Transcript of Interview
Henry S. Thompson
Gordie Slack

21 February 1997

Center for Theology and the Natural Sciences

Religious Background and Spiritual Trajectory

GS: To start us off, would you please say a few words about your religious background, tracing your spiritual trajectory?

HT: Sure. My name is Henry S. Thompson. I was born in 1950 in Philadelphia. I'm an American by birth, although I've been in Edinburgh for sixteen years now, in Scotland. My parents were, roughly speaking, non-practicing atheists. They were the generation that fought the battle to not be terribly concerned about religious observance. Although I was baptized, we never went to church on Sundays. And I have fairly clear memories of going to an ordinary church service for the first time when I was thirteen or fourteen, when I was visiting some friends. But when I was in fourth grade, I changed from the local suburban public school that I was attending to a Quaker school, Germantown Friends School in Germantown in Philadelphia. By the time I left there at seventeen and three quarters, I considered myself a Quaker, not only because it's part of the life of most Quaker schools, going to Meeting for Worship during the week as part of the school schedule, but also by the time I was sixteen and a half or so, and pretty mobile on my own, I started coming to Meeting for Worship on Sunday. But when I went out to Berkeley, to University, in 1968, I didn't particularly find the Berkeley Quaker Meeting to my taste. Although I attended it sporadically, I didn't attend with anywhere near the regularity that I did before. It wasn't actually until after I moved to Edinburgh, sixteen years ago, that I started going to Meeting regularly again and, not too long after that, I formally applied for membership and was accepted as a member of London Yearly Meeting of the Religious Society of Friends, or Quakers. So I've sort of been a Quaker off and on all my adult life, but only formally for the last twelve or fourteen years. I am an active member of my Meeting; virtually every Sunday I go to Meeting for Worship, I served as an Elder for six years, I play in the Meeting ceilidh band ().

My kids have a moderately confused life because my wife is a practicing Catholic. That's also been a significant influence on my theology and my religious perspective. But our kids most Sundays will go to Mass and to Meeting. So I'm also fairly well plugged into the Dominican Chaplaincy here at the University of Edinburgh, which is our Parish Church, essentially. Dominican Catholicism is a particular intellectual breed of Catholicism and it's the one I know best. But there is actually a surprising amount in common between Catholicism and Quakerism. They sort of meet around the back, leaving the other Protestant denominations in a different category. The Catholics think so too. It's not just the Quakers that think this. It's somewhat curious. So that's the basis for my religious perspective. And that happens to be the reason why I don't use my title, "Dr.", because, for complicated historical reasons to do with life in the seventeenth century, Quakers have not used titles since then. It has to do with having been sent to jail for refusing to take their hats off in front of the king in the seventeenth century. And so, I try to remind myself about that testimony by not referring to myself or to other people with titles. It's a sort of a petty spiritual discipline, shall we say. I don't know if that's what you meant by "your spiritual trajectory."

GS: It is, although I'd like to know what happened when you were at Quaker school, other than the obvious influence of the context, that made you become interested in spiritual matters? Was it just a gradual deepening of interest? Or was there what you would describe as a spiritual event of any kind?

HT: It certainly was gradual. It's probably not surprising that conversion experiences are not high on the list of things that Quakers talk about, because of the silent worship. Silence is at the heart of the Quaker approach to spiritual discipline in general and worship in particular. Certainly, for me, what happened was a growing realization that I couldn't ignore the fact that on a Thursday morning when we had Meeting for Worship, these little old ladies who came in from outside school, who weren't teachers necessarily, but were members of the Meeting that had oversight of the school, and who came to Meeting for Worship with the school, would stand up and tell the truth in ways that I found compelling, that I couldn't address in any way other than to say taking these people seriously means taking their belief seriously, as well as the particularly useful things that I heard them say.

My take on Quakerism

There was space within Quakerism for the kind of hyper-intellectual approach to religion and theology that was comfortable and natural to me because I'm a scientist. I'm an analyst, a left brain kind of person. I ended up in linguistics and cognitive science and artificial intelligence, all of which are sciences, as various people have said, that pay you to take out your brains and play with them. And so an inarticulate, conversion experience kind of religion was never likely to be attractive to me. I think that's also why I've found the Dominicans so congenial, because, as I said, along with the Jesuits, they are the sort of intellectual wing of the Catholic Church. As I got older, and spent more time at it, experience in Meeting for Worship gave me a sense of the Divine Presence in a way which I wouldn't in any way describe as a thunderclap, but which validated my belief or the necessity of feeling that there was more to things than the physical world. There is a phrase "the gathered meeting," when you have a Meeting for Worship which works, after which everybody stands up and looks at each other and says, "Oh, there was something going on there, wasn't there." There is in a very small way a breaking through into the real world of the transcendental, of the ineffable. And that has come to be the sort of spiritual dimension to my life in one sense.

