By Daniel C. Dennett

Viking. 324 pp. $25.95 The 18th-century English lawyer Oliver Edwards famously said, "I have tried too in my time to be a philosopher; but, I don't know how, cheerfulness was always breaking in." I suspect a lot of us feel that way. Philosophy may be important, but it sure is dull.

Tufts University professor Daniel C. Dennett is a wonderful counterexample. In a series of sparkling books, written for the intelligent general reader, Dennett has taken on really big issues, made them clear, dealt with them seriously and given us much on which to reflect and (most important) with which to disagree. "Consciousness Explained" took on the mind-body problem. "Darwin's Dangerous Idea" took on evolution and its implications for us as humans. And now, in "Freedom Evolves," Dennett tackles the problem of free will.

What is this problem? It is easy to explain and hard to solve. We humans seem to have a dimension of freedom -- an ability to choose -- not possessed by inanimate objects or by plants and most lower animals. A bee barring an intruder from the hive cannot choose instead to be warm and friendly. But we humans do have this power. President Bush could decide to give the order to attack Iraq, or he could decide to restrain his troops and go another route -- negotiation, for instance. He has the option, unlike the bee.

As Dennett rightly stresses, free will is not some incidental thing, for on it hangs all of human morality. You choose freely to do good or ill. If people are not free, then they are not moral agents. But how do you reconcile free will with our thinking about determinism? Science has shown with increasing certainty that the world works rather like a clockwork machine, with everything -- including us humans -- part of a process governed by fixed laws. No options. No choices. No free will. Nor do you escape the paradox if you protest that modern science is indeterministic, rather than deterministic; that the world is not a clockwork machine but something with events governed purely by chance. A person whose actions are random is crazy, not free.

In an earlier book, "Elbow Room," Dennett favored the solution of the great Scottish philosopher David Hume. Known as "compatibilism," the position argues that the real contrast is not between freedom and determinism, but between freedom and autonomy or lack of the latter. People can be governed by laws but still be free, so long as they are not in chains or under the influence of drugs or hypnotized or whatever.

In "Freedom Evolves," Dennett continues this thinking, now brought into line with and enriched by his enthusiasm for Darwinian biology. The point is not that we humans uniquely have something like an ethereal soul that makes us free, but rather that evolution through natural selection has made us uniquely able to reason and to reflect on and to control our actions in ways unavailable to other beasts. Especially through our culture, we are able to manipulate situations, particularly social situations with our fellow humans, and to promote our own interests. It is here, in this biologically conferred autonomy, that true freedom resides.

As always when Dennett is writing, there is much of great interest along the way. This is a man who truly loves science and enjoys reporting on it and trying to relate it to the philosophical points he is making. He is particularly good when dealing with the work of those social psychologists who are, both in theory and in practice, trying to relate our biological needs to our behaviors in groups, showing how basic norms of moral behavior might have emerged naturally rather than on stone tablets carried down from on high. Dennett is crisp and critically insightful on all sorts of flabby presuppositions, such as those about the inevitability of genetic determinism, those claiming the supposed self-interest of all actions, and assumptions about the essential value of being natural or of cherishing what Mother Nature has done for us.

Arguing against the last of these, Dennett tells us that in his sweaty adolescence, one of his favorite magazines was a nudist publication, American Sunbather. We learn that it was here that he first read of and came to doubt claims hymning the virtues of being natural -- in this case, going around stark naked. I suspect that his memory is playing tricks here. My extensive experience of these matters tells me that no teenage male ever reads the text of nudist magazines.

I accept the conclusions to which Dennett points. Without compatibilism, I do not see how one escapes the paradoxes of free will. But for all its virtues, I doubt that "Freedom Evolves" will last as one of Dennett's better books. As a whole, it seems shaggy and unfinished. Too often, he stops to recap something said elsewhere, or to engage in a fight with a critic, or to emote over something that he has just read and enjoyed. I wish his editors had sent back his manuscript and told him now to produce a final draft, cutting and smoothing and writing the book as a whole. "Freedom Evolves" is good, then, but not as good as it might have been.