Mysticism and Logic, Chapter X

“Knowledge By Acquaintance and Knowledge By Description,” pages 209-232

What if we know some general fact – the candidate with the most votes will win – without knowing who that candidate is? In such cases, what do we really know from the general description of the winner alone?

Being “acquainted” with something is to “have a direct cognitive relation [p. 209]” with it: in a sense, you have met that thing. “Acquaintance” requires both a subject (you) and an object (the thing) – but the object doesn’t have to be near at hand for you to maintain an acquaintance with it.

People are acquainted with their sense-data, like smells or sounds. (Sense-data have played a major role in the later stages of Mysticism and Logic, beginning in Chapter VII.) A sense-datum can itself be rather complex, with multiple parts connected spatially.

In this regard, consider consciousness. We can be aware of some object we see or some desire we have, while simultaneously being aware that it is us who is seeing the object or having the desire. But can we be aware of ourselves in isolation, without a connection to other elements of awareness? To return to our previous terminology, we are acquainted with the dual construct “Self-acquainted with A” (like “self aware of some object”) but also know “I am acquainted with A”. But what of this “I”? Is it known only by description – like being ‘the subject-term in awarenesses of which I am aware [p. 212]’ – or as a direct acquaintance? (So much for the consciousness detour.)

Awareness of sense-data is an example of awareness of particulars; we can also be aware of concepts more general, universals. For instance, we can be aware of many particular shades of yellow, but also aware of the concept of yellow, a universal. Statements such as “yellow is different from blue” are meaningful – we can be aware of the difference between yellow and blue, the relation of these two universals. And because we can aptly deduce from a few cases (A before B and B before C implies A before C), we must be able to know the universal relation “before” beyond just knowing some cases.

OK, we are aware of particulars (this and that, and properties (yellow) of this or that, and relations between them, this before that) and of universals. All non-particular objects are universals. The particular/universal dichotomy is akin to concrete/abstract.

While we are acquainted with sense-data, we are not acquainted with physical objects or with the minds of others. These we know by description, not by acquaintance. These descriptions can be definite (“the” object) or ambiguous (“an” object).

[Russell then (page 214) constrains himself to speaking of definite descriptions.] What if we know that there is an object that fits a definite description, but we are not acquainted with any such object? What is the nature of our knowledge? We know there was a man in the iron mask, though we don’t know his identity. We do not know who will get the most votes even though we know that the winner will be the person who receives the most votes: we have “'merely descriptive knowledge [page 215].'”

Someone who knew Bismarck (but was not Bismarck) is acquainted with sense data about Bismarck’s body and mind, but the body and mind themselves are known only by description, as reflecting the sense data. One might have various descriptions of Bismarck, but they all refer to the same person. For those who do not know Bismarck, our descriptions of him are presumably some bits of historical knowledge, such as “’the first Chancellor of the German Empire [p. 217].’” These predominantly abstract words, if we are to be sure we are dealing with an accurate description of Bismarck that allows us to form some judgment, must eventually be connected with “a particular with which we are acquainted [p. 217]”. If we claim that this first German Chancellor was gifted in diplomacy, we can only be confident in our claim if we reference some sort of testimony or related particular we are acquainted with. (Bismarck’s diplomatic acuity we can know to be true, even though we are not directly acquainted with him, and thus don’t know the proposition concerning his skill. That is, we communicate with propositions that we can know to be true without being acquainted with the proposition.) Place names like London or the solar system similarly implicate acquainted particulars. But logic applies to things beyond the actual existing, and requires no connection to specific particulars.

“The fundamental epistemological principle in the analysis of propositions containing descriptions is this: Every proposition which we can understand must be composed wholly of constituents with which we are acquainted [p. 219].” [Russell goes on to support this principle, citing his own Philosophical Essays and endorsing (though without threatening the principle) a criticism lodged by Wittgenstein. How can we make a judgment without knowledge of what it is we are judging? Russell also dismisses the notion that we don’t know objects of judgments such as Julius Caesar, but only know some abstract mental idea of Caesar. This (misguided) notion, he believes, leads to the view that we never know something, only ideas about something – but why not go further, and say that we only know ideas of those ideas, ad infinitum?]

“Meaning” and “denotation” frequently are distinguished; for instance, in the phrase “featherless bipeds,” the denotation is humans, though the meaning is more complex. A statement such as “’men are the same as featherless bipeds [p. 224]’” (implicitly) presents an assertion that the denotations are the same. But Russell claims that actual propositions do not concern denotations, excepting the case of proper names. We can know the truth of a proposition without being acquainted with the denotation. [Russell is well known for his work on denotation, and some of his previous efforts he follows here.]

Consider (p. 226) the phrase “Scott is Sir Walter” – these are two names for the same person. But if Sir Walter Scott often was called “the author of Waverley,” that would not make him the author – the author of Waverley is whoever wrote Waverley. The authorship is a fact that goes beyond mere names. [Russell, getting into what for me are deep philosophical weeds, cites (p. 228n) a fuller discussion in Volume I of Principia Mathematica.]

These developments allow the denotation of a phrase like “the author of Waverley” to be defined. “If we know that the proposition ‘a is the so-and-so’ is true, we call a the denotation of the phrase ‘the so-and-so’ [p. 229].” When we assert a proposition about “the so-and-so,” its truth or falsity generally can be assessed by checking its truth with respect to a. This useful feature of denotations hinders a proper logical understanding, as it leads us to mistake the denotation for the description, and to believe (again, incorrectly) that propositions concern denotations: to understand propositions, “we need acquaintance with the constituents of the description, but do not need acquaintance with its denotation [p. 230].”

 To recapitulate: Knowledge by acquaintance can be distinguished from knowledge by description. It is sense-data, universals, and possibly ourselves with which acquaintance exists – we are not acquainted with other people and objects. We can have descriptive knowledge of an object, say, if we know that it is the sole object having a property (or properties) with which we are acquainted (page 231). Propositions that we can understand must only directly relate to things we are acquainted with. A judgment (connected to a mind) must also concern only objects of acquaintance to that mind. Thus when we see descriptive phrases in propositions, “the objects denoted by such phrases are not constituents of judgments in which such phrases occur (unless these objects are explicitly mentioned) [p. 231-232].” When we assert the proposition that the author of Waverley is the same as the author of Marmion, Sir Walter Scott plays no role in our judgment of that proposition.

[End Matter: My copy of Mysticism and Logic proper ends on p. 232, but there are some more printed pages: two pages of Index; a page with the elaborate St. George-slaying-dragon emblem of the publisher (George Allen & Unwin); and five pages advertising fifteen other books by Russell for sale, with prices – the “Fifth Impression” of Marriage and Morals, for instance, apparently can be had for 8s. 6d.