Copyright © Norman Swartz 1997, 1998
Original version: September 24, 1997
Department of Philosophy
Simon Fraser University

Epistemic Modes

"P is knowable experientially" =df "It is humanly possible to know P either by direct appeal to experience or by valid inference from propositions one or more of which is known experientially".

  1. "experiential" and "a posteriori" are synonyms.

  2. "experiential" does not mean either "experimental" or "empirical".

  3. The phrase "valid inference" is not to be taken to mean "deductively valid inference". It is to be understand to mean "deductively or inductively valid inference".

  4. Be careful not to read into this definition more than it states. In particular, do not assume that one's knowledge must derive from one's own experience. Indeed a very great deal of our knowledge is obtained from other persons – persons who were present at events at which we ourselves were not, persons who have more acute senses than the rest of us, etc. Knowledge need not be firsthand: it can be second-, third-, or ...-hand.

  5. The definition (above) is recursive, as is the next. See "Definitions, Dictionaries, and Meanings".

"P is knowable ratiocinatively" =df "It is humanly possible to know P by appeal to reason (e.g. by an analysis of concepts) or by valid inference from propositions which themselves are known by appeal to reason."

  1. Note that the parenthetical phrase begins "e.g." The meaning would change dramatically if it were instead "i.e."

  2. Mathematicians will insist that "valid" be taken to mean "deductively valid" for the case of mathematical logic.
The two definitions above – although certainly standard and familiar – have some peculiar and unexpected, indeed perhaps paradoxical, consequences. If you are interested in learning what they are, see "Getting from P to Q: Valid Inference and Heuristics", by Norman Swartz, in Dialogue, vol. XXXII (1993), pp. 527-540.
"P is knowable empirically" =df "It is humanly possible to know P only experientially."

"P is knowable a priori" =df "It is humanly possible to know P other than experientially [i.e. without experience]."


(Note "" stands for "implies".)
  1. empirical experiential (but not conversely)

  2. ratiocinative a priori (but not conversely)

RATIONALISM is the philosophical theory that there are at least some contingent propositions that can be known a priori.

For example, (it is sometimes alleged that) it stands to reason that (i.e. it may be known by reason alone [ratiocinatively], and hence a priori, that)
  • Heavier objects fall faster than light ones.

  • If there are 5 persons in a room and none leave, then if 6 persons enter that room, there will be 11 persons in that room.

  • If a table measures 110cm when measured left-to-right, then it will measure 110cm when measured right-to-left.

  • If a train is traveling 30km/h due east, and if a passenger is walking in that train from the rear towards the engine at 5km/h with respect to the train, then that person is traveling east at 35km/h with respect to the ground.

  • To get greater fuel economy in a car, one should turn off the air-conditioning and open the windows.

  • Everything that is colored is extended.

  • Nothing can be in two places at the same times (or, more exactly, nothing can have its entire spatial extent present at the same time in two unconnected spatial locations).

  • Nothing can be both red and blue all over at the same time in the same place.

EMPIRICISM (for present purposes) is the philosophical theory that contingent propositions are never knowable a priori, i.e. that if a contingent proposition is knowable at all, then it is knowable empirically, i.e. only by experience.

Empiricists will deny of all the claims immediately above, and indeed of any and all others, that they are both contingent and knowable a priori. Of any specific claim that a Rationalist might put forward as a candidate for being both contingent and knowable a priori, an empiricist will argue that:
  1. the claim is not contingent, but is noncontingent; or

  2. the claim, although contingent, is not knowable a priori; or

  3. the claim is not knowable (by any means – either empirically or a priori).

Twentieth- and twenty-first-century empiricism needs to be distinguished from eighteenth-century, or, as it is often called, British Empiricism. Eighteenth-century empiricism claimed two things:
  1. (The equivalent of) what has been said of Empiricism above; and

  2. All our concepts are derived through sensory experience. There are no innate concepts.
The former of these two theses is (now known as) Judgment Empiricism; the latter, Concept Empiricism. All twentieth-century empiricists are judgment empiricists; some, but not all (perhaps even not many) are concept empiricists.

If you find that an author cursorily dismisses 'Empiricism' (or 'British Empiricism'), he/she is likely railing against Concept Empiricism. But you'll have to examine the context of the argument to be sure.

Return/transfer to Norman Swartz's Philosophical Notes

Return/transfer to Norman Swartz's Home Page