Any corrections, disputations, or other comments from participants here, I welcome. In the last paragraph of this piece, I mention further discernments yet to be drawn out in the scope of this topic. These I am working on at present using works of a number of modern scholars on Aristotle, and of course, his own text. The work presented below is to become a portion of the book I’m writing (probably in a second chapter, following one on Nietzsche/Rand). My target audience for the book would be anyone with the ability and interest to read and absorb such a composition as Galt’s speech. Surely, additional explanation for the stretch of text below, such as the basics of syllogism, will be needed, at least in a note for the book version. In the end, I don’t want to suppose that all readers in the target audience have had a first course in logic. I realize that the citations would not be of interest to some such readers, but in this blog, as in the book, I expect scholarly readers as well. And anyway, I require that aspect for my own interest and roadmap.
On Logic, Aristotle and Rand
Rand set out:
To exist is to be something, as distinguished from the nothing of non-existence, it is to be an entity of a specific nature made of specific attributes. Centuries ago, the man who was—no matter his errors—the greatest of your philosophers, has stated the formula defining the concept of existence and the rule of all knowledge: A is A. A thing is itself. You have never grasped the meaning of his statement. I am here to complete it: Existence is Identity, Consciousness is Identification. (AS 1016)
By “greatest of your philosophers,” Rand meant Aristotle. Unlike moderns such as Leibniz, Baumgarten, Kant, or Rand, Aristotle did not connect a “law of identity,” in so many words, with his principle of noncontradiction. Aristotle also did not connect the law of identity that speaks to the distinctive natures of things with a formula such as “A is A” or “A thing is itself.” Aristotle would say “A thing is itself” is nearly empty and useless, and he would not connect that proposition to “A thing is something specifically,” which he thought substantive and important.
Aristotle observed that saying “Man is man” or “The musical is the musical” and so forth are all occasions of affirming that a thing is itself. Someone who had said “Each thing is itself” might have meant “Each thing is inseparable from itself; and its being one just meant this” (Metaph. 1041a1). This meaning in A is A states a truth of all existing things—an existing thing is one, one with itself—and that is, I’d say, sufficient for some tare weight of objective meaningfulness in the statement. That much of A is A is a background assumption in all other meaningful statements.
In mathematics if we can show that an equation can be reduced to the equation 1=1 or 2=2 or sinq=sinq, and so forth, we have proven that the initial equation (apart from any physical application) is true. That is a usefulness of “A thing is itself.” In science one aim (and thrill) is discovery of A is A sleepers. This has been accomplished through the join of observation, mathematics, and induction. Examples are the discovery that the evening star is the morning star and that light is electromagnetic radiation (within a certain range of radiation frequency).
For Leibniz “the primary impossibility is this: A is not A; just as the primary necessity in propositions is this: A is A” (1678, 187). The identity of which Leibniz speaks as a basis of logical necessity is the identity of sameness. A demonstration establishes a sameness between the subject and predicate in the conclusion. It shows that the conclusion’s predicate is contained in the conclusion’s subject. Among the premises could be observations or intellectual truths. The demonstration proceeds by recognizing definitions and by substitution. For effective use in proofs, definitions must not contain contradictions, manifest or concealed. It is not enough that we understand what we say in a definition, for it can still be the case that our definition is of something impossible. Natures of things are implicated in deductive proofs by the observations, intellectual truths, and definitions employed in the demonstration.
Loemker writes of Leibniz: “Contradiction, . . . or the principle of impossibility, is implied in identity, and the two are opposite aspects of the same law, which Leibniz sometimes calls the basic law of being” (1969, 24). In Leibniz’ view, the law of identity entails that predicates of affirmative propositions are contained in their subjects. As with Aristotle, with Leibniz the primary form of being is substance. Identity-containments by subjects of their predicates record the relation of substance to its modifications. Such identity-containments look rather like Rand’s identity of character, or nature.
Rand did not continue with Aristotle’s central concept of substance, rather, she made existence most fundamental and made natured entity the primary form of existence. Entity, not substance, takes the role of bearing attributes and actions. Rand’s conception that “logic is the art of non-contradictory identification” and that “logic rests on the axiom existence exists” embed logic in her fundamental metaphysics: Existence exists and is identity; consciousness is of existence and is identification. In amplification of her compact statement “Existence is identity,” Rand goes on to say that the law of identity (and lack of contradiction) applies to objects, to attributes, to actions, and to their compositions into larger wholes. In Rand’s metaphysics, identity as to nature is tied at the most basic level to identity as self-sameness. For Leibniz identity as self-sameness is the deeper reality of the two. Not so for Rand.
Rand shared with Leibniz the view that the principle of noncontradiction rests on the law of identity. In the 1960’s lectures Basic Principles of Objectivism, Nathaniel Branden held forth and explained Rand’s idea that the law of identity is the basic principle of metaphysics and of epistemology.
The three . . . laws of logic are: The Law of Identity, the Law of Contradiction, and the Law of Excluded Middle. The last two are merely corollaries or restatements of the first.
[The law of identity] is the link between the two sciences, the bridge between existence and consciousness, between reality and knowledge.
As a principle of metaphysics, the Law of Identity tells us that everything which is, is what it is. As a principle of epistemology, it tells us that contradictions cannot exist, that a thing cannot be A and not-A. (Branden c. 1968, 66–67)
Those relations on identity and noncontradiction were also presented by Leonard Peikoff in his 1972 lectures on the history of philosophy, where he indicated historical philosophic puzzles resolved by these Randian conceptions. In Objectivism: The Philosophy of Ayn Rand, Peikoff writes:
The law of identity acts as a bridge linking existence and consciousness, or metaphysics and epistemology. The law acts as a bridge in a second respect also. The law defines the basic rule of method required for a conceptual consciousness to achieve its task. In this regard, the law tells man: identifications must be noncontradictory.
. . . Aristotle’s law of contradiction states . . . nothing can be A and non-A at the same time and in the same respect. This is not a different fact from the law of identity. It is a corollary of the latter, a restatement of it for the purpose of guiding human cognition. (1991, 118–19)
Rand’s sense of identity basing noncontradiction goes beyond Leibniz to include natures at the most fundamental level of identity. Identities of nature or character are not modifications of existence. Existence is identity, not only identity of sameness, but identity of character. There are no existents without both of those aspects of identity.
There are occasions in which one could say “Man is man” by way of stressing that some stunning actions of man are among human capabilities, parts of human nature. Seeing the first man walk on the moon, one might say “Man is man.” Seeing Romeo take his life, one might say “Man is man.” The reader of Atlas Shrugged finds Ayn Rand proclaiming “Man is man” with a meaning along these lines. Her proclamation was to stress that, notