1 Wilson, G., “Cheap Materialism,” Midwest Studies in Philosophy, 4 (1979): 51–72.CrossRefGoogle Scholar

2 Post, J. F., The Faces of Existence: An Essay in Nonreductive Metaphysics (Ithaca, NY: Cornell University Press, 1987).Google Scholar

3 In this paper we make the case for excising Post's realism, but leave further trimming to others. So, for example, Haugeland, J. (“Ontological Supervenience,” The Southern Journal of Philosophy, 22, Supplement [1984]: 1–17)CrossRefGoogle Scholar has argued that the principle of physical exhaustion (Post's inventory) is unnecessary for a physicalism. Haugeland agrees with Post (Faces of Existence, p. 186) that exhaustion is independent of determination. It could also be asked whether Post's reasons for dividing the determination principle into the principle of the physical discernibility and what he calls truth determination are convincing. His motives for distinguishing between them are mentioned in note 31 below. For our part, we believe that these are significant questions, but we pay scant attention to them in this paper.

4 Post, Faces of Existence, p. 159. It has sometimes been argued that, since physics needs mathematics and mathematics needs abstract entities like sets and their properties, physicalism is false. That is, even physics needs non-physical entities. Post wants to avoid having physicalism stand or fall on the nature of mathematical entities. If it turns out that a nominalist construal of mathematics is inadequate, then, for Post, physicalism would need to be defined as the claim that “everything is mathematical-physical; everything is either an abstract entity of mathematics or a concrete entity of physics” (ibid., p. 167–68). There is something unsatisfactory about this strategy. Consider the following extension. If it turns out that there are ghosts that have spiritual natures then physicalism could still be maintained. We would simply define ‘physicalism’ in terms of the claim that everything is spiritual-mathematical-physical. If physicalism is not just going to be the vacuous stipulation that whatever exists and whatever its nature, it shall be called ‘physical’ then something needs to be done to remedy this defect.

5 Post, Faces of Existence, p. 116.

6 Although Post does not draw the implication, this would seem to entail that ‘physical’ cannot be defined solely in terms of any compositional relation since the ultimate components are also thought to be physical.

7 Post, Faces of Existence, p. 121.

8 Ibid.

9 The stipulation that it satisfy a positive predicate is included to avoid the possibility of calling some posited entity physical by use of ‘is not an electron’ (Post, Faces of Existence, p. 122).

10 It is true that shoes and ships and sealing wax are not referred to in those terms in contemporary treatises on physics, but all such objects satisfy predicates like ‘has mass n’, ‘is moving with velocity v’, etc. Post notes that all objects would seem to satisfy gravitational predicates but enters this qualification for extraordinary objects which might not.

11 Post, Faces of Existence, p. 163.

12 Post, Faces of Existence, p. 204–5. Post seems to identify physics with the search for absolute invariants. His examples are of highly abstract principles: no object can have a velocity greater than the speed of light, in all interactions mass-energy is conserved (ibid., p. 205). Post's definition of physics seems clearly dysfunctional. Is plasma physics not physics? What about gas laws? There are branches of physics which deal with specific classes of physical phenomena and their distinctive properties. In fact, most of physics does not take the highly abstract form operative in Post's discussion (cf. Stafleu, M. D., Time and Again: A Systematic Analysis of the Foundations of Physics [Toronto: Wedge, 1980]).Google Scholar

13 Post, Faces of Existence, p. 288.

14 Ibid., p. 203.

15 Ibid., p. 163.

16 Ibid., p. 205.

17 Post, Faces of Existence, p. 162. Contrast Mario Bunge (Scientific Materialism [Dordrecht: Reidel, 1981]) who insists on defining emergence in exclusively ontological terms. For Bunge, properties are traits of natural systems and a property is emergent relative to a given level if no part at that level possesses that property. There are important issues here that are obscured by Post's definition of emergence (presumably an ontological affair) in terms of theory irreducibility (an epistemological affair).

