes empiricism, and that any such would be in violation of empiricism’s own tradition (the third characteristic). As brought out by Ho, however, any philosophical position suffers the same liabilities of circularity and regress. The solution proposed by van Fraassen—to characterize a philosophical position as a stance, complete with emotions, values, and whatever else supports reasonable subjective evaluation—is an attractive one. But, if we extend the “permissible philosophical moves” to include these non-rational judgments, then the distinction between philosophy and other forms of persuasive writing begin to dissolve. As Ho points out, this may not be so bad in real life, where acknowledgment of the non-rational element of belief is an inducement, not to despair, but to a life of tolerance and humility. But it will be a bitter pill for philosophy.
The reason I found Ho’s essay so provocative is that although his sparse, assertive prose invited me to question many of his basic premises, I still found his overall logic compelling. If I inserted my own premises, the argument still held. For instance, he writes of Van Fraassen’s original empiricist position that “the problem… is that empiricism is supposed to be the ultimate methodological theory in the sense that it tells us the source of knowledge tout court” (320). Really? I did not take it that way. There is no doubt that empiricism is meant to be pretty thorough-going. But, “the source of knowledge tout court” is a bit extreme. Later on, “value conflicts, on the other hand, are typically thought of as being potentially unresolvable” (327). Similarly, it seems to me that the mere existence of aesthetic, ethical, and political discourse shows that in general we think value conflicts are in fact resolvable. At least, they are not any more unresolvable than the many disputes about matters of fact in which the conditions for the establishment of facts (such as perception or semantics) are subjective and controversial. But no matter. I insert my own views about the aim of empiricism or resolving value conflicts and his argument stands. The only problem that remains is the constructive project of providing guidance for how we ought to resolve conflicts (whether over values or facts), given that we acknowledge as proper the role of the non-rational in our coming to beliefs of all kinds.
Should Christians be Realists?
In conclusion, we return to our main question: should Christians be scientific realists? In answering this question, we would do well to remember that although reasons to reject constructive empiricism are not reasons to accept realism in and of themselves, reasons to accept empiricism are indeed reasons to reject scientific realism
Many Christians will think that to be both consistently and distinctly Christian requires realism about some, perhaps many, things. I agree. Thus, for instance, the Apostle Paul instructs the Corinthians clearly to take the doctrine of the resurrection literally: “And if Christ is not risen, then our preaching is empty and your faith is also empty.”11 This is a claim of factual content and very much at the core of both Christian belief and practice. Of course, it is possible to relax either consistency or distinctiveness and remain in one’s heart a Christian. Nonetheless, I submit, there is no consistent, content-free Christianity.
There is a parallel in science. Although, at least according to constructive empiricism, this condition does not hold for science (the unobservable content of any scientific theory is open to revision), it still is the case that to participate meaningfully in the community of science, one must accept the standard scientific theories, or most of them. And acceptance is not a trivial attitude to take. So, constructive empiricism is not a solution to all possible conflicts between scientific and religious inquiry. Still, it may be a solution to some. And the step suggested by Lipton may be generalized further still. Belief is not a dichotomous option. Lipton’s generalized version of acceptance says that one stipulates a subset of propositions that are the ones to be believed in. Very well. But what attitudes do we take to the rest?There are more attitudes toward propositions than simple belief and non-belief (such as agnosticism), and a generalized version of generalized acceptance must take up the task of specifying which attitudes should be taken to which subsets of propositions and when. I submit that a more nuanced view of the way believers in both science and religion actually accommodate seeming contradictions might go a great way toward upending some of the supposed conflict. Of course, this is an option that is not available to fundamentalists of any stripe, for whom literal belief in all doctrines is the characteristic sign of the true believer.
What, then, of reasons for realism? Van Fraassen’s original reason for being a constructive empiricist is that it brings into concordance our understandings of scientific knowledge and practice. It allows us, for instance, to understand the transition in scientific revolutions without appealing to unbridled irrationality. Lipton, by contrast, has a practical reason for being a constructive empiricist: to manage contradictory doctrines in an epistemically respectable way. Thus, scientific realism is not the only game in town and constructive empiricism, particularly, is a promising alternative. In executing this empiricist program, van Fraassen has identified an important, general philosophical issue (philosophical positions as stances) with ramifications for understanding the relationship of science to religion in particular (materialism is a stance, not a doctrine). The implications of this view for understanding both science and religion are addressed first by van Fraassen in The Empirical Stance and then developed further in secondary critical literature. The reasons for constructive empiricism hold no less for the Christian than for others. There may yet be additional reasons for realism, even some which are especially Christian. But I do not know them.
Cite this article
- A. E. McGrath, A Scientific Theology, Volume 1: Nature (Grand Rapids, MI: W.B. Eerdmans,2001); McGrath, A Scientific Theology, Volume 2: Reality (Grand Rapids, MI: W.B. Eerdmans,2002); McGrath, A Scientific Theology, Volume 3: Theory (Grand Rapids, MI: W.B. Eerdmans,2003).
- T. F. Torrance, Theological Science (Oxford: Oxford University Press, 1969).
- B. C. Van Fraassen, The Scientific Image (Oxford: Oxford University Press, 1980).
In his 2007 introduction to a chapter in one of the books under review here (Images of Empiri-cism), Maarten van Dyck wrote, “Undoubtedly, The Scientific Image has been one of those fewbooks that really had a profound impact on the philosophy of science… Exaggerating only alittle, one could even say that trying to refute van Fraassen’s position in that book soonbecame one of the standard exercises that one had to pass to qualify as a truly realist philoso-pher. And who didn’t want to be a realist—in one of its many guises?” For earlier critiques,see especially Images of Science, eds. P. M. Churchland and C. A. Hooker(Chicago: ChicagoUniv. Press, 1985).
- B. C. Van Fraassen, Laws and Symmetry. (Oxford: Oxford University Press, 1989).
- B. C. Van Fraassen, The Empirical Stance (New Haven, CT: Yale University Press, 2002).
- T. S. Kuhn, The Structure of Scientific Revolutions, 3rd ed. (Chicago: Chicago University Press,1996).
- Luke 18:25.
- Luke 18:27.
- Bradley Monton, ed. Images of Empiricism (Oxford: Oxford University Press, 2007).
- 1 Corinthians 15:14.