|1. Absolute vs. Objective|
The absolute/relative distinction is related to, but different from, the objective/nonobjective distinction. There can be objective truths about issues which are relative. The right shoe size is relative to the size of your foot, but there are objective truths about which shoes are the right size for which feet. Similarly, the velocity of an object is relative to a frame of reference, but there are objective truths about the velocity of an object relative to any frame. Sometimes we discover (as with velocity) that a certain class of facts are relative in an unexpected way. But then we attempt to take a step back and uncover the (objective) relationship between these facts and whatever it is they are relative to.
|2. An argument for realism: van Inwagen's anti-antirealism argument|
This is a quick and dirty, unsophisticated version of an argument van Inwagen uses. It's a sort of tu quoque argument. (Is it a problem that tu quoque is usually counted as a logical fallacy? I think in this case it's actually a sound criticism, but there might be food for thought here.) The antirealist says that there are no objective truths. Is that supposed to be an objective truth? If so, the view is inconsistent. If not, then it seems that realism is equally true. But the antirealist is supposedly criticizing realism.
I like this argument, but I feel a little guilty about it. (Especially about my presentation above, which is very brief and not at all nuanced.) Maybe I'll try to present a fairer version of the state of play later.
|3. An argument for antirealism: the incoherence of the view from nowhere|
Here's another sketchy, unsophisticated presentation of an argument. It is sometimes suggested that the idea of an objective conception of the world is the idea of a view of the world from no point of view, a "view from nowhere." (Thomas Nagel's phrase, from his book of the same name, but Nagel is actually a defender of objectivity.) But, it is suggested, there is no such thing as a view from nowhere. Every view is inevitably situated, is inevitably from a particular perspective.
Response 1: we are not limited to a single perspective. We can collect information about how the world appears from various points of view, from various perspectives. This gives us something more closely approaching objectivity. (It's rather like the point in section 1 above that, when we discover that something is relative, we step back and try to see how the dependence works.)
Response 2: a representation of the world does not have to be a "view." It can be a more basic sort of representation that is not a view at all. Analogy: ray-tracing graphics programs. These involve having a "model" of what we want to depict, and then adding information about light sources and the location of the frame through which the objects are to be viewed, and using the physics of light reflection, refraction, etc. to determine the intensity and color of the light at each point in the frame. The model is not pictorial at all (it's a purely geometric description of the space occupied by the represented objects), but we can use it plus information about the viewer to generate lots of specific views of the modeled world.
And I think this is actually a very nice metaphor for what science tries to do in general. We try to find an objective, observer-independent representation of the world which we can combine with information about observers to yield information about how the world will appear from various points of view.
|4. The social construction of absolutely everything|
One version of antirealism is social constructionism. The first widely disseminated use of this term that I know of (which does not at all imply that it is actually the first!) is Peter Berger and Thomas Luckmann, The Social Construction of Reality (1966). I haven't looked at this for so long, I'm not sure whether the authors actually make any metaphysical claims or not. But many more recent books with similar titles do make rather radical antirealist claims: the claim that a particular kind of thing is "socially constructed" typically goes along with the claim that it therefore is not objectively real (in the realist's sense, though it may be "objective" in some weaker sense, e.g. intersubjective). One example in this genre is Bruno Latour and Bruce Woolgar, Laboratory Life, which argues for the social construction of TRF(H), or Thyrotropin Releasing Factor (Hormone). Another is Andrew Pickering, Constructing Quarks (1984). Ian Hacking's book The Social Construction of What? discusses, as a review put it, "the social construction of science, madness, child abuse, and rocks." (Hacking is not exactly a gung-ho social constructionist, though; he is trying to mediate the dispute between constructionists and realists.)
It seems to me that extreme social constructionists constantly make what looks to me to be a serious mistake or confusion: they give persuasive evidence that a given theory, or model, or picture of the world is socially constructed, and go on to conclude that the world itself is socially constructed. But the fact that we construct theories can hardly show that we construct the world the theories are about!
Sometimes constructivists reply to this objection that they are distinguishing between the "world" we know and love, which is socially constructed, and the real world which underlies it, which is not. However, they claim that the real world, in this sense, is entirely unknowable, and in fact cannot even coherently be spoken of. (This view obviously bears a strong resemblance to Kant's distinction between appearances and things-in-themselves, except that for Kant appearances are constructed by means of categories and forms of intuition that are necessarily shared by all humans, while the constructionists thing that our basic concepts may change over time and may differ between different cultures.) The world we can know about, they insist, is only the world as categorized or carved up by our concepts and categories, and by carving it up in one way rather than others, we are creating the world we know about.
However, the fact that we can only talk about the world by using (humanly constructed) language does not show that there is anything humanly constructed about the world we are talking about! It's entirely possible that the theories we have constructed to understand the world are actually correct, so that in using the theory we are in fact describing the really real world. (It's true that concepts or categories, unlike theories, cannot be correct or incorrect, and that we might have used different concepts to understand the world. It remains the case, though, that, given a set of concepts or categories, statements about the world made with them are made true or false by the world, not by human constructions. --Is merely using one set of concepts rather than another, "carving the world up," a way of constructing a world, or at least constructing an organization of the world? No -- it is a way of selecting one organization rather than another for our use. The fact that we could have selected a different organization doesn't show that we have somehow done something to the world by selecting the one we did.)
A couple of representative quotations from Latour and Woolgar, Laboratory Life: The Construction of Scientific Facts (Princeton: Princeton University Press, 1986; first edition 1979).
"[W]e have chosen to study the historical genesis of what is now a particularly solid fact. TRF(H) is now an object with a well-defined molecular structure, which at first sight would hardly seem amenable to sociological analysis. If the process of social construction can be demonstrated for a fact of such apparent solidity, we feel this would provide a telling argument for the feasibility of the strong programme in the sociology of science" (p. 106).
"[I]t would be incorrect to conclude that the TRF story only exhibits the partial influence of sociological features. Instead, we claim that TRF is a thoroughly social construction" (p. 152).