Another thing which I've found entirely satisfying and continually supporting about Quakerism is the pervasive emphasis on "walking the walk" as well as "talking the talk." There is a particular Quaker injunction which says, "Let your lives speak." If you want to understand the Christian message, it's about how to live your life. Watching what people do at a Quaker Meeting is a very useful, practical reminder. Another way of putting it is that, like Judaism, and unlike almost any other Western religion, Quakerism is an orthoprax sect rather than an orthodox sect. It matters what you do. There are virtually no beliefs in common among late twentieth century Quakers, but they all go to Meeting for Worship on Sunday and sit quietly and say what they have to say if they feel bidden to do so. That's what they share.

GS: I was talking to Arno Penzias last week, who said, "When you ask a Jew if he wants to become a Christian, he says, What do I need to do? And when you ask a Christian if he wants to become a Jew, he says, What do I need to believe?"

HT: Exactly. It's interesting that he backed that up. I got the word "orthopraxy" from a close friend of mine who's head of the Computer Science Department at Penn, Mitch Marcus, who's quite a serious practicing Jew, a thoughtful one who has talked a lot with me. Brian Smith and Mitch and I were grad students at the same time, and have been in touch ever since. I don't know whether Brian has ended up on the list or not, there was some question about whether he could make the meeting.

GS: Yes, I actually talked with him just this past weekend.

HT: He and I are old friends. Mitch is somebody who we both talk to a lot about religion. And so that's where I get my information about Judaism. Anyway, it occurs to me that I should have suggested him. Brian may have.

As a friend of mine once quoted somebody, maybe Kant, as saying there are two basic questions: "Why is there something, rather than nothing?" And, "How ought I to live my life?" Quakers typically have little interest, and nothing to say, about the first question, but a great deal to say about the second. The thing which, aside from the disinterest in churches and the existing established religions, most distinguished the founding Quakers of the seventeenth century, was that they believed in the perfectibility of humankind and the possibility of the Kingdom of God on Earth. And they were therefore mixed up at that time with people like the Ranters and the Levellers and the Diggers and were persecuted relatively indiscriminately along with the rest of them. From their perspective this was grossly unfair, as there were some fairly important differences between them. And I suppose one of the things one says about going to Meeting for Worship on Sunday is that one is practicing living the Kingdom. And we refer somewhat grandiosely to the means by which we manage our Church. One of the main aspects of Quakerism is that there is no laity, as we like to say. Outsiders say there's no clergy, but we like to think about it the other way around. What that means is we run the place ourselves and we call that gospel order. Or the way in which we try to run our Meetings qua churches, the institution, goes by that name, although it's being a bit generous in various ways. The standing joke is of course that somebody hears about Quakers and is asked, "Would you be interested in becoming a Quaker?" And they say, "Oh I couldn't possibly. I'm not good enough." And the Quaker response is, "Neither are the rest of us. We're just practicing."

Computer Science and Religion

GS: I've been interviewing both biologists and information scientists, or computer scientists, and the interface between biology and religion raises very natural questions. The line of questioning is pretty clear for me. It's been a bit trickier in the information science realm because a lot of computer scientists are able to carry on their work without really thinking very much about the religious ramifications of it. Possibly there are clear ethical questions. Maybe this is less true for you as someone whose emphasis is orthopractic...

H.T: I wish that were true. But, like the rest of the Quakers, I'm practicing and it doesn't break through very often. But what is also different, to some extent, is that the particular area in which I work raises those questions. I'll take the liberty of actually giving you a reference. It's a long time ago now but I did actually write one scholarly paper about this issue in the Proceedings of the 1985 International Joint Conference on Artificial Intelligence, called "Empowering Automatic Decision Making Systems, General Intelligence, Responsibility and Moral Sensibility." That arose, and I'll talk about it a little bit, because it is a place where the religious sense and the strict technical computer science sides of my work really did come together. It's also possible that Brian has talked to you about the fact that in the late 1970s and early 80s, he and I and a bunch of other people were involved in something called Computer Professionals for Social Responsibility. It largely arose out of concern about the dominance of military funding for computer science research in the U.S., and subsequently came to focus first on issues to do with launch on warning, which was especially in the European context where I carried this on and helped start a similar group in the U.K. when I moved here in 1980. Launch on warning was a particularly poignant topic here because there was a political dynamic which said, "We have missiles in the Eastern parts of Western Europe. The Soviets have missiles in the Western parts of Eastern Europe. Their flight times are such that their locations are only 15 minutes apart. What that means is there is no possible human mechanism which can be given the responsibility for getting those missiles off the launch pads in time to avoid destruction in case of a Soviet attack. It therefore follows that we must put those missiles under automatic controls to be a credible deterrent." And the political dynamic mandated what the technical necessity was, despite the fact that no responsible computer scientist anywhere would say, "Oh yeah, we can do that." And so we were originally involved in organizing a technical response to that, basically saying, "You can't do that. It's fundamentally irresponsible to put fallible computer systems in control of nuclear weapons. It is essentially to commit suicide."

