«RECONCILING SEARLE AND DREYFUS Closing the Gap: Phenomenology and Logical Analysis By Sean Dorrance Kelly I I AGREED TO WRITE THIS ARTICLE ON T IS ...»
RECONCILING SEARLE AND DREYFUS
Closing the Gap: Phenomenology
and Logical Analysis
By Sean Dorrance Kelly
I AGREED TO WRITE THIS ARTICLE ON
T IS WITH BOTH PRIDE AND HESITATION THAT
phenomenology and logical analysis. John Searle and Bert Dreyfus are
for me two of the paradigm figures of contemporary philosophy, so I am extremely proud to have been offered the opportunity to engage with their work. The editors of The Harvard Review of Philosophy, it seems to me, have shown a keen sense of what is deep and important in our discipline by publishing extended interviews with these two influential thinkers. At the same time, writing this article meant entering into a debate between Searle and Dreyfus about the priority of their respective philosophical methodologies, and this, I am afraid, is at best a no-win situation. My strategy, therefore, has been to try to engage as sympathetically as possible with the programs of each philosopher, and to draw from their work lessons that seem to me important for any philosophical account of intentionality and social reality. I have not shied away from criticism; indeed the whole paper is an extended series of criticisms of the work of both philosophers. But I hope they will recognize this for what it is: an engagement with the work of each that is based on the deepest respect and admiration.
Before I jump into the fray, I should mention that I owe a personal debt of gratitude to Searle and Dreyfus as well. They were not only my two advisors in graduate school, but were the two main reasons I entered philosophy. Having become dissatisfied with the formal worlds of mathematics and computer science, their respective critiques of artificial intelligence and cognitivism showed me the rich possibilities of philosophy and struck me as a reason to become a philosopher. It is a difficult discipline, sometimes unforgiving; but I am glad I followed them into it.
§1. Phenomenology and the Analysis of Intentionality
ONE OF THE FIRST THINGS THAT ATTRACTED ME TO JOHN SEARLE’S WORK WAS THEChinese Room Argument. Searle’s famous argument against the prevailing background assumptions of cognitive science—especially against the functionalism that most cognitive scientists took as a working principle at the THE HARVARD REVIEW OF PHILOSOPHY vol.XIII no.2 2005 Closing the Gap: Phenomenology and Logical Analysis 5 time—seemed to me to focus in a clear-eyed and brilliantly straightforward way on exactly the right kind of issue: it emphasized the importance, and the deeply puzzling nature, of the first-person perspective. As I worked with Searle over the years I came to appreciate that one of the striking characteristics of almost all his work is how, standing bravely against the mainstream of contemporary analytic philosophy of mind, he managed to highlight again and again the fundamental importance of the first-person perspective. This focus is essential to such canonical Searlean doctrines as the intrinsic intentionality of human mental life, the inadequacy of functionalism, the “internalist” account of meaning and of names, and many others.
Searle’s emphasis on the importance of the first-person perspective ought to make phenomenology a natural ally for him. It ought to make Searle a natural ally for the phenomenologists, too. And in a way it has. But alliances take many forms, and over the years this one has taken the form of a series of sometimes acrimonious debates. Since both parties agree that the first-person perspective is essential in some way or another to philosophical theorizing about the mind and world, these debates have focused on what role exactly it ought to play.
Although they have often been rich and interesting, I have always felt that the debates between Dreyfus and Searle overplay the differences between their respective methodologies. My view is that both phenomenology and Searlean logical analysis are essential to any full account of intentionality and social reality.
One goal of this paper, therefore, is to narrow the gap between them.
Sometimes the debates between Searle and Dreyfus have been based on misunderstandings. One of the latest rounds, which has centered on the ontological question of what there is, is a case in point. When one is deeply committed to the importance of the first-person perspective, as both Searle and the phenomenologists are, one runs the risk of falling into Idealism—the ontological view, roughly, that the only things there are are those things that are revealed from the first-person perspective. Searle has claimed recently that a crucial difference between himself and the phenomenologists is that they are, but he is not, committed to this kind of Idealism.1 Now, there is no doubt that many great philosophers who have taken the first-person point of view seriously have also succumbed to the Idealistic itch. Berkeley endorsed perhaps the most radical version of Idealism, of course, but many phenomenologists have fallen into the trap as well. Husserl, for instance, was a kind of Idealist from the middle of his career forward and Merleau-Ponty was an avowed Idealist throughout.
Despite this circumstantial evidence, however, there is no necessary connection at all between phenomenology and Idealism. Early Husserl was a committed Realist, and although Heidegger does, in some of his moods, seem to deny the independent existence of the universe, there is a healthy debate in the Heidegger literature about whether Idealism or Realism was his considered view. Since Dreyfus is well-known as the defender of the Robust Realist interpretation of Heidegger, this ontological issue actually turns out to be an area of agreement rather than of disagreement between him and Searle: both are committed to defending a Realist view about the nature of the universe.
One place where phenomenology and logical analysis may come more into conflict, however, is in the analysis of the contents, or as Searle calls them, vol.XIII no.2 2005 THE HARVARD REVIEW OF PHILOSOPHY 6 Sean Dorrance Kelly the conditions of satisfaction, of various intentional states. Here I believe that the conflict is genuine, but it is every bit as much a conflict within Searle’s own work as it is a conflict between him and the phenomenologists. Fortunately, there is an easy way to resolve the conflict. Before I get to the resolution, however, let me begin by laying out the issues.
