We've been traveling this week. First, a night and parts of two days visiting Jo's mother in Brunswick, Maine, followed by a night in St. Andrews, New Brunswick, and then a couple of hours ago we arrived at Brackley Beach, Prince Edward Island where we hope to remain for a few days if we can find a comfortable place to stay. We have a third-floor room in Shaw's Hotel for tonight, but the hotel is booked for the weekend and we'd really like to settle down for a few days and relax.

Last year around this same time we visited P.E.I. and Shaw's hotel for the first time. We had a very pleasant visit and decided to try to recreate it again this year - usually a bad idea but this time it seems to be working out. I found a record of that earlier trip on my laptop this afternoon. Last year, I used the laptop to keep a trip log and store the images that we took with our digital camera. On the long stretches of the drive, I would turn the notes and images into vacation web page that I downloaded to our web site and then sent email to family and friends enclosing the URL of the web page as an electronic postcard of sorts.

Today, as we rode up the southern coast of New Brunswick along the Bay of Fundy, I went over about half the entries for this journal adding HTML tags to turn the entries from simple text documents into hypertext documents for an on-line version of the journal. I doubt I'll take the time to produce a record of this year's trip other than adding a few trip-related comments in the entries of this journal.

As we drove through New Brunswick, we listened to the books-on-tape version of a collection of short pieces by Tom Wolfe (titled "Hooking Up" [Wolfe, 2000]) which included a characterization of the life and times of Robert Noyce (computer industry pioneer, Intel co-founder and co-inventor of the integrated circuit along with Jack Kilby of Texas Instruments) and another piece - possibly it was several pieces whose introductions and endings I missed and so strung them together to form a single thread - that discussed the ideas of Marshall McLuhan, Pierre Teilhard de Chardin, Richard Dawkins, Edward O. Wilson and several other writers and scientists. In the midst of this latter piece or concatenation of pieces, Wolfe claimed that the web and the internet constituted just another mode of communication and expressed skepticism that it would somehow transform or lift humans and human society to a new level as some evangelists and members of the digerati (the digital literati) have prophesied.

Wolfe believes that E.O. Wilson's writings on sociobiology provide a convincing argument that humans are incapable of radical deviations from their fixed patterns of behavior and that we and the society in which we live are irretrievably a product of our genes. Wolfe is skeptical of theories for accelerated transformation of society such as those based on Richard Dawkins's "memes" - ideas (software) that infect minds like catchy advertising jingles, spread through the population like viruses and reproduce like genes in accord with Darwinian evolution. Until we understand real genes and in particular how our genes influence the wiring of our brains, we simply won't be able to reprogram ourselves to any significant degree.

The crux of Wolfe's argument as I understand it is as follows. Consider the genes as the machine code for powerful firmware that directs much of our behavior including the most primitive fight-or-flight responses and those responses governing social interaction. For a meme to effectively propagate, it has to find a host willing to entertain the meme and possessed of suitably powerful software to process the meme and counteract any contradictory genetically-mediated influences. This combination powerful memes and extensive preparation sounds a lot like the basis for modern education which some would say has not proved terribly effective at countering our deeply ingrained gene-mediated behaviors. Moreover it seems unlikely that educating people will be made significantly easier by the advent of the world wide web - propagating information may be faster and easier, but learning difficult concepts will continue to require patience, persistence and intellectual effort. By the way, it's certainly possible to believe in Dawkins's theory of memes without believing that memes can single-handedly bring about a radical transformation of society.

As I was listening to this - just about the time we missed the turn for the bypass that should have taken us around Charlottetown thus avoiding a considerable amount of traffic, I found myself becoming annoyed with Wolfe and hostile to what he was saying. I blamed my annoyance on Wolfe's glib dismissal of ideas without what I considered a careful analysis or at least an analysis that he was willing to share with the reader. But then I realized that on the basis of my own reading of Wilson, Dawkins and McLuhan, I agreed with Wolfe and by doing so I was toying with abandoning certain utopian fantasies that I entertained when I was younger and was now loathe to finally and definitively bury.

I was also confusing the prospects for transforming the individual with the prospects for transforming the population of the entire global village to use McLuhan's term (closely related to Teilhard de Chardin's "noosphere", the converging "cybersociety" and other utopian visions of a transformed and enlightened global society). I want to believe that I can transform or reprogram myself. I think of my mental state in terms of software and my reading, studying and thinking as a form of self programming. And while I find myself agreeing with Wolfe and scoffing at the utopian idealists much as a physicist scoffs at amateur scientists claiming to have invented perpetual motion machines, I cling to the idea that I can change, that, as a self-taught student of the mind, I can locally reverse the cognitive analog of entropy without violating the corresponding laws thermodynamics.

I say this without a great deal of conviction just now, having recently experienced situations in which, despite extensive preparation on my part, I found myself helplessly deviating from my carefully planned behavior and reverting to primitive, deeply wired responses that I could not resist. Still, the very idea of a mind at war with itself and capable of recognizing and mounting defenses to thwart its own built-in biases and wired responses gives me hope. I suppose there are people who have no wish to change the way they are; however, for many, I expect that it's important to believe that change, positive change, is possible. As a computer scientist aware of the power of programmable machines, it's natural to want to tinker with your own programming.

The very idea of programming yourself may seem a little bizarre to some of you. I offer the following brief account of what I mean by self programming and what I think of as a natural extension of life writing or self-reflective journal keeping.

Why would programming your own responses be any different from programming a robot with a fixed set of sensors and manipulators and an existing layer of software providing basic services and low-level drivers for all the robot's sensors and effectors? Well, aside from the fact that we don't have access to the source code for the brain and have no idea of the APIs for the low-level software (much less the high-level software), it seems that low-level driver code is fundamentally different from a software perspective than the "subroutines" and basic firmware that governs our behavior. For one thing, our firmware appears to be constantly adapting; every time you think about something you change the way you think about it, every time you do something you change the way you do it and your motivations and predelictions for doing it in the future.

Imagine writing code to get a robot to walk across the room, code that relies on a particular set of low-level driver routines. Suppose you finally get the robot to perform as you wanted, but the next morning when try your code again you find out that the drivers have perversely rewritten themselves so that, instead of walking across the room, your code causes the robot to plug itself into the battery charger and go into sleep mode. People talk about being compelled to do things, about being wired to respond one way rather than another, about being unable to overcome their biases and predispositions.

So what possible mechanism might we employ to modify our biases and predispositions. I realize that this discussion is right on the edge of some sticky philosophical issues which I really don't want to discuss but it might be a good idea to expose some of my biases so you won't have to guess them. We'll start with a biggy. I don't believe in free will - I'm not even sure that the notion makes sense but to the extent that I can make sense of it, I don't believe in it. I believe that I'm a very complex machine, so complex indeed that no human including me can accurately predict what I'll do next. But I can imagine a more complex machine that would find my behavior so simple and predictable as to be uninteresting. I even believe that it's only a matter of time before such machines exist if they don't already somewhere in the universe. (I guess this also answers the question of whether or not I believe "artificial intelligences" are possible.

If free will means that I could do something (usually the line is "I could choose to do something") that this imagined complex machine could not predict, then I certainly believe that I don't have free will. I'm programmed to behave in a particular way. If, as I believe, I can reprogram myself, that capability doesn't change anything. Computer programs that rewrite themselves, that are subject to and even actively subject themselves (or versions of themselves) to simulated natural selection are common. Programs that react to, adapt themselves to and draw inspiration for their adaptations from their environment are similarly common. I believe that more of the same (self replicating, adapting, evolving) coupled with machines of exponentially increasing power and ability to sense their environment is all that it will take to eclipse our role as the most intelligent life form on this planet.

So with that introduction - we're nothing more than programmable machines - what mechanism might we employ to program ourselves? Well, we are just programmable machines but some of our programming is pretty powerful. We have the ability to process language, use logic to follow and critique arguments, construct models of our environment, hypothesize, evaluate and predict outcomes, absorb knowledge provided in books and by hearsay, form plans, and formulate and perform experiments to test hypotheses. I admit that it's quite extraordinary that we came to have these capabilities but have them we do and some of us exercise them every day.

So, it's simple, right? We just list and evaluate a set of possible outcomes, evaluate them to determine the ones we'd most like to come to pass, and then formulate and carry out plans to realize those outcomes. I can and have written code to enable robots to carry out a sequence of steps very much like this, and the fact that I can articulate the steps means that I could probably carry them out myself with one caveat: I may not feel like it! Does this mean I'm stuck, unable to "overcome" my programming and my hardwired biases and predilections? No. It just means that some of my programming is of a sort that requires a very different sort of reprogramming.