18 Post, Faces of Existence, p. 205.

19 Ibid., p. 207.

20 Ibid.

21 Ibid., p. 202–303.

22 Ibid., p. 206.

23 Ibid., p. 204.

24 In Post's terminology, the identity posited in this principle is a token identity, not a type identity (Faces of Existence, p. 163). Post's differentiation of types, kinds, and sets is weak and is worth restructuring. Some of the difficulties in the token-type distinction are discussed by Haugeland, “Ontological Supervenience.”

25 For Post, ‘discernibility’ is an ontological feature. In other words, ‘physical discernibility’ means ‘physical difference’.

26 Cf. Post, Faces of Existence, p. 177.

27 Ibid., p. 176.

28 Ibid., p. 185–86.

29 Ibid., p. 185. This formulation is offered as a variant on his formulation in terms of truth: “Given any two [physically possible] worlds, if the same physical sentences are true in both, then the same non-physical sentences are true in both” (ibid., p. 185). The examination of his truth theory is not important at this point, but we will discuss it below.

30 Post, Faces of Existence, p. 185.

31 The contrast between discernibility and determination is associated with the structure of the possible world formulations used by Post. Physical discernibility “operates on” (addresses differences between) pairs of entities within worlds, while determination operates on pairs of worlds at a time (Post, Faces of Existence, p. 177). That is, physical discernibility addresses differences between entities (as defined by properties), while determination deals with differences between properties and relations (which are, of course, defined for entities). It is not clear to us that much is gained by framing the distinction into separate principles. Post argues that the division is worth making because discernibility and determination are independent in the sense of it being logically possible to have one without the other (ibid., p. 189). Even if this distinction has merit, it would seem best incorporated into a theory of determination which describes the internal structure of the set of determination relations. (We explore this point further below.) In our critique of Post we ignore this distinction since there is very little cost and some ease of expression involved in simply using “no difference without a physical difference” and “all truth is determined by physical truth” as broad, and roughly synonymous slogans (ibid., p. 188f.).

32 Post, Faces of Existence, p. 26.

33 Ibid., p. 181.

34 Ibid., p. 182.

35 Ibid., p. 186.

36 Ibid., p. 183.

37 In other words, if domain determines domain B and domain B determines domain C, then it follows that domain A determines domain C.

38 Post, Faces of Existence, p. 29. The behaviouristic (anti-mentalist) prejudice is evident in much of the literature on the subject of truth. Quineian suspicions about beliefs and the like seem to have encouraged the adoption of the fiction that sentences are harder, more real and manageable than beliefs. In passing we would like to register the conviction that this is mostly silliness. It is argued, for example, that truth is a feature of interpreted sentences. What exactly is a sentence? (A series of physical markings, no doubt.) An interpretation? (Another sentence, of course.)

Why would anyone believe that interpretations are ontologically firmer than beliefs? There seems no good reason to believe that the relationship of beliefs to the brain is any more mysterious or ontologically suspect than the relationship of interpretations to sentences. Post, having argued for a pluralism which accepts at least the possibility of emergent non-physical properties, should feel free to talk of true beliefs as well as of true sentences and to abandon the behaviourist claptrap.

39 Post, Faces of Existence, p. 30.

40 Ibid., p. 32.

41 Ibid., p. 33, 50.

42 Ibid., p. 50.

43 Ibid.

44 Ibid., p. 287, quoting Rorty, R., Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979), p. 368.Google Scholar

45 Post, Faces of Existence, p. 287.

46 Every way the world is nonetheless corresponds either to a physical entity or to a spatio-temporal sum of physical entities. We have also said that, for Post, truth is not to be equated simply with propositional truth (Post, Faces of Existence, p. 285). All truth, not just truth in the sciences, is determined by physical truth but this does not mean that only physics is a domain of truth.

47 A qualification of this point is that some differences specified by non-physical predicates may require a host of physical predicates of different kinds to cover all cases of that non-physical difference.

