Why There Isn’t a Ready-Made World
Metaphysical Realism is true.
(2) There are no Essential or Intrinsic Properties.
Why do people accept (1)?
Why do people accept (2)?
Arguments against conjunction of these two claims:
1. No intelligible correspondence relation. Why? Would require an act of intentional stipulation.
This would require that one hold the word and the object before one’s mind and then stipulate that the first corresponds to the second.
But, one can’t hold mind-independent objects before one’s mind.
2. Furthermore, one couldn’t hold a correspondence relation before one’s mind, since this is also supposed to be a mind-independent item.
Putnam’s observation: The tension or incompatibility between metaphysical realism and the denial of intrinsic properties has not gone unnoticed by modern materialists.
Problem, however: One cannot accept the existence of essential properties and still remain a materialist. Why? Because the sorts of notions which need to be invoked to give content to the notion of an essential property are not, themselves, physical notions.
Putnam’s goal: To show that one cannot give an account of essences that is genuinely physical. One cannot be both an essentialist and a metaphysical materialist. Or, Essentialism à (Not) Materialism. In particular, one cannot try to use essentialism to support one’s materialism.
Why focus on materialism?
Putnam’s objection to materialism – Scientism
How does the materialist try to make sense of essentialism on the physicalist world-picture? By invoking causation or essences.
There are, broadly, three different strategies for using the notion of causation: (1) Humean analysis in terms of constant conjunction; (2) Sui generus talk about causal powers; (3) Counterfactual analysis. Finally, we can invoke (4) essences directly.
(1) Humean analysis of causation:
“Now if’A causes B’ simply meant ‘whenever an A-type event happens, then a B-type event follows in time’, ‘causes’ would be physically definable. Many attempts have been made to give such a definition of causation-—one which would apply to genuine causal laws while not applying to sequences we would regard as coincidental or otherwise non-causal. Few philosophers believe today that this is possible.”
Why isn’t it definable? Think about Hume.
But, ignore this problem for the time being. Would causation, so construed, be a truly physical relation?
Putnam says “no”. Why?
“Mill pointed out that in ordinary language ‘cause’ rarely (if ever) means ‘total cause’. When I say ‘failure to put out the campfire caused the forest fire’, I do not mean that the campfire’s remaining lit during a certain interval was the total cause of the forest fire. Many other things—the dryness of the leaves, their proximity to the campfire, the temperature of the day, even the presence of oxygen in the atmosphere—are part of the total cause of the forest fire. Mill’s point is that we regard certain parts of the total cause as ‘background’, and refer only to the part of interest as ‘the’ cause.”
We refer to our interests and purposes in providing an explanation. Thus, the notion of a cause is not simply a notion of physics. If it were, then the causal relation could simply be discovered in the world. What counts as the cause of event b could not depend upon what one happens to know or what one’s interests and purposes are in asking the question.
(2) Sui generus talk about causal powers.
“‘Causes’, we have just seen, is often paraphrasable as ‘explains’. It rarely or never means ‘is the total cause of’. When Boyd, for example, says that a certain micro-structure is a ‘causal power’ (the micro-structure of sugar is a ‘causal power’ in Boyd’s sense, because it causally explains why sugar dissolves in water) he does not mean that the micro-structure in question is the total cause of the explained events (sugar will not dissolve in water if the water is frozen, for example, or if the water is already saturated with sugar, or if the water-cum sugar is in an exotic quantum mechanical state). ‘Causal powers’ are properties that explain something, given background conditions and given standards of salience and relevance.”
Richard Boyd’s suggestion. Leave the notion of causation unanalyzed. Talk, e.g., about the causal power of various micro-structural features of sugar to cause sugar to dissolve in water and leave it at that.
Putnam’s response to this suggestion: There is nothing physicalistically acceptable about the notion of a sui generus concept of causation. Why not?
Once again, our interests and purposes help explain why we identify solubility of sugar with its microstructure. We can only do this if we take various standards of relevance and salience to inform nature itself (what if the water is frozen or already saturated with sugar, or in some freakish quantum mechanical state?)
(3) Counterfactual analysis
Idea here is that we analyze “A causes B” as “If A had occurred, then B would have occurred” or as “If A had not occurred, then B would not have occurred.”
“Lewis chooses to give truth conditions for contrary-to-fact conditionals. He postulates that there actually exist ‘other possible worlds’, and that there is a ‘similarity metric’ which determines how ‘near’ or how ‘similar’ any two possible worlds are (Lewis, 1973). A contrary-to-fact conditional, ‘If X had happened, then Y would have happened’, is true just in case Y is actually true in all the nearest ‘parallel worlds’ to the actual world in which X is actually true.”
Problem here is the same as before. What we really want to say is that “If had occurred, then B would have occurred” in normal circumstances, or “If A were to occur, then B would occur” whenever background conditions are standard, where this provides us with the explanation that we want, and the explanation we want is a function of our interests and purposes. We cannot give an account of “normal” or “standard” without writing these man-made salience criteria into nature, which would make nature very much like a mind.