The easiest way to see the problem is to start with an example. In Intentionality, Searle argued that causal self-referentiality is one of the conditions of satisfaction of perceptual states.2 In order really to see an object in front of me, he argued, as opposed to merely seeming to see it, the object must cause me to have the visual experience I am having. For example, if I have a visual experience of a truck, but the truck I seem to see is not what is causing me to have that visual experience, then the conditions of satisfaction for the visual experience are not met, according to Searle. That is because, he claims, the world turns out to be different than my experience represents it to be.
This analysis of the conditions of satisfaction of perception buys into a certain bit of Grice’s causal theory of perception. In particular, it buys into the idea that when I am seeing an object, the object I am seeing is the one that I am in causal contact with. Searle builds this Gricean condition, or something like it, directly into the contents of the perceptual experience: among other things, if there is no object with which I am in causal contact, or if the object I am in causal contact with is different from the one I seem to see, then the world is not the way my experience represents it to be. The causal self-referentiality condition, therefore, is an essential part of Searle’s analysis of the contents of perception.
The conflict I am interested in comes about when one recognizes a different, but also seemingly natural, condition on the content of perceptual experiences. For surely, one might think, what counts as the content of a perceptual experience ought to be tied in some close way to its phenomenology.
If that is right, then we need to ask ourselves this question: Is it a part of the phenomenology of seeing things—is it, in other words, either what I am or what I could be conscious of when I have a perceptual experience of an object—that the thing I seem to see is causing me to have my experience of it? In their debate with Searle over this issue Dreyfus and Dagfinn Føllesdal argued that it was not. After all, what would it be like, experientially, to see something to be causing the experience I am having of it? Perhaps we will say that I can see one billiard ball causing another to move by hitting it. This is contentious, of course, and a strict Humean theorist of causality, or even a strict sense-datum theorist of perception, would deny it outright. But even if we accept the possibility of seeing that kind of object causation, it does not seem possible even to conceive of the more complicated case in which I see something to cause the experience I am having of it. After all, what would the difference be, experientially, between a visual experience of the truck in which I did see it to be causing me to have the experience, and an otherwise identical visual experience of the truck in which I did not? If we take the phenomenology as a guide to the content, therefore, the causal self-referentiality condition does not belong in the content of perception.
One option at this point is to try to find reasons for believing that causal self-referentiality really is part of the phenomenology of visual experience. In personal communication Searle himself has pressed this line.3 The way to focus THE HARVARD REVIEW OF PHILOSOPHY vol.XIII no.2 2005 Closing the Gap: Phenomenology and Logical Analysis 7 on the phenomenology of causal self-referentiality is to consider the case of a person with an extraordinarily good visual imagination. The person in question is so good at imagining visual scenes that she can have an imagined visual experience of, say, a yellow Volkswagen that is equivalent in every visual detail to the experience of the yellow Volkswagen she would be having if she were actually looking at it. The vibrant shades of yellow, the shape of the car, the elements of the visual background, all of these are identical in the visual experience and the imagined visual experience. Still, Searle argues, there is an important difference between the two. Namely, it is part of the imagination, though not part of the visual experience, that I am causing this visual experience.
The imaginary experience, in other words, lacks the causal self-referentiality.
Since ex hypothesi this is the only difference between the two experiences, it shows that causal self-referentiality must be part of the phenomenology of normal visual experience.
I believe this is an ingenious, but flawed, attempt to save the causal self-referentiality clause. It is flawed because it does not give us any reason to believe that causal self-referentiality as Searle understands it is part of the content of visual experience. To see this, think again about the phenomenology of vivid imagination. Let us accept, at least for the purposes of argument, that imagined scenes feel as though they are caused by me. And let us even accept that visual experiences are such that they feel like they are not imaginings.
Even giving Searle all this, the most one is justified in putting into the content of the visual experience is that the thing seen is not caused by me. In short, if visual experiences feel like they are not imaginings, and imagined scenes feel like they are caused by me, then visual experiences feel like they are not caused by me. But to say that the visual experience I am having is not caused by me is substantially weaker than to say that the visual experience I am having is caused by the object it is an experience of. The imagination case only justifies the weaker of these two claims.
Instead of arguing that causal self-referentiality is a part of the phenomenology of visual experience, therefore, Searle sometimes seems to accept that it is not. 4 When pursuing this line of thought, however, he does not conclude that causal self-referentiality is excluded from the content of experience, but rather that phenomenology is impoverished. He argues that since causal selfreferentiality is a part of the conditions of satisfaction of perceptual experience, and since this causal self-referentiality condition is beyond the reach of phenomenology, phenomenology is not sufficient for logical analysis. As he says in his interview with The Harvard Review of Philosophy, “[T]here is a whole area that is beyond the reach of phenomenological analysis” (p. 114).
If Searle’s project were merely a third-person analysis of the conditions that need to be met for something to be a perception, then this claim about the impoverished nature of phenomenology would be defensible. Grice’s own account was a third-person analysis of perception like this, and I believe it is true that if your only resources were phenomenological you would not be able to generate Grice’s causal theory of perception. The problem is that when Searle builds the causal self-referentiality constraint into the intentional content of perception, Searle’s own commitment to the importance of the first-person vol.XIII no.2 2005 THE HARVARD REVIEW OF PHILOSOPHY 8 Sean Dorrance Kelly perspective in analyzing intentionality creates an internal conflict. To see this we have to introduce a principle that makes clear Searle’s commitment to the first-person perspective in the analysis of intentionality. This is what Searle calls the Connection Principle.