# Thought, Fact, and Reference: The Origins and Ontology of Logical Atomism

Herbert Hochberg
Edition: NED - New edition
Pages: 512
https://www.jstor.org/stable/10.5749/j.ctttt490

Export Selected Citations Export to NoodleTools Export to RefWorks Export to EasyBib Export a RIS file (For EndNote, ProCite, Reference Manager, Zotero, Mendeley...) Export a Text file (For BibTex)
1. Front Matter
(pp. i-vi)
2. Preface
(pp. vii-viii)
H. H.
3. Introduction
(pp. ix-xii)

Moore and Russell turned a significant segment of contemporary philosophy onto an analytic and linguistic course with their early attack on idealism. Together with Wittgenstein’s early contributions, doctrines of the Cambridge philosophers formed what Russell called “Logical Atomism.” From that period to the present, philosophers have constantly heard of and read about “analysis,” “what can be shown and not said,” “logical and grammatical form,” “ordinary meaning,” “simple and unanalyzable,” “knowledge by description and knowledge by acquaintance,” “atomic sentence,” “denoting expression,” and “the present King of France.” In spite of the varied criticism and controversy spawned by the doctrines of the...

(pp. xiii-2)
5. I The Analysis of Perception
(pp. 3-29)

Moore’s most systematic attempt to handle the problems of intentionality occurs in connection with his analysis of perception inSome Main Problems of Philosophy. He begins the book with the following account of belief.¹ Suppose one believes that an object is white. Then, there is a belief which is a particularactoccurring at a specific time. In one sense, the phrase “a belief” stands for a particular occurrence of what we might call a generic kind,believing, just as the page upon which these words appear is a specific individual of the generic kindpage. As philosophers traditionally speak...

6. II Idealism, Realism, and Common Sense
(pp. 30-52)

The relatively simple and classic argument that Moore offers against the phenomenalist in “The Refutation of Idealism” goes something like the following.¹ We distinguish the act,${m_{\text{a}}}$, from its object, say${s_1}$, in a case of an awareness of the datum. As in the later pattern ofSome Main Problems of Philosophy, we have an act, an object, and a relation between them. The idealist claims that “To be is to be perceived.” Therefore, he claims that from

(a)${s_1}$exists

it follows that

(b)${s_1}$is perceived,

(b) may be construed as,

(c)$(\exists {\text{x}})\;{\text{DA(x,}}\;{s_1})$

where ‘DA’ indicates the relation...

7. III Thought and Belief
(pp. 53-86)

Moore’s account of perception depends on the discussion of belief and the introduction of propositional entities. A thought, say$m{}_1$, that occurs at a particular time has the content it does because of the proposition it directly apprehends. Propositions give content to thoughts. Such thoughts would also have certain generic properties. For example, a thought that$a{}_1$is W is different from a doubt. We can consider a set of predicates, ‘T’, ‘D’, etc., to stand for generic properties that we would ordinarily indicate by ‘thinking’, ‘doubting’, etc. (or ‘is a thought’, ‘is a doubt’, etc.). Thus, when there is...

8. IV Moore and Bradley on Particulars, Predicates, and Predication
(pp. 87-121)

Bergmann’s notion of a bare particular is at the heart of both his version of realism and his analysis of mind. Moore’s concept of a particular was similarly crucial, not only in the formulation of his views regarding belief and perception, but for Russell’s analyses of thought and reference. Moreover, understanding Moore’s early analyses will help us to understand both Russell’s very early views, which provided a foil for the emergence of his theory of descriptions, and Frege, whom Russell explicitly attacked. As we shall see, Frege and Moore developed quite similar themes. Finally, Moore’s views on particulars provide supporting...

9. V Names, Individual Concepts, and Ontological Reduction
(pp. 122-146)

Throughout the discussion of relations and individuation, two issues have been left unresolved. One concerns the assumption that properties or relations, to individuate, must be “parts” or “constituents” of objects. The second concerns the assumption that the proponent of relational individuation may not use proper names of objects in a perspicuous schema, but must construct definite descriptions for them.

The use of the phrase ‘part of’ lends a specious plausibility to the argument against relational individuation in that one implicitly uses the idea that colors and shapes, being located in space “where” the object is, may be spoken of as...

10. VI Frege’s Account of Reference and Thought
(pp. 147-169)

Frege resolved the problem about the content of thoughts, expressed by sentences employing names or labels, by holding that names, like definite descriptions,expressed sensesthat were constituents of propositions expressed by sentences. Thus, names and descriptions stood in two relations to entities: theyreferredto objects andexpressedsenses. Sentences, likewise, referred to truth values and expressed propositions or thoughts. Predicates, logical connectives, and quantifiers were “function signs” thatstood for“functions” or “concepts.” There has long been a controversy over the interpretation of his claims regarding predicates and function signs, and their connection with his views about meanings...

11. VII Russell’s Critique of Frege and the Origin of the Theory of Descriptions
(pp. 170-197)

One of the most abused and misunderstood arguments in the most celebrated philosophical paper of the twentieth century opens with the phrase “The relation of the meaning to the denotation involves certain rather curious difficulties” and ends with the claim “Thus the point of view in question must be abandoned.”¹ Here, I am not primarily concerned with a line by line exposition of the discussion that the passage requires and deserves. That task has been performed elsewhere.² Rather, I am concerned with the merit of the argument, actually several arguments, as directed against a Fregean type of analysis. Therefore, I...