Not long after that came Star Wars. I actually ended up on breakfast television here for about two minutes as part of the organization that we did to get computer scientists in this country to say that they did not want any part of the proffered U.S. funding for S.D.I. research in this country. And that was reasonably successful. We managed to get a pretty unanimous declaration of disinclination on the part of the British technical community to have anything to do with that. What that all led up to was a general concern with the moral dimension of empowering computational systems to act in the human realm, and it's an issue which remains almost completely ignored, but which seems to me to be of profound moral and practical significance. And the particular issue which I addressed in that paper was the circumstances which would have to surround any rational empowerment of an artificially intelligent computational system in any significant human context. The context I used as a thought experiment was the judicial one, since that is one that science fiction has sometimes discussed along the lines of: "Human judges are well known to be fallible. In the millennia future, we will have computational judges who aren't tired and don't get bored and who will render fair and just judicial verdicts utterly reliably." And leaving aside all the immediate questions about the fact that artificial intelligence just isn't up to that, and won't be for the foreseeable future, there was a moral dimension to this question, which I thought hadn't been discussed. So I thought about what it meant to submit oneself to judgment, and into people's hands, in circumstances where you have to give them responsibility for a significant aspect of your life, whether it's literally your life as in the case of doctors, or your education, or your children's education, or indeed your handling in the legal process. That there was never any discussion about giving these computational judges any spiritual life, actually seemed to me to be quite important. It seemed to me that submitting to judgment depended, in no small part, on my recognition that the judge took responsibility for his actions. And responsibility is fundamentally a moral category. It's not something which you typically think of as a sense that an autopilot has. Or at least not self-conscious responsibility. So it's not just "imputed" responsibility. We can say, "Well it was the failure of this cable which was responsible for those people's deaths." But you wouldn't suppose that the chairlift was conscious of its responsibility for the integrity of its cable. Whereas, a teacher's consciousness of their responsibility for the children that they teach, or a judge's conscious recognition of his or her responsibility for the judgments that they render and for the people that they render them on, are absolutely constitutive of those roles. If that's true, then saying something about the origins of that self-conscious responsibility is eventually going to be necessary. And at that point, if not before, a thoroughgoing understanding of the grounding of our moral behavior in our religious sensibilities becomes necessary.

The Origins of Moral Responsibility

GS: If, even in a religious life, it is really most important to behave properly, not to believe particular things, then isn't sufficient for a computer judge just to behave properly, and not to have to believe it is responsible for those judgments? Is there a necessary link between believing I have a moral responsibility and actually doing the right thing or being a good judge?

HT: That's a very good question. I hadn't thought about that before. Let me start with the usual anecdote, which is the famous, I think the original form of the joke is, "How do you paint perfectly?" "That's easy, just become perfect and paint naturally." You're right to put me on the spot on this, because of course, one of the points that George Fox, one of the principle articulate founders of Quakerism, based his belief in the possibility of living the Kingdom, was "Be ye therefore perfect." Personally I can't do the Scriptural quotation business, that's not my strong point, but that's the text he started from. Now the fact of the matter is that most of us aren't perfect. It's only the holy fools who can act in the right way without thinking about it first. And especially for hyper self-conscious intellectuals such as myself, I have to think about what I do even if I wouldn't directly start out by saying, "Here are ten propositions which I acknowledge the truth of and my actions follow from them." That's not the way one tends to do it. One tends to think in terms of role models, in terms of precedent. That's a dry word for something that's more active than that. And in terms of self-consciously practicing moral behavior so that you will do the right thing when the time arises, when the circumstances arise. The slightly jocular way of putting it is, "If you have to think about it before doing something, does it count?" If you have to weigh it up, if you have to tell yourself, "Now the right thing to do here is to give the money back," or to say that you'll clean up the garbage, or whatever it is because, "I know that's the right thing to do so I really have to do it, right." If you have to talk yourself into doing the right thing, can you take any credit for it? Well, of course you can. But asking that question reveals a naive wish that one could just do the right thing naturally. But given that, historically, certainly in 1985 when that paper was written and the years leading up to that, no-one conceived --- ("no-one" is too strong) the mainstream view of artificial constructs was of constructs whose behavior was designed into them. So if you didn't design into this construct a moral sensibility, it wouldn't have one. Now, in the twelve years since then, the debate has become much more evenly balanced between two perspectives, which if I can put them simplistically are: on the one hand, we will build out-of-the-box intelligent artifacts that will spring forth fully formed from their creator's brow (this was by far the dominant orthodoxy in the late seventies and early eighties). And on the other hand, that's too hard: We will build the computational equivalent of a child and let it learn to be an adult, which is easier, because babies are much simpler behavers than adults. That's pretty hard too, because we don't know very much, still, about how babies get to be adults in terms of their intelligence.