You've probably heard about B.F. Skinner's notion of operant conditioning. The idea is that by repeated exposure to a stimulus coupled with an appropriate (positive) reinforcing signal an organism can be trained to produce a response to a stimulus even in the absence of the reinforcing signal. Complex responses (behaviors) can be "shaped" by reinforcing their component parts (also important in reinforcement learning). Skinner got a bad rap since, as is often the case for a scientist single-mindedly pursuing a theory, he tended see operant conditioning and behaviors determined by simple stimulus-response associations at work in every aspect of human behavior. But Skinner had one thing right - operant conditioning is an important factor in explaining human behavior and a crucial tool in the repertoire of a self programmer.

Suppose that I've had a bad experience in school and so I'm reluctant to attend class. But logic and a quick look around at people whose lifestyles I find appealing tells me that getting a good education is a ticket to realizing the outcomes I most want. But I can't seem to overcome my aversion to being in school. So I engineer situations such that the stimulus, being at school, is associated with a positive experience, such as taking a class I really like or working with a supportive teacher, and eventually my reluctance and aversion are extinguished and my interest and attraction become the dominant responses. Believe it or not you can even do this in your head by simulating the stimuli and reinforcement signals as it were; you simply imagine being at school and having a good time or learning amazingly useful stuff. It sounds too good to be true but it works and most of those "power of positive thinking" seminars you read about in the back of airline magazines are based on this simple idea.

So I can use self-inflicted operant conditioning to rewire parts of my firmware if I can recognize what it is that I want to change and devise an effective method of conditioning to instill the target stimulus-response behavior. You need to know what makes you tick and hence there are probably limits to how much you can alter your behavior but it's possible to make significant alterations. With a little insight, you could probably condition those perverse adaptive motor drivers that I mentioned earlier to get your robot to walk across the room reliably. So, for me, self programming consists of a combination of searching for and recognizing desirable outcomes, when appropriate using operant conditioning to adjust my desires to bring them in line with the analyzed results of experiments designed to test relevant hypotheses, and then means-ends analysis to figure out how to bring about the desirable outcomes. Sounds so simple doesn't it? Well, it isn't but then neither is programming a silicon-based machine to do anything halfway interesting.

I should note that you can act randomly simply by flipping a coin, consulting an astrologer or otherwise selecting how to act on the basis of some random or pseudo-random oracle. (A sequence of numbers (or bits or boolean values) is said to be random if it is unpredictable; computer programs called "random number generators" provide sequences of so-called pseudo-random numbers since, while it is very hard to predict the next number in a sequence of pseudo-random numbers, they are obviously predictable if you know the algorithm used to generate the sequences.) If indeed you have access to a sequence of random numbers then by definition your behavior is unpredictable. Does this mean that no machine could ever predict your behavior? Well, yes, but then it doesn't seem that you are in charge of your behavior either, and, in any case, your behavior is "predictable in expectation", i.e., a suitably powerful machine can predict your most likely behavior which is all that can be expected in the presence of true randomness.

Think it odd for an engineer, scientist or, specifically, a computer scientist to fly off the handle and discourse on themes usually the province of philosophy or theology? Not so. Scientists, mathematicians and academics of all stripes appear to have rather lose handles. In the fall of 1999, Don Knuth (The Art of Computer Programming) gave a series of six public lectures at MIT about interactions between faith and computer science. The sixth and final lecture was concerned with computer programmers as creators of new universes and computational complexity as an approach to thinking about the question of free will. The lectures were later published as a book [Knuth, 2001]. Over one hundred years earlier, William James (1842-1910) gave an an address to the Harvard Divinity Students entitled "The Dilemma of Determinism" (it was later published in the Unitarian Review, September, 1884, and included in a collection of James writings [James, 1962]. It is clear that we make decisions every day in an attempt to order our lives, and, thus, according to James, we must believe (or at least act as though we believe) in free will. In addition to being a noted philosopher and ethicist, James was a scientist and, in particular, a very influential psychologist (see Louis Menand's "The Metaphysical Club: A Story of Ideas in America" [Menand, 2001] for a fascinating account of the life and times of William James and his contemporaries including Charles Sander Peirce whose writings on the foundations of probability, statistics and the scientific method helped shape how we think of these concepts today). For the philosophically inclined, it's very instructive to read Knuth's account of determinism and free will followed by James's account.

