Monday, July 27, 2009

Rand on Concepts, Relation to Induction (Part 1)

In her Introduction to Objectivist Epistemology (ItOE), Rand presents a theory of concepts, which describes what concepts are (as opposed to what they are not) and how they are properly formed (and how they are improperly formed). A concept, Rand maintains, is a “mental integration of two or more units possessing the same distinguishing characteristic(s), with their particular measurements omitted.” (ItOE, 2nd Edition, p. 13)

(Some elements of that definition may be familiar to anyone who has looked up the word “concept” in a dictionary. The online Merriam-Webster Dictionary, for instance, states that a concept is “an abstract or generic idea generalized from particular instances,” which is nearly the same as Rand’s “mental integration of two or more units.” What is distinctive and deserves study within her theory is her view of similarity and abstraction as measurement-omission, the process by which the concept is formed.)

There is quite a bit I want to discuss on the subject of concepts. One example being the fact that Nietzsche and Schopenhauer (among others) were mistaken on the nature of concepts due to their not having the concept “measurement-omission.”

For now, I want to compare Rand’s view on concepts with what I think is true regarding induction.

1) Concepts versus complete enumeration:

Given Rand’s definition of concepts, it is not difficult to guess that she would deny that concepts require complete enumeration of instances to be formed, or require a statement of the exact number of instances. (In order to form the concept “bottle,” for example, one would not need to know about all instances of bottles, or even know how many of them exist.) As long as there is more than one instance or unit of something, Rand seems to be saying that a concept could be formed in relation to those instances.

On this subject, she says that a “concept is not formed by observing every concrete subsumed under it, and does not specify the number of such concretes.” (Ibid. p. 17)

Concepts, then, are the result of ampliative reasoning, of gaining general knowledge which extends beyond the knowledge gained about particular things one has encountered or experienced. Rather than only knowing about the objects one observes day-to-day, concepts allow us to know facts about countless unobserved things (a distant galaxy, for instance) and higher-level conceptual information (Newton’s theory of Tides) that would otherwise be unavailable to us. (Just as it is unavailable to all other known living things besides humans, that is, to beings who lack concepts.) As Dr. Leonard Peikoff puts it in his work “Objectivism: the Philosophy of Ayn Rand,” concepts allow us to know “facts pertaining to all trees, every pond and drop of water, the universal nature of man.” (p. 73)

Similarly, this ampliative feature is one of the requirements I consider for any genuine induction.

Induction as identification of cause, not complete enumeration:

A concept, as I’ve said, vastly expands what we can know about the world, by serving as a sort of filing or classificatory system in which we input and retain information. A properly formed induction is also a classificatory system, one that allows us to retain causal relationships and apply them to particulars, including ones we‘ve never encountered before.

When I consider the inductive generalization “all people are mortal,” it isn’t my copious, enumerated observations of people (even if complete) which demonstrates that we are, in fact, mortal--it’s identifying the cause of mortality. Once I’ve identified the cause of mortality, that is, realized that mortality is “being vulnerable to the cessation of life,” I will know what can and cannot count as a mortal being, meaning I have a valid concept of “mortality.” Assuming I already know that all people are living things, demonstrating that we are all mortal becomes a matter of relating this knowledge to our concepts of “life” and “people.”

A related advantage of causal induction to enumerative induction is that the first has an answer to proposed counter-examples.

Say, someone presents a person who appears, by all lights, to be immortal. The induction by enumeration would have no way of countering this, and would be instantly refuted--exactly what Francis Bacon pointed out as a chief weakness of the enumerative induction method (or, rather, "anti-method").

The defender of causal induction, on the other hand, can simply respond "that immortal thing isn't a person," or "that person isn't immortal; she's mortal." Instead of instantly conceding, the causal inductivist is in a position to request that his opponent retrace the steps by which he came to the idea that this person was immortal, that is, to identify the cause of immortality.

(I’ll have more to say about the causal view of induction and the close connection between induction and concepts when I discuss Bacon in more detail.)

Part II will probably discuss abstraction as it is found both in Rand's theory and in my understanding of induction.

No comments:

Post a Comment