GS: That transformation itself is probably one of the most complex things that a human will ever do, moving from the infant into the adult.

HT: Absolutely. So it begs an awful lot of questions to say, "We will create computationally sophisticated, but behaviorally trivial artifacts and let them learn to be judges or doctors or whatever else it might be." There are very simple mechanisms which achieve that, namely that seven or eight pounds of protoplasm which starts out life as a newborn baby. But the complexity of that system is already well beyond anything that we understand in any interesting way. So it seems to me that that debate about the origins of moral sensibility and computational artifacts is one which really hasn't happened. It's one that, it still seems to me, ought to happen and in due course will have to happen. It has the potential for informing us, in the way that the best kinds of intellectual inquiries do. That is it will be good not only for the computer scientists and the AI people, but also for the theologians. If I had to say that the theology which I find comes closest, mainstream theology that I'm aware of, and I'm a long way from being well read in modern theology; the theology that comes closest to lining up with both this inchoate perspective from the AI and the Quaker perspective, is by Alasdair MacIntyre, entitled After Virtue. And that book, which is all I've read of his, is a long way from being, in the terms of the science, "implementable." One of its virtues is that it is pretty concrete. It does touch down in this notion of walking the walk as well as talking the talk in some fairly obvious ways. But it's a long, long way from being something which a computer scientist with a behaving artifact could say, "Oh Yeah, I could take that and I can work with that. I can start thinking about adding some virtue into my robot."

Morality, Virtue and the Spiritual Life

GS: I haven't read MacIntyre, but I'm wondering now what the difference would be between programming virtue into a computer would be and giving it a spirit. Somehow it's easier for me to imagine a virtuous machine than it is a spiritual machine. I can see awarding a medal of honor to a machine but not wanting to Baptize it.

HT: Well that's an interesting thought. I really think that's right. One obvious reason, although I don't think it's the interesting one, why that's true is there are lots of different spiritual bases which evidently induce virtuous behavior. You have lots of different existence proofs. Across a fairly broad spectrum of religious traditions the consensus about what constitutes a virtuous life is not perfect by any means, but it's much better than chance. Not surprisingly, there are internet newsgroups and mailing lists for Quakers as well as for every other sect and religion under the sun. And somebody happened to post there yesterday or the day before, "When my kids were born, I felt like I needed to get back into church because I felt there was no possibility of them growing up with a reasonable moral sense unless we were going to some kind of church regularly." That's how he eventually became a Quaker. But he went on to say "Since then I've realized that there are plenty of people out there who are leading eminently moral lives without any evident religious sensibilities." There certainly are. So a range of religious backgrounds, or none, seems to be a possible basis for virtuous behavior. I suppose I'm enough of a believer in the value of spiritual life, that I'm going to commit the Catholic sin, and to some extent it's a Quaker sin too, of saying they really are believers even though they don't think they are. Both the Catholics and the early Quakers tended to do this with respect to other religions and say, "They're worshiping the same God we are. They just don't know it." And/or, "They won't acknowledge it, but they really are." This tends towards a pretty unpleasant sort of triumphalism, but in a more responsible way is a fairly reasonable basis for what in Quaker circles is known as Quaker Universalism. Which says, "Quakerism started out as Christianity but for me, I'm a Quaker Buddhist, thank you very much." Different divisions within international Quakerism take more or less kindly to that perspective, but there are certainly plenty that do take kindly to it.

Coming back to your question, it seems to me that in some sense, to solve the problem that you posed, namely isn't it easier to program in virtue than it is to program in spirituality. It's not that understanding how to provide a basis for virtuous behavior will necessarily involve an understanding of spirituality, but that's certainly a possible, indeed the most likely, source. But I'm just speculating. I'm a long way from understanding how to make that connection, how to run it in either direction. I guess it comes down to a belief that the virtuous reflex, the ability to do the right thing, to engage in right livelihood, must spring from some principled basis. And I'm skeptical about the ability of people in general, or scientists in particular, to succeed in the longstanding humanist project of deriving moral behavior from selfish first principles. Without the acknowledgement of something that's bigger than I am, I don't think I could do it. I've never observed any attempt to do it be even remotely convincing. [I just found these words in an article about Arthur Koestler by Bernard Avishai in The New Yorker of 6 January 1997: "[W]e cannot be relied on to do what is right if all we have to work with is our pain or history or interest."]