48 This means, for example, that anthropology may also be concerned about objective truth. The primary difference between anthropology and physics does not lie in objectivity but in its comprehensiveness. Anthropological truths are true only of humans and their works, while physical truths are true of everything that exists. Physics, according to Post, is concerned with all objects and with properties that characterize all objects. (But see note 12.) Objective truth is truth invariant across perspectives.

49 The invariances expressed in eternal sentences are supposed to explain the variations experienced from different perspectives. So, for example, the roundness of the top of an oil drum accounts for the shapes from straight line to circular that we see when we perceive it from different angles. “We accord objective existence to the shape the would most simply account for all of these varying projections” (Post, Faces of Existence, p. 67). Apparently not only physics expresses invariances and so not only physics is a domain of truth expressing a face of existence.

50 Post, Faces of Existence, p. 309.

51 Ibid., p. 15; cf. p. 307.

52 Haugeland (“Ontological Supervenience,” p. 3) characterizes the metaphysical vision of unity as “the call of the one.” In like spirit, we can identify the plurality theme as “the call of the wild.” Thus Post's vision can be helpfully, if somewhat whimsically, characterized as a synthesis of the call of the one and the call of the wild.

53 Post, Faces of Existence, p. 161.

54 Ibid., p. 22.

55 Ibid., p. 308.

56 The only global anti-realist would be the sceptic who claims to know (not know?) that nothing, including him/herself, is real. Since this global anti-realism is not likely to be the actual view of anyone, it seems silly to counter it with a doctrine called ‘realism’.

57 The primary reason for the apparent vacuity is the excessive generality of the determination notion. The process necessary for specifying determination more adequately is to develop connective theories relating the relevant domains under consideration at any one time (but see also note 58). Then one can base one's epistemological access on the theory of the domains and their interrelations. We discuss Post's underspecification of determination further in subsection 2.2.1.

58 Minimally “the world determines truth” must be meaningfully linked to our belief formation and truth determining efforts via a general epistemology which connects theories of reference, perception, conceptual change, etc. This crucial gap is an important context for the development of connective theories discussed below, although we cannot elaborate these concerns in this essay.

59 Post, Faces of Existence, p. 69–70. We should note in passing that even though Post claims that objectivity is not a matter of intersubjectivity, it is not at all clear that his discussion of invariance when applied to the question of truth does not amount to the same thing. It is interesting that he explains invariance by means of a “Do you see what I see?” question. This is clearly just a question of intersubjective agreement or consensus. Intersubjectivity is just a question of invariance of observation given a substitution of observers, an invariance across perspectives or with respect to different knowers.

60 Post, Faces of Existence, p. 290.

61 Ibid., p. 288.

62 Ibid., p. 67–68.

63 Ibid., p. 69–70.

64 Ibid., p. 249; original emphasis.

65 Ibid., p. 244.

66 Ibid., p. 312.

67 Thomas Kuhn's arguments about the role of paradigms or disciplinary matrices and Michael Polanyi's arguments about the tacit dimension of knowledge are relevant here.

68 Post, Faces of Existence, p. 282.

69 Ibid., p. 69.

70 Ibid., p. 68.

71 Ibid., p. 287.

72 Ibid., p. 309.

73 Ibid.

74 It is arguable that properties are simply powers to produce effects. See Harré, R. and Madden, E. H., Causal Powers (Oxford: Blackwell, 1977)Google Scholar. Some properties are comprehensive. So, for example, virtually all physical objects have gravitational effects on each other. Other powers (properties) are more restrictive. Roses smell sweet only to organisms with the necessary olfactory organs. Is gravity a face and odour a mask? Is odour a mask if I “mistakenly” believe that it is something that roses possess or emit?

75 There are many aspects of determination worth exploring. For example, in his refutation of the Principle of Sufficient Reason, Post describes determination as an “ultimate fact” but glosses over the nature of his primitives (Faces of Existence, p. 187). The odd point about his argument here is that he merely asserts the possibility of the inexplicability of determination; he offers no substantive considerations to support his supposition. Nor does he explicitly posit determination as an axiom or as a useful but primitive assertion. These features of his argument make sense as another consequence of the excessively general nature of the determination notion noted below. And the same is most likely true of the epistemological-ontological divorce noted in subsection 2.1.2 above. These and other points are worth consideration over the long run, but are not developed in any detail in this paper.

76 See also Haugeland, “Ontological Supervenience.”

77 We are not requiring some kind of complete explanatory theory of determination here. We are merely claiming that mere logical compatibility is not an adequate criterion for establishing intelligibility at least in this instance. This is not a novel claim, and Post himself seems to agree when a similar move is made in an opposing position (e.g., Post, Faces of Existence, p. 324, 332).

78 Post, Faces of Existence, p. 215.

79 When Post sets out to develop the notion of determination, he frames the issue as “how we could ever tell whether various truths are determined by truths from physics, if not by way of reduction, translation, or even extensional isomorphism?” (Faces of Existence, p. 215). This is the ascertainment question which he acknowledges does not follow from generic determination by itself (ibid., p. 186–87, 258). Our point is that we not only need methodological guidelines with relevance for epistemological criteria. We also need a demonstration that such guidelines are viable in precisely the crucial place for making the Post-physicalist claim: weak determination. Part of what needs to be “put up” is a specification of weak determination that addresses in principle the ways by which such relations might be ascertained.

80 Post, Faces of Existence, p. 201–2.

81 That this is not a trivial issue becomes evident when one starts to sort out this question in the context of similar concerns following from part-whole relations. See, for example, Hart, H., Understanding Our World: An Integral Ontology (Lanham, MD: University Press of America, 1984)Google Scholar and his references on enkapsis. As for kinds of weak determination relationships, the most likely candidate Post mentions is extensional isomorphism. This concept is not applicable to Chaco Canyon, and its status is probably borderline between weak and strong, emergent and reductive determination relations.

82 By way of definition, a connective theory consists of three components according to Post: (a) an interpreted theory internal to the determining domain X, (b) a set of auxiliary assumptions required to apply the theory in (a) to particular cases, and (c) a set of bridge principles whose vocabulary is made up of terms from X and from the determined domain Y. This is basically Ernest Nagel's scheme.

The missing theory of relations between culture and geology in the Chaco Canyon example is precisely what Post calls a connective theory (or a set of such theories). He provides no connective theory for either identifying the kind(s) of determination relationship realized in that example, or for comparing the example with other kinds of determination relations.

83 It is worthwhile to be careful about this point. Consider the extremes of a continuum: at one end are tasks performed by philosophers as “metascientists” and at the other end are tasks taken on by members of the special sciences. For example, characterizing the logical structure and consequences of scientific theory is a task often taken on by philosophers and ignored by scientists. Likewise, the review of empirical research relating two domains (like conditioning and neural structures) is usually taken on by specialists in one or the other or both of the subdisciplines in question. But what about reviewing the historical development of specific theoretical traditions or the development of metatheoretical and methodological foundations to a discipline? It is not hard to find arenas of study where the philosopher and special scientist work on the same or similar topics (and where the overlap is reflected in both cooperation and counter efforts). Recognition of the overlap between philosophical and special scientific tasks, for example, makes it easier to avoid using “passing the buck” as a viable strategy.

84 Post claims that any causal relationship is also a determination relationship (Faces of Existence, p. 225), but does so without elaborating a model of causality. Although we can sympathize with his desire to avoid that particular nest of hornets, his escape effort nonetheless led to getting stung. The problem is this. Determination of Y by X claims merely that “Given the way X is, there is one and only one way yean be.” As Post is at pains to point out, this does not imply that some other X' might not also determine Y. For example, one's thought about the Oilers taking the Stanley Cup in 1990 could be determined by two different sets of physical states in the world (including, perhaps, two possible brain states). Thus X is a sufficient, but non-necessary condition for Y to occur. Yet most models of causality are by no means restricted to sufficient but non-necessary conditions. For example, a necessary but non-sufficient cause of a human thought is the activation of a human brain (until we develop an artificial medium for human consciousness, assuming that is possible). And while the properties of neural functioning are causal in an important sense, this form of causality cannot be characterized exclusively as fitting the definition of a determination relation.