I have to admit to being somewhat disingenuous in my discussion of free will; clearly, I believe in some variant of free will. In discussing free will (as in discussing other baggage-laden terms such as "machine intelligence"), I've tried to be somewhat controversial to probe your biases and get you thinking. If you want to gain more insight into what philosophers mean by free will, I highly recommend Daniel Dennett's aptly titled "Elbow Room: The Varieties of Free Will Worth Wanting" [Dennett, 1984]. But new insights from neuroscience, cognitive science and artificial intelligence are forcing us to look at the issue of free will and determinism from different perspectives. Steven Pinker [1997] cautions that we should remain clear in our purposes when talking about various models concerning what it means to be a human being.

"Free will is an idealization of human beings that makes the ethics game playable. [...] As long as there is no outright coercion or gross malfunction of reasoning, the world is close enough to the idealization of free will that moral theory can meaningfully be applied to it."

- pg. 55 in [Pinker, 1997]

"A human being is simultaneously a machine and a sentient free agent, depending on the purpose of the discussion, just as he is also a taxpayer, an insurance salesman, a dental patient, and two hundred pounds of ballast on a commuter airplane, depending on the purpose of the discussion. The mechanistic stance allows us to understand what makes us tick and how we fit into the physical universe. When those discussions wind down for the day, we go back to talking about each other as free and dignified human beings."

- pg. 56 in [Pinker, 1997]

In addition to the various sorts of free will deemed worth having in Dennett's book [1984], there are proposed mechanistic accounts of free will that agree with what we know of how the brain works and that allow for the application of moral theory. Drew McDermott [2001] claims that "[a] system has free will if and only if it makes decisions based on causal models in which the symbols denoting itself are marked as exempt from causality." McDermott's notion of causal model is similar to the sort of environmental model posited above for hypothesizing, evaluating and predicting the consequences of acting. Being exempt from causality in this case simply means that the system is running a program using such a causal model in which the actions performed by the system are determined by the program. Implicitly, the system believes (or acts as if it believes if you're uncomfortable attributing beliefs to a machine) that it, the system, is in control of itself.

Some of the questions and issues surrounding determinism and free will are currently beyond what science can answer and some of them may remain forever so. Not everything admits to or warrants a scientific explanation. Our attitudes concerning who we are and how we relate to one another and to the universe tend to be very personal. My attitudes today are very different from my attitudes when I was twenty-six and those very different from my attidudes when I was seventeen. I don't consider my earlier views as being necessarily wrong nor would I bet on my current views being right. I'm not even sure how the words "right" and "wrong" apply to my attitudes as they pertain to determinism and free will. You and I could probably have a discussion in which we hashed out exactly what we meant by determinism and free will and even agree on what it would mean for a theory regarding these concepts to be right or wrong, but my guess is that the resulting concepts would be so filtered and sanitized that we would lose interest in them. Grappling with the difficult problem of what these terms mean to each of us is an important way in which we come to terms with ourselves and learn to understand and appreciate the perspectives of others.

I should say, however, that as far as I can tell no one has the definitive answers to the "big" questions concerning "life, the universe and everything" (borrowing the title of Douglas Adams's third volume in the original "Hitchhiker's Guide to the Galaxy", Ballantine Books, 1995). These are questions that require (and merit) a personal lifetime investment in seeking their answers and one of the best ways of seeking such answers is to listen to others and particularly others who don't hold the same opinions as you might. I would defer to Don Knuth on a wide range of questions concerning computer science and mathematics but I disagree with (or perhaps misunderstand) his views on determinism and free will. I also got a lot out of listening to what he has to say. In re-reading William James and Charles Sanders Peirce today, I find some of what they had to say embarrassing (James was the president of the "Society for Psychical Research" from 1894-1895 and told his brother, Henry, to remain watchful for evidence of James's continued presence after his death) but much of their work is as relevant today as it was over one hundred years ago. I suppose that in some very real sense William James is with us today as much as he was during his lifetime (as is the American Society for Psychical Research so who am I to scoff and snicker).