GS: On the other hand, the ineffability of that "bigger than I am," or the source for virtuous behavior poses a real problem for people trying to recreate it in any way.

HT: It sure does. One of the things I would hope to get out of these discussions in Berkeley in June, if nothing else, is to actually get some contact with some people working this problem from the other end. Since I moved over here and stopped having the occasional conversation with Brian's father, Wilfred Cantwell Smith, who's a comparative religionist and theologian in his own right, I haven't had much conversation about this side of things, with anybody. As I said, I wrote this paper over twelve years ago. And I haven't pursued it. But this invitation has made me start thinking about it again.

GS: I often feel sorry that the biologists and the computer scientists aren't getting together as well because some of the biologists are working at this from the third end, which is trying to look at just what, from an organism point of view, constitutes the spiritual life of a human being.

HT: I find that that boggles my mind, but then I'm sure that the converse is probably true as well. This I suspect is a direction which your discussions with the other computer scientists has not gone. Are there directions that they have taken that it would be useful for me to respond to?

Mechanical Metaphors for Mind

GS: Yeah, I think so. There are a lot of metaphors for the human mind that are borrowed from computational science now. And I'm wondering how, leaving spirit aside for a moment, how powerful a metaphor or a model for human cognition, a computer is. And then, I'd like you to discuss the possibility of spirit and the less tangible aspects of mind being an epiphenomenon of creating some kind of a digital mind-like machine. I'm interested too in what the religious ramifications are of mind or spirit being essentially emergent properties.

HT: Another question that might follow along the same line is: if we did have an artificial intelligence, you're a Quaker, you believe in the essential importance of human life, your pacifism stems from that, would it then be a crime equivalent to murder to throw the "off" switch? Which is to romanticize the question; but it is a question which definitely arises. Going back up and answering your first question, or addressing it, I certainly can't answer it.

Someone, and I've got no idea where this observation came from, but as far as I can see with my limited knowledge of the history of science and philosophy, it's true, observed that our dominant metaphor for the brain in any given era is the most complex technical artifact available. And so the ancient Greek model was plumbing, waterworks. And in the late nineteenth century it was the London Underground. I've actually seen a wonderful illustration from an early twentieth century/late nineteenth century encyclopedia, in which the mind is depicted as a very complex railway network. And then it was the electrical grid; and now it's computers. And, it's sort of obvious why that might be true.

There is, it seems to me, a qualitative difference now. I'm sure you must have talked about this at length with Brian Smith, because this is his basic interest in life, namely that the claim now is not as it was 2,000 years ago that the mind was like a waterworks system, or was like a subway system, or was like the electricity grid, but the mind is a computer. There is a world of significance in the difference between that "like" and that "is."

Just what is special about sophisticated brains, and where in the phylogenetic hierarchy they become sophisticated enough to have these properties, are things I'm not qualified to talk about. But at least one level of sophistication is that they are meaning-bearing artifacts, that they are systems or complex behavioral structures, physical and behavioral structures, which are about things, which mean things, which represent the world, and whose behavior is influenced by those representations. So there is this two-way causal connection between the world and our representations of it, and our representations and actions in the world. And there is a wealth of complex philosophy, some of it due to John Searle, in Berkeley, which addresses that very complex interaction. At that point, the question becomes, If you had a constructed artifact, as opposed to a biological artifact, which had both those causal connections, that is, where the world influenced the representation in the way that our perceptual interaction with the world does, and those representations influenced the behavior of the artifact in the world, in the same way that our beliefs and desires influence our behavior, then would you have something which was in any interesting way, different from us? And the so-called "strong AI position" that as far as the things that AI and cognitive psychology care about, the answer has to be no. That you've done it if you've done that. But the question of whether that would be sufficient for: a) self consciousness, and b) following on from self consciousness, necessarily, a spiritual life, is something which people don't usually talk about. The first term is something that's discussed. There's a tremendous toing and froing about whether self consciousness is an intrinsic or epiphenomenal result of sufficient semantic and computational complexity on the part of any artifact, whether it's biological or mechanical, or whether self consciousness is something we need to understand and build into our systems just like we build in visual processing, auditory processing, and motor control and all the other things we'll need in due course. I don't have an opinion about that. I find that debate extremely difficult and one which I tend to agree with whoever I've heard most recently, which suggests that we're not getting there. There's not a very clear understanding in that whole area. And, unfortunately, it does seem to me that that question is a necessary prior question to the question about spirituality.

