On the Plurality of Worlds [Hardcover]

It's obvious to me that some of these reviewers don't understand either what Lewis was doing in this book or the standard philosophical response to it. He wasn't arguing that there are multiple universes connected to ours in some way explorable through science. Those would be parts of this world, in Lewis' sense. The worlds he's talking about are possible worlds. They aren't actual. That is, they don't exist in any way spatiotemporally or causally connected with the actual world. And yet he says they're as concrete as we are.

Philosophers' responses to this view are incredibly interesting. They think the idea is nuts, and yet they have no way to resist the conclusion that he gives compelling arguments that his view solves numerous philosophical problems that no one has been able to deal with in a perfectly satisfactory way. This doesn't convince many people that his view is correct, but the response has been pretty strong among those who want to use his system without thinking that it's true. They call it a modal fiction, and the view is called fictionalism. This is becoming incredibly influential among metaphysicians.

Aside from all that, most metaphysicians today recognize this book as just incredibly fruitful and creative in bringing together so many different strains in metaphysics. He deals with so many problems in such a lucid way that the book serves to introduce many problems in metaphysics, making advances in the discussion even apart from the contribution of his main thesis.

Lewis' work always produces mixed feelings in me. There is a long list of good things to be said on his behalf, and, so far as I can see, only one really bad thing. But the bad thing is bad enough to taint my otherwise unalloyed (and profound) respect and admiration for this superb thinker, quite probably the best philosopher alive, and almost certainly the best I've ever read. (Aaron, if you are reading this, you were right about your teacher; he deserves all the plaudits you kept showering on him, and for all I know he may even be right about the crazy things he says.)

First, the pros. Lewis offers a modal metaphysics that is a) technically brilliant, elegant, and well-motivated, b) capable of providing reductive analyses of a vast range of otherwise obscure notions, and c) the best game in town if considered simply on it theoretical merits. He shows how his modal realism can be used to analyze modality (necessity, possibility, and the like, as well as restricted modalities) and mental and semantic content, and to make room for properties and counterfactual claims. He shows how to dissolve the debate among essentialists and anti-essentialists using counterpart theory, how to avoid various apparently serious objections to modal realism, and how to understand the debates about de re modality and "transworld identity". He offers a clear account of ersatzism, especially of the linguistic variety, which (I agree with him in holding) is the only real alternative to his modal realism among the theories so far offered, and is as such clearly an inferior theory. He offers a devastating argument against "magical ersatzism", probably the most commonly held view on modality apart from linguistic approaches. The argument is decisive, I think, and shows that approaches like those of Plantinga or Stalnaker cannot succeed. His modal realism has its limitations; because he cannot help himself to worlds bigger than a certain size and shape, and because he has all the worlds as concrete objects existing "side by side", so to speak, he can't countenance the full range of possibilities that an ersatzist can, and his analyses come apart at the edges. (For instance, the thesis of the plurality of worlds itself is, on Lewis' account, necessarily false; thus he must believe in necessarily false truths. Again, he cannot admit the possibility of two spatiotemporally disconnected universes, or worlds that differ as to which non-spatiotemporal abstracta exist in them.) When these defects are compared against the defects of Lewis' competitors, however, his view emerges as the clear winner.

Nonetheless, in the end all of this theoretical advantage is not enough. The thesis of the plurality of worlds is simply incredible. Indeed, I would go so far as to say that its falsehood is something like a Moorean fact; it disagrees so badly with common sense that no philosophical considerations in its favor could constitute an adequate reason for accepting it. Lewis's methodology is right; he proceeds in theorizing by weighing the costs of departure from pretheoretical opinion against the benefits in theoretical economy and utility. Unfortunately, this method is here being used by the wrong person; for all his brilliance, Lewis just doesn't have the *good sense*, or wisdom, to see how crazy his theory is--and this is the only thing that lessens my admiration for him. Faced with the evident inadequacies of all the competitor accounts that Lewis surveys, I am forced to make the choice Lewis says I will be forced to make:

"Paradise on the cheap, like the famous free lunch, is not to be had. Make of this what you will. Join the genuine modal realists; or foresake genuine and ersatz worlds alike."

Since I cannot join the modal realists (literally, "cannot"; I am psychologically incapable of it), I choose to foresake possibilia altogether. Ersatz or concrete, they must go; and if I can think of no alternative to these approaches--which are after all the only ones now on offer--I shall simply suspend judgement on whether there are really possible worlds of *any* sort, and on their nature and how we can do what we do with them. But I insist on maintaining the hope that some future theorist will tell us how we can help ourselves to possibilia, in a way that we can live with--for it seems absurd to give up the benefits of talking in terms of possibilia just because we have no adequate and credible metaphysics of them.