12. VIII Descriptions, Substitution, and Intentional Contexts
(pp. 198-230)

Russell made use of a nonextensional context in his presentation and defense of the theory of descriptions in “On Denoting.” There, he claimed that his theory enabled us to solve the puzzle of why the argument

(A1) 1. George IV wished to know whether Scott was the author ofWaverley.

2. Scott is the author ofWaverley. Therefore,

3. George IV wished to know whether Scott was Scott.¹

is not valid, while retaining the rule that two names for the same thing can be interchanged in any context where they areused, as opposed to mentioned. Thus, his theory of...

13. IX Existence, Predicates, and Properties
(pp. 231-270)

Russell declared existence not to be a property. The claim, subject to some doubts of Moore’s, was fairly common doctrine until concerns about non-naming names, so-called free logics, and “possible worlds” surfaced to reraise Moore’s classical concerns about assertions of existence and nonexistence. Just as Russell’s claim that existence was not a property was reflected by the lack of a predicate for such a property, in a standardPrincipia-type schema, and the consequent use of the existential quantifier to make existential claims, current claims have to do with the reintroduction of a predicate, say ‘E’, into such schemata. The claim...

14. X Facts and Possibilities
(pp. 271-308)

Like properties, facts have long been the target of philosophical attacks. These have ranged from purported paradoxes, like the fallacy we considered earlier, to the profound and genuine problems that Bradley pointed out. The latter we shall consider in the next chapters. Somewhere in between lies the kind of criticism that considers the appeal to facts to be empty and trivial, just as some nominalists see the reference to properties to be empty of explanatory value. The idea is that to say that a fact is the ground of truth for an atomic sentence is not really to say anything....

15. XI Russell’s Theory of Judgment and Sellars’s Critique of It
(pp. 309-346)

InThe Problems of Philosophyof 1912, Russell proposed to construe a sentence like

(S1) Othello believes that Desdemona loves Cassio

as expressing a four-term relation, say Bel⁴, holding among Othello, Desdemona, Cassio, andloves as terms. Thus,

(S2) Bel⁴ (Othello, Desdemona, loves, Cassio)

would more perspicuously exhibit the form of thefactstated by (S1). He advocated such a view because he held that to take (S1) to express a two-term relation between Othello and a proposition (that Desdemona loves Cassio) or other complex (Desdemona’s love for Cassio) would be to introduce entities corresponding tofalsesentences. This would...

16. XII The Structure of Thought: Part I
(pp. 347-379)

InThe Analysis of Mind, Russell abandoned mental acts on much the same grounds he was to use to decisively “exorcise” bare particulars inAn Inquiry into Meaning and Truth. Both kinds of things conflicted with his principle of acquaintance and, hence, were not to be acknowledged by one with empiricist scruples.

The first criticism I have to make is that theactseems unnecessary and fictitious. The occurrence of the content of a thought constitutes the occurrence of the thought. Empirically, I cannot discover anything corresponding to the supposed act; and theoretically I cannot see that it is indispensable....

17. XIII The Structure of Thought: Part II
(pp. 380-413)

Our pattern permits the derivation of

$({{\text{B}}_{1'}})$$\ulcorner {\text{Desdemona}}\;{\text{loves}}\;{\text{Cassius}} \urcorner$M Desdemona loves Cassio.

Hence, we considered permitting the substitution ofnameswithin the corner quotes insuchcases, i.e., in contexts like ‘$\ulcorner \ldots \urcorner$M . . . ’. The question then arises about substitution within the corner quotes in all contexts. An obvious problematic case is provided by identities like:

┌Desdemona loves Cassio┐ = ┌Desdemona loves Cassio┐

and true denials of an identity such as;

┌Desdemona loves Cassio┐ ≠ ┌Desdemona loves Cassius┐;

since, although they are true, to permit the substitution of names on the basis of ‘Cassio = Cassius’ would...

18. XIV Logic, Fact and Belief
(pp. 414-443)

Russell argued that negative facts exist.¹ The argument was quite clear, but the implicit criteria used never became explicitly recognized or stated. If they had, I believe that he would have been less hesitant about rejecting disjunctive facts (and, by implication, other molecular facts) and that he would not have accepted universal and existential facts in the way he did.² We have already considered one line of argument for negative facts in the arguments for possible facts. I have claimed that, for certain issues, the appeal to negative and positive facts does not significantly differ from the appeal to possible...

19. XV Difference, Existence, and Universality
(pp. 444-454)

As I noted in an earlier chapter, we need not recognize existential facts like ‘$(\exists {\text{x)Wx}}$’. Given that ‘Wa’ is true, the fact it indicates suffices to ground the truth of ‘$(\exists {\text{x)Wx}}$’. Just as the rule for conjunction shows we need not recognize conjunctive facts, the rule of existential generalization does the same for facts like ‘$(\exists {\text{x)Wx}}$’. But, as Russell observed, we do not have

$\frac{{{\text{W}}{a_1}\;\& \;{\text{W}}{a_2}\& \; \ldots \;\& \;{\text{W}}{a_{\text{n}}}}}{{({\text{x)Wx}}}}$

as a valid argument form. He thought we must recognize the universal fact that${a_1},\;{a_2},\; \ldots ,\;{a_{\text{n}}}$wereall the objects. With such a claim as afurtherpremise, we can deduce ‘$({\text{x)Wx}}$’ from the appropriate conjunction...

20. Notes
(pp. 455-482)
21. Name Index
(pp. 483-486)
22. Subject Index
(pp. 487-489)