I guess I don't believe that an organism, whether biological or mechanical, that wasn't self conscious, could have a spiritual life. And that ties back to the questions that we were discussing earlier. Because it's related to this business of good and evil, and sin, and other hot-button words. Notice that one typically doesn't say "That tree will burn in Hell" for having fallen down and crushed that woman in her car yesterday. Certainly it will not. Nor, would we say, "That bacterium will burn in Hell for having infected my daughter and killed her." Gosh, here comes the other hot button word that we've managed to avoid so far. There is, inescapably, tied up with notions of good and evil, and therefore with any notion of virtue that makes any sense, the notion of free will. And if you're not making choices, then you aren't characterizable as being good or bad, it seems to me. If you're making choices, then you're self aware. So there's a sort of circularity here, with all these things chasing each other's tails. And it doesn't seem obvious to me that merely creating semantically robust artifacts, ones with real semanticity as Brian Smith would say, is going to be sufficient to create either self conscious beings or moral beings. But I haven't thought about that one much.

Spirtual Life, Grace and the Inward Light

GS: I wonder too, if there isn't a way in which spirituality could be said not to be reliant on self consciousness, but rather some other kind of connection to the noumenal world.

HT: Certainly. Early Quakers used to refer to the "inward light." Modern Quakers have sort of lost their understanding of this word by and large, and I think most Catholics, for instance, too. Nowadays people talk about it as the inner light and seem to think it's something that they've got, inevitably and inescapably. Whereas the inward light is much closer to what a Catholic would call grace. And it's something you are given. It doesn't come for free. It doesn't come with the equipment as it were, it comes by the grace of God. You get that noumenal access as a gift. And it therefore follows if you take all that language literally, that it would be up to God whether this particular constructed artifact had that access or not, had that awareness, had that spiritual life. I don't find that language congenial in the sense that one of the ways in which I do part company with my wife and with even fairly thoughtful Catholic doctrine is on the whole business of providence, God's ability to act in the world. The phrase that many Quakers use, although I don't think it's a Quaker phrase originally, is "No Hands" you know "My Hands are Thy Hands." or "No Hands but Mine." That there is no way for God to act in the world except via human beings seeking to do "His Will," in inverted commas. That is to say that physical intervention in the world, in any sort of direct sense, is not part of my conception of the spiritual dimension of things.

Free Will and the Nature of Decisions

GS: Well what about the will in inverted commas? This is something that I'm obliged to touch on in these interviews. It's a big problem obviously for a lot of biologists, especially evolutionary biologists, to look at the world in terms of anyone's will, or teleologically. I wonder how you, whether you do juggle these two different interpretations of how things operate in the world: On the one hand, looking at things purely mechanically without any intention, and on the other hand, looking for God's will and ways that you can manifest it.

HT: It's very tricky, that one, isn't it? The field is littered with, in my opinion, corpses of people who have attempted to derive free will from some computational or physical properties. It's something which tends to be, it seems to be a disease which infects people at a certain stage in their lives. Roger Penrose is the most recent famous example of someone from outside the field who basically said, "Oh yes, we've got it. We can derive free will from quantum indeterminacy via computational modeling," which just made all the people in this building, and anywhere within a hundred meters of any center of reasonable expertise in the area, fall about laughing.

GS: It's a sort of a God in the Gaps argument.

HT: It's a God in the Gaps. That's right.

