There's a lot of stuff in the world: trees, cars, galaxies, benzene, the Baths of Caracalla, your pancreas, Ottawa, ennui, Walter Mondale. How does it all fit together? In a word… Supervenience. (Pronounced soo-per-VEEN-yence. The verb form is to supervene.)
Supervenience is a shorthand abstraction, native to Anglo-American philosophy, that provides a general framework for thinking about how everything relates to everything else. The technical definition of supervenience is somewhat awkward:
Supervenience is a relationship between two sets of properties. Call them Set A and Set B. The Set A properties supervene on the Set B properties if and only if no two things can differ in their A properties without also differing in their B properties.
This definition, while admirably precise, makes it hard to see what supervenience is really about, which is the relationships among different levels of reality. Take, for example, a computer screen displaying a picture. At a high level, at the level of images, a screen may depict an image of a dog sitting in a rowboat, curled up next to a life vest. The screen's content can also be described as an arrangement of pixels, a set of locations and corresponding colors. The image supervenes on the pixels. This is because a screen's image-level properties (its dogginess, its rowboatness) cannot differ from another screen's image-level properties unless the two screens also differ in their pixel-level properties.
The pixels and the image are, in a very real sense, the same thing. But — and this is key — their relationship is asymmetrical. The image supervenes on the pixels, but the pixels do not supervene on the image. This is because screens can differ in their pixel-level properties without differing in their image-level properties. For example, the same image may be displayed at two different sizes or resolutions. And if you knock out a few pixels, it's still the same image. (Changing a few pixels will not protect you from charges of copyright infringement.) Perhaps the easiest way to think about the asymmetry of supervenience is in terms of what determines what. Determining the pixels completely determines the image, but determining the image does not completely determine the pixels.
The concept of supervenience deserves wider currency because it allows us to think clearly about many things, not just about images and pixels. Supervenience explains, for example, why physics is the most fundamental science and why the things that physicists study are the most fundamental things. To many people, this sounds like a value judgment, but it's not, or need not be. Physics is fundamental because everything in the universe, from your pancreas to Ottawa, supervenes on physical stuff. (Or so "physicalists" like me claim.) If there were a universe physically identical to ours, then it would also include a pancreas just like yours and an Ottawa just like Canada's.
Supervenience is especially helpful when grappling with three contentious and closely related issues: (1) the relationship between science and the humanities, (2) the relationship between the mind and brain, and (3) the relationship between facts and values.
Humanists sometimes perceive science as imperialistic, as aspiring to take over the humanities, to "reduce" everything to electrons, genes, numbers, and neurons, and thus to "explain away" all of the things that make life worth living. Such thoughts are accompanied by disdain or fear, depending on how credible such ambitions are taken to be. Scientists, for their part, sometimes are imperious, dismissing humanists and their pursuits as childish and unworthy of respect. Supervenience can help us think about how science and the humanities fit together, why science is sometimes perceived as encroaching on the humanist's territory, and the extent to which such perceptions are and are not valid.
It would seem that humanists and scientists study different things. Humanists are concerned with things like love, revenge, beauty, cruelty, and our evolving conceptions of such things. Scientists study things like electrons and nucleotides. But sometimes it sounds like scientists are getting greedy. Physicists aspire to construct a complete physical theory, which is sometimes called a "Theory of Everything" (TOE). If humanists and scientists study different things, and if physics covers everything, then what is left for the humanists? (Or, for that matter, non-physicists?)
There is a sense in which a TOE really is a TOE, and there is a sense in which it's not. A TOE is a complete theory of everything upon which everything else supervenes. If two worlds are physically identical, then they are also humanistically identical, containing exactly the same love, revenge, beauty, cruelty, and conceptions thereof. But that does not mean that a TOE puts all other theorizing out of business, not by a long shot. A TOE won't tell you anything interesting about Macbeth or the Boxer Rebellion.
Perhaps the threat from physics was never all that serious. Today, the real threat, if there is one, is from the behavioral sciences, especially the sciences that connect the kind of "hard" science we all studied in high school to humanistic concerns. In my opinion, three sciences stand out in this regard: behavioral genetics, evolutionary psychology, and cognitive neuroscience. I study moral judgment, a classically humanistic topic. I do this in part by scanning people's brains while they make moral judgments. More recently I've started looking at genes, and my work is guided by evolutionary thinking. My work assumes that the mind supervenes on the brain, and I attempt to explain human values — for example the tension between individual rights and the greater good — in terms of competing neural systems.
I can tell you from personal experience that this kind of work makes some humanists uncomfortable. During the discussion following a talk I gave at Harvard's Humanities Center, a prominent professor declared that my talk — not any particular conclusion I'd drawn, but the whole approach — made him physically ill. (Of course, this could just be me!)
The subject matter of the humanities has always supervened on the subject matter of the physical sciences, but in the past a humanist could comfortably ignore the subvening physical details, much as an admirer of a picture can ignore the pixel-level details. Is that still true? Perhaps it is. Perhaps it depends on one's interests. In any case, it's nothing to be worried sick about.
NB: Andrea Heberlein points out that "supervenience" may also refer to exceptional levels of convenience, as in, "New Chinese take-out right around the corner — Supervenient!"