Notes from Inductive Inference and It’s Natural Ground by Hilary Kornblith
Kornblith immediately establishes this book as a response to Quinean naturalised epistemology. He establishes that whilst Quine had set in motion an evolution of a naturalist epistemology, he could not be held directly responsible for presentations of naturalism in the contemporary context. Never-the-less, naturalism is referred to as the ‘Quinean project’.
“For me then the problem of induction is a problem about the world: a problem of how we, as we now are (by our present scientific lights), in a world we never made, should stand better than random or coin-tossing chances of coming out right when we predict by inductions which are based on our innate, scientifically unjustified similarity standard.” (Quine, Ontological Relativity and Other Essays)
note: think out what is meant exactly by similarity standard.
1.1 Epistemology as a Part of Natural Standard
Epistemology in general, and Quinean epistemology in particular, addresses two questions: “(1) What is the world that we may know it?; and (2) What are we that we may know the world?” (Kornblith, p2)
“Quine has argued that knowledge of the world is possible, in part, because the world is divided by nature into kinds. At the same time, our psychological processes are so shaped by evolutiion as to be sensitive to those very natural kinds.” (Kornblith, p2)
Inductive inference is framed, via evolutionary psychological dispositions, as dovetailing with our scientific account of the structure of nature. Therefore, our ‘native processes of belief acquisition are well adapted to revealing the structure of the world.’ (p2) Kornblith seeks to elaborate on and doubt this argument. He notes that Quine himself was spare in this argument.
“Evolutionary processes do not optimize for their environments, and, in many cases, processes which produce false beliefs will have tremendous survival value.” (p3) See also Stephen Stich
Kornblith still seeks to make the argument that our psychological processes (inductive inference) dovetail with the causal structure of the world. This fit between our psychological processes and the world is the mission of the descriptive philosophical approach of this book. Thus, inference will be comfirmed as a justified form of belief. This fit will be explained as a product of an evolutionary process. However, evolution is invoked as an after-thought to explain how the fit came about, rather than the fact that the fit is good. The fit itself is established independently of evolution.
For Quine, epistemology falls into place as a ‘chapter of psychology and hence of natural science.’ (Quine, Ontological Relativity and Other Essays) So what of the normative dimension of epistemological inquiry? Kornblith asserts that Quine’s approach is not at odds with normativity. Normative questions themselves fit into empirical inquiry. Kornblith refers to Goldman (1986) and divides the inquiry into two parts.
1) the account we provide of the conditions which a process of belief acquisition must meet if it is to account as normatively correct (independent of psychological inquiry)
2) the account of which processes meet these conditions
Kornblith portrays this approach as not sufficiently naturalistic enough for Quine, and chooses to side with Quine. He relates the question of our psychological processes to the holistic character of confirmation:
“Our account of the processes we ought to reason by thus cannot float free of our account of the range of processes which are in fact open to us. But this is just to say that our normative account must itself be influenced by the results of empirical investigation. We may not insulate any part of epistemology from empirical work.” (Kornblith, p5)
Quine, followed by Kornblith, propose an a posteriori account of epistemology. Kornblith asserts that foundationalism and the coherence theory do not provide any useful or accurate epistemic advise. These theories do not put us in any more accurate contact with the world. Hence, the Quinean project of naturalised epistemology promises to change this via the all-important dovetail between our best current theories of the nature of the world, and our best current psychological theories.
1.2 The Importance of Natural Kinds
“How then does the existence of natural kinds help to explain what it is about the world that makes inductive knowledge of it possible? The existence of natural kinds brings with it a certain causal structure. Only certain combinations of properties may be found together in a single individual, for not all combinations are causally possible or causally stable. Natural kinds involve causally stable combinations of properties residing together in an intimate relationship.” (Kornblith, p7)
If kinds are shown to be accurate descriptions of the world, then inferences made based on probability of properties across/within kinds are proven to be reliable.
“The causal structure of the world as exhibited in natural kinds thus provides the natural ground of inductive inference.” (Kornblith, p7)
It is also noted that Quine did not consider natural kinds as substantive enough to provide a proof, to be scientifically acceptable.
1.3 Our Psychological Constitution
What are we that we may know the world?
To follow the line of reasoning that our psychological character of inductive inference is a violation of the canons of good statistical inference (as is suggested by numerous studies highlighted by Kornblith) is to arrive a skeptical account of knowledge. The entire scientific project takes on deep contradictions. When science is turned on the world it gives us an accurate account of its features. When turned on us, it should then explain how we arrive at such an account. However, a feedback loop emerges as we tell ourselves that our cognitive equipment is not well suited to providing us with accuracy. How then can we confirm the reliability of any statement? One ‘escape route’ is to suggest an elite intellectual class, comprised mainly of scientists (which is now a defunct escape route given the fallibility of science after the ideological corruption of environmentalism). This also suggests a discontinuity between ordinary and scientific inference, which according to Kornblith, has not been proven to exist (this would be an interesting argument to take to historical events in scientific development; to what extent does ordinary inference provide the foundations for the advance of science since the enlightenment?)
Essentially Kornblith is optimistic. He seeks to confirm the existence of natural kinds, the reliability of ordinary inductive inference, and its grounding in the advance of science. Kornbliths picture also accounts for fallibility, allowing for errors, and a pattern to the errors of our inductive inferences. This is a descriptive account of inductive knowledge. The account sets up a standard by which its success or failure can be measured, thereby providing also a normative element to naturalised epistemology.