Perhaps Post just slipped at this one point but does realize that causality does not generally entail determination. This awareness might be behind his discussion on p. 46–47 of Faces of Existence or his gloss “the intended sense” (ibid., p. 225). In any event, the interconnections between causality and determination are important and complex, no doubt requiring more than a few passing comments from either Post or us to sort them out.

85 Post, Faces of Existence, p. 218–20.

86 Ibid., p. 219.

87 Ibid., p. 219–20.

88 Ibid., p. 279f.

89 Ibid., p. 280.

90 If you invite us to a social event, we should like to know whether it's a costume party, an opera, or the roller derby! Although the analogy to social invitations is not strict, the most relevant distinction between them is that adopting metaphysical principles might well have a larger impact on our lives than committing ourselves to a social evening. To take either proposal seriously, we deserve a hint as to what it is we are getting ourselves into.

91 We agree with Post that an adequate determination theory must be applied significantly to domains outside of science. We are not promoting scientism by our choice of example; we are providing the greater challenge. For he has not even demonstrated adequacy in scientific domains even though his work pays special attention to science as subject matter.

92 Note that Post has no difficulty incorporating events from an evolutionary past or from contemporary and immediate past ecosystem parameters (including food availability). Since it is the whole world which determines, there is nothing odd if the special sciences identify widely scattered domains of significance for determination. It is in fact precisely this kind of flexibility which Post built into his system.

93 For those who want to round out Aristotle's four causes, formal causes are intertwined with all of the other categories. For example, the efficient causal exposition of reproduction assumes the DNA pattern in parent cells as templates, patterns of constraint (formal causes) upon which the efficient and material causes of cellular mechanisms act.

As for the notion of historical cause, we are simply attempting to point out that the ostensibly unique and “coincidental” events at crucial junctures in the process of evolution are not characterized by the replicability criterion often used to define cause. That is not to say that such events are uncommon, but it is to say that we usually consider most of them unimportant. Yet we cannot ignore historical events in the domains of evolution or cosmology, for example.

94 When he characterizes his position (“‘Downward Causation’ in Hierarchically Organised Biological Systems,” in Ayala, F. J. and Dobzhansky, T., eds., Studies in the Philosophy of Biology: Reduction and Related Problems [Berkeley, CA: University of California Press, 1974], p. 179)CrossRefGoogle Scholar, D. T. Campbell talks about being an anti-vitalistic emergentist. Of course, we should also acknowledge that he also calls himself a reductionist of some sort (p. 181–82).

95 Campbell, “Downward Causation,” p. 181.

96 The available terminology certainly betrays the history of discussion in this area. There are a variety of distinctions within the strong determination relations, but relatively few within weak determination, even given the approximate nature of the sequence as provided here.

97 This concern was raised above in the discussion of Post-realism.

98 Post, Faces of Existence, p. 207–8; original emphasis.

99 Ibid., p. 308.

100 Ibid., p. 282.

101 Ibid., p. 283.

102 Ibid., p. 282.

103 Why not trivial? Rather than the designation of values to ‘unreality’ by reductionist materialisms, it becomes reasonable, in the sense of being logically compatible with physicalism, to assert values. Despite the epistemologically irrelevant “correct distribution of truth values over value judgements,” the moral connective theories we develop enter the race for knowledge on much the same basis as other theories. In the face of traditional reductionism and various non-cognitivisms, this is not entirely trivial.

104 Post, Faces of Existence, p. 284.