On the other hand, I feel like it's an area where my experience of computational systems leaves me even less clear about what free will might be than the naive lay person. I sometimes occupy myself with this, it has to be said, sort of casually because I never can get anywhere with it. I actually once tried to quietly understand what constitutes "deciding" for an individual. If there is an external procedure by which it's agreed decisions will be made, whether it's a vote or a consensus, or the sense of the Meeting as we practice it in Quaker business meetings, that's not what I mean. What I mean is what is called "making up one's mind." It's one of those things that when you push hard on it, it just runs away from you. If you turn to computational systems and you say, "Well okay, you build behaving artifacts, you build things which people interact with, and are prepared to impute rationality to, whether they are as simple as chess playing robots, or as complicated as English language interaction-based help systems for Microsoft Word, shall we say. People ask such systems questions or pose problems for those systems and they give answers, or they exhibit behavior and those behaviors represent choices." And you can say, "Yes, but I can look at it and I can tell you in no uncertain terms exactly where those choices came from." And even if, at the end of the day, the computation that was supposed to lead to an answer, came up with two alternatives with different expectations or different certainty values, and flipped a weighted coin in order to decide between them, well that's still not a very satisfactory notion. It's not a useful starting point for an understanding of free will or what's going on when you have an internal monologue of the form that we mentioned a while ago, of, "Should I pick up that money and give it to that woman who's just dropped it? Or will I keep it for myself?" At some point, you either run after her or you don't. But I find I have no language for talking about what happens at the cusp of that decision. And I personally haven't found that knowing what the closest analogs to that in the computational context, are any help in understanding it. But maybe if some people sat around and talked about that.... The thing that one says about cognitive science, which is the sort of human-oriented end of what we do, is why isn't it a prodigious arrogance to suppose that you, relatively ordinary folk, are going to have any more luck at cracking these problems than the greatest minds of the successive ages between, let's say, the early Greek Philosophers and today have had? Because there's been a singular lack of success for well over two thousand years. And all these fundamental questions, both of understanding the human mind, and of understanding human spirituality remain. Why should you guys be any different? Why do you act as if the world has changed? And the answer is, well there's a little bit of "We're standing on the shoulders of people who've gone before." And a little bit has been learned over time and we can do a little better. But there is also the belief that we have a qualitatively different intellectual tool in the form of that computational metaphor, and its realization in real physical silicon and metal and plastic. The ability to not only use that computational metaphor to assist our discussion, but to construct computational artifacts whose behavior we can then tear apart right down to the lowest level if we need to, in a way that in the relatively civilized late twentieth century we, for good reasons, can't do with human beings, does make a potential difference. I think the utility of that metaphor, and its attendant artifacts, has been much better explored in the semantical, or mental and psychological domain than it has been in the spiritual or theological domain. And maybe there's some leverage there.

The way it's worked in this domain is that in the last twenty years we've gotten people whose fundamental training is in psychology or linguistics or other areas of the human sciences who have gained profound computational sophistication and then been able to leverage off of that conjunction. To a lesser extent, you get people coming from the other direction. They are more often likely to look somewhat foolish when they try. But sometimes that happens. I'm not aware of many examples at all of that happening on the spiritual or theological side.

God's Will and the Kingdom of Heaven

GS: Fascinating. Part of it, and I guess we touched on this before, may be that it's so difficult to say what spiritual life looks like. It's not what ethical life looks like; that's relatively easy to describe. But spiritual life is basically invisible.

HT: And the people who are inclined to model it are going to have such idiosyncratic takes on it, right? I would be very surprised if we don't find when we sit down in June, that the range of spiritual perspectives is going to be very dramatic. My experience is that the intellectuals, and by and large people who are going to be at something like this are going to be intellectuals, tend more toward the--this is a line from Islam via Cantwell Smith--"I know that God is greater than I know Him to be." The really extreme to mystic perspective, that there is nothing that you can say about God that's true. Because you're just not going to get there. Any attempt to frame a proposition is by definition inadequate and misleading. That doesn't leave you much room for scientific inquiry.

I'd like to know more about Thomas Aquinas. He seemed to me the most intellectually rigorous theologian who didn't take the man with the white beard starting point, that I'm aware of, but I've never really had the time to study Thomas.

GS: What about God's will on a day to day basis? Is that a part of your world?

HT: Absolutely. It's crucial. It's because of the Quaker business meeting, the way we run our lives, the way we run our meetings as institutions. We say very explicitly, we worship together with a concern for business. We take the procedure of the Friend's Meeting for Worship as the mechanism by which we do business. And what that means is that in serious issues, we are seeking God's will for the Meeting. We are not trying to achieve consensus. Consensus is a secular process. The Meeting for Worship for business, when it works, and it's under threat in various ways, which would take us too long to discuss here, but when it works, it is because everybody engaged in it, or a sufficient number of the people who are engaged in it, see what they are doing there as seeking God's will for that Meeting. Now people mean different things by that, and somebody put it not very long ago, when what you are talking about is whether to have chairs or benches in the meeting room, it is unlikely that God has an opinion. The best you can hope for is something that will be pleasing in the sight of God. And all of this is of course code, because we are stuck using these words which project a humanoid perspective onto the referent of that word, that G word. But what doing God's will means for me, and I think for many more intellectual Quakers, and those of other religions, too, I'm sure, means acting in a way which is consistent with bringing about the Kingdom of God on Earth. There is this incredible quotation, from George Fox, and it depends on in some sense on what day of the week it is and how my breakfast went and what I'm feeling, whether I take that literally, whether I actually go along with the original Quaker perspective, which says that the Second Coming has already happened: "Christ has come to teach his people himself." You can, in the gathered Meeting, or even on your own, receive Christ's guidance. He promised it; it says so right there in Acts. The Paraclete will come. I will be there. And all that is very much God talk, very much Christ talk, which in one way I find totally off-putting and unhelpful. On the other hand, maybe it's really real. Maybe what all that is talking about is that the ineffable, the noumenal, the abstract God ground is accessible to all of us and if we get around the various problems of being misled and misleading oneself, that we have access to that, that it is universal.

There is an answer to the question, "Of these behaviors, which is the one which is most consonant with what that noumenal essence would have. What is consistent with that? What would it mean to have the Kingdom of God on Earth?" And doing God's will is acting in as consonant a way with the Kingdom of God on Earth as we can, where the Kingdom of God on Earth means, in non-God talk, living the life we are meant to live, the life that is the best expression of our humanity.

At the end of the day, that's the answer for me, and for a lot of Quakers, I think, to the question, "How ought I to live my life?"

GS: How does that notion of consonance with God's will, or consonance with the project of bringing about the Kingdom of Heaven on Earth, jive with the idea that's so popular among evolutionary biologists these days that life is pretty much the product of repeated application of algorithms, that we're not moving from one place toward another, that Heaven on Earth is something that will be defined by the course of evolution, if it's ever defined at all.

HT: You know, that's a good point. First of all, as a linguist, if nothing else, I got the teleology beaten out of me long ago. One of the things we've learned in the twentieth century about human languages is that they change and they evolve but they aren't going anywhere. That's not an opinion. It's a fact. Languages move around. English has gone from one end to the other of various possible spectra, in terms of things for instance whether the verb comes in the middle of the sentence or at the end, to take something trivial. We've changed our minds about that three or four times in the recorded history of English and its predecessors. So there isn't any right answer. It's not going to be the case that five thousand years from now all languages will be verb final, or all languages will be verb middle. Languages aren't going anywhere, they're just going. I have no difficulty with the notion that there are lots of evolutionary systems in various senses of the word systems for which that's the case. It seems to me that the progress toward the Kingdom is orthogonal to all that. This is something that the professor of New Testament theology who happens to be an attender at the Quaker Meeting here, said at a little series that we did on the New Testament recently at Meeting. It's a bit from the Kabbalah or Hasidic Judaism. There are, it turns out, seven hundred and some odd clauses in the law. It comes back to orthopraxy, that what a good Jew tries to do, is observe the law. Well there are seven hundred and how many clauses in the law once you get them all down, once you work your way through the Pentateuch and you get it all sorted out. If everyone on Earth obeyed all those laws for one day, that would bring in the Kingdom of Heaven. That's all it would take; it's simple. It just takes all of us to obey the whole law for one day and we've done it, right. There would be no going back after that. You can think about that as almost like the business that it is within the bounds of possibility that all of the air in this room will coalesce in the upper left hand corner of the room for long enough for me to choke. You can measure the probability that that will happen. It just needs all the right things to come together at the right time. And so in some sense, there's no progress toward that event; it just may happen one day. In a very detached perspective about the Jewish Law, you could take the same perspective: it just may happen one day, that we'll all manage it.

GS: But is that law like languages in that it, also, may not go anywhere, but it also changes?

HT: That's right. At the end of a certain number of hundreds of years of work, there was agreement about what those 735 clauses were, and then it was okay. Then we knew what the task was from there on out. No, you're absolutely right, for the rest of us it isn't like that. But I guess I do think that it is each individual's project to understand for themselves as best they can, what living the Kingdom means, and to try to do that. Yet it is a matter of faith, and not, by any means, proof, that if we all did that and did it well, we would agree. This comes back to this business of there being only one God, whatever you call it. That if there is a transcendant, it's the transcendant. Our human access to it is inevitably limited. But we do have access to it; that's the article of faith, I suppose. What's interesting about the different religions, and the differences between them, and why Quakerism is particularly interesting in one small respect, is their approaches to the question of, "How do we know whether the bit of the transcendant you claim to have gotten hold of, is accurate." That is, the question of authority. That's one of the respects that the Catholics and the Quakers are much more alike than either of them are like many of the other Protestant sects. That is for neither the Catholics, nor the Quakers, is the answer to that, "Oh, it's in the Bible." For both of them, more or less depending on exactly what kind of Catholic you are, the answer is, "In the tradition of the Church." We work it out together. The Pope has a particular role in that. There is no equivalent role to the Pope's in Quakerism, but there is a surprising amount of consensus about the fact that it is the consent of the faithful, in the Catholic phrase, that determines what the fact of the matter is about what the transcendent is telling us. What underlies that is a belief that there is a consistent story, and we can get access to it. Imperfectly. From time to time. That's as far as I get.

GS: And that's as long as we've got. Thank you very much. I've enjoyed this tremendously.

HT: So have I. You're welcome.