In the book _On the Plurality of Worlds_, analytic philosopher David Lewis presents his case for modal realism (also called "extreme modal realism", though why it is "extreme" is questioned). As Lewis explains, "This book defends modal realism: the thesis that the world we are part of is but one of a plurality of worlds, and that we who inhabit this world are only a few out of all the inhabitants of all the worlds." Lewis will argue for this position based on the fact of the utility of modal realism for philosophy provides reason to believe that it is true (just as the utility of set theory provides reason to believe that there are sets), then he will consider the common objections to this position and argue against them. Lewis maintains that he does not provide an argument that effectively requires one to believe in a plurality of worlds, but only that he will provide a cost-benefits analysis of this idea. The idea of possible worlds is a recurrent one in the philosophy of Leibniz; however, Lewis does not mention him because he feels that any interpretation he might offer of Leibniz's works would be inadequate. David Lewis (1941-2001) who taught at Princeton was one of the more interesting thinkers in analytic philosophy and this book is certain to remain a classic and can be appreciated even if one cannot go all the way with Lewis in accepting his arguments for modal realism. It should also be pointed out that Lewis' arguments while often subtle can be followed with some effort and that he presents a very effective case.

This book consists of four chapters. The first chapter "A Philosophers' Paradise" lays the case for modal realism and argues that a plurality of worlds has utility in philosophy. Lewis begins by noting that he will restrict himself to possibilia and by writing off impossibilia. Lewis then goes on to explain how the usefulness of a plurality of worlds makes it likely for one to think that it is indeed true. Lewis considers the case of a possible worlds analysis of modality (showing how the "diamond" ("possibly") and "box" ("necessarily") symbols are mutually interdefinable and can be understood in terms of possible worlds). Lewis also considers the idea of "closeness" of possible worlds and defines such notions as counterfactuals and verisimilitude. Following this, Lewis considers content of possible worlds and considers analyses of thought, belief, and language. Lewis then shows how modal realism may be used to explain properties (universals and tropes), the fact that possible worlds are isolated and not spatio-temporally connected, and the fact of concreteness, plenitude, and actuality. Lewis maintains that actuality must be understood by making use of an indexical analysis. The second chapter "Paradox in Paradise?" shows some of the commonly made objections to modal realism and argues for why Lewis thinks they are inadequate. Lewis considers the objections made that according to his analysis "everything is actual", objections that try to trap modal realism in paradoxes akin to those of naïve set theory, an objection made that there are more worlds than there are (relying on a paradox in set theory), epistemic objections that we cannot know, objections made that the existence of possible worlds provides a road to skepticism or indifference (i.e. ethical paradoxes brought about by the existence of possible worlds), an objection made that modal realism causes the arbitrariness of this actual world to be lost, and finally "the incredulous stare" (the objection that modal realism flies in the face of common sense). Lewis provides good arguments against most of these objections, but ultimately the most powerful seem to be the objections from common sense. The third chapter "Paradise on the Cheap?" argues against the alternative to Lewis' view which attempts to make use of the utility of a plurality of worlds without committing oneself to modal realism (what Lewis calls "ersatz modal realism"). The "ersatzist program" replaces possible worlds with mere abstract representations. Lewis considers what he considers to be the three leading possibilities for ersatzism: linguistic ersatzism, pictorial ersatzism, and what he calls "magical" ersatzism. He finds each of these alternatives to be problematic and ultimately not be able to provide the benefits of possible worlds given the modal realist understanding. The fourth chapter is "Counterparts or Double Lives?". Here, Lewis considers the question of "counterparts" across worlds and attempts to resolve some of the apparent paradoxes which may ensue. He considers the cases of trans-world identity and overlap. Then, he provides an argument against the existence of trans-world individuals and an argument against what he terms "haecceitism". Finally, he provides an argument against constancy. From these arguments he concludes that his notion of counterparts is the correct analysis and is superior to the alternatives. Thus, stands the case for Lewis' modal realism.

Among analytic philosophers in the last century, David Lewis stands out as one of the most important. His arguments for modal realism in this book are fascinating and are important if one seeks to understand the issue of possible worlds. Ultimately even if one finds such objections as ethical paradoxes brought about by modal realism or the common sense objection convincing, Lewis' work still remains important in that it must be seriously considered. It should also be pointed out that while the idea of multiple worlds or universes appears to be playing some role in the more abstract reaches of theoretical physics (including the possibility that quantum mechanics may best be resolved by an appeal to multiple worlds) that these multiple worlds considered by physicists differ from the possible worlds of philosophers such as Lewis. To begin with there is the fact that the possible worlds considered by Lewis are spatiotemporally isolated. This does not appear to be the case with the multiple worlds considered by some physicists. Further, there may be reason to believe that the speculations of certain physicists regarding multiple universes are unwarranted scientifically (perhaps by an appeal to Occam's razor). Nevertheless, Lewis' analysis of possible worlds remains important and should be read by those interested in the further development of analytic philosophy.