Alternative Reading of Counterfactual Approach: We would say “A is followed by B in all nearby possible worlds.” But, nearby possible worlds are possible worlds which are relevantly similar to our own, and, once again, we come up with such a relevant similarity metric. We do not find it in nature. And besides, possible worlds are themselves hardly physical entities.
“According to Kripke, the statue and the piece of clay are two objects, not one. The fact that the piece of clay has a modal property, namely the property ‘being a thing which could have been spherical in shape’, which the statue lacks (I assume this is not one of those contemporary statues) already proves the two objects cannot be identical, in Kripke’s view.”
One can try to define the essential properties of a type of thing as it is in itself in terms of its modal properties. This would allow the clay and the statue to be two different objects, since the clay, qua clay, can be spherical; but the statue, qua statue, cannot be spherical.
Main problem with this account:
“The difficulty is that Kripke individuates objects by their modal properties, by what they (essentially) could and could not be. Kripke’s ontology presupposes essentialism; it cannot be used to ground it. And modal properties are not, on the face of it, part of the materialist’s furniture of the world.”
To individuate objects through reference to their modal properties is to individuate objects through reference to what they could and could not be.
One last effort to make naturalistic sense of the semantic component of Metaphysical Realism: Functionalism
“The contemporary metaphysical materialist thinks about reference in the following way: the brain is a computer. Its computations involve representations. Some of these (perhaps all) are ‘propositional’: they resemble sentences in an internal lingua mentis. (They have been called ‘sentence-analogs’.) Some of them could be sentences in a public language, as when we engage in interior monolog. A person refers to something when, for example, the person thinks ‘the cat is on the mat’ (the sentence-analog is ‘subvocalized’) and the entire organism-cum-environment situation is such that the words ‘the cat’ in the particular, sentence-analog stand in a physical relation B (the relation of reference) to some cat and the words ‘the mat’ stand in the relation B to some mat.”
But what is this relation R? And what on earth could make anyone think it is a physical relation?
What is this functionalism here? How does it differ from a mere causal account?
The essence of functionalism is in the seeming possibility of multiple realizability.
Sentient computers or aliens made out of very different kinds of stuff than ourselves.
We don’t want to rule out the very possibility a priori. So, we say that the Martian and ourselves have something in common. My brain state and his rocky brain state are both implementations of a common functional state . . . where I might understand the functional property not just as a property of the organism in question, but of the entire organism/environment system.
Putnam has three objections to this proposal.
(A)The first one regards informativeness? He uses the example of chemical valences, where the valence of an element is the combining power of atoms. Similarly with theories of truth. What is Putnam’s point here?
(B)The second example involves disembodied spirits.
“Thirdly, the list is too specific. Reference is as ‘abstract’ as causation. In possible worlds which contain individual things or properties which are not physical (in the sense of ‘physical,’ not definable in terms of the fundamental magnitudes of the physics of the actual world), we could still refer: we could refer to disembodied minds, or to an emergent non-material property of Goodness, or to all sorts of things, in the appropriate worlds. But the relevant situations could not, by hypothesis, be completely described in terms of the fundamental magnitudes of the physics of our world. A definition of reference from which it followed that we could not refer to a non-physical magnitude if there were one is just wrong.”
(C)The really critical objection:
The crucial point is that functional properties are defined using the notions of cause and effect. This is no problem for Lewis; Lewis believes he can define cause and effect using counterfactuals, and, as already mentioned, he gives truth conditions for counterfactuals in terms of a primitive notion of ‘similarity of possible worlds’. Since he has a non-physical primitive in his system, he does not have to show that any of the notions he uses is physically definable. But the notion of ‘similarity of possible worlds’ is not one to which the materialist is entitled; and neither is he entitled to counterfactuals or to the notion of functional organization’.
Problem ends up being this. A functional state ends up being defined in terms of its causal role within the overall causal economy of the organism and environment in which it occurs.
We could only understand the notion of a functional state in terms of the notion of causal connection. But Putnam purports to have already shown that the notion of causal connection is not materialistic.
Without causation as a physical relation, we can’t define functional relations as physical, and thus we can’t define reference as physical. So, we can’t make sense of the correspondence relation.
Conclusion: Combination of materialism and essentialism is internally incoherent. Why? Because we can’t give an account of the essences of objects while remaining materialists.
In Particular, essences and intrinsic properties cannot be understand in terms of causation because causation isn’t physical.
What can we do? We can no longer make sense of the idea that objects have special essences which put them into contact with our thought. Nor can we suppose that there is a special non-sensory intuitive faculty. Both of these options reject materialism. At most we can become “pragmatists” or “internal realists.”
If all this is a failure, where do we go from there? One direction, the only direction I myself see as making sense, might be a species of pragmatism (although the word-‘pragmatism’ has always been so ill-understood that one despairs of rescuing the term), ‘internal’ realism: a realism which recognizes a difference between ‘p’ and ‘I think that p’, between being right, and merely thinking one is right without locating that objectivity in either transcendental correspondence or mere consensus. The other main direction—the one that does not make sense to me—is natural metaphysics, the tendency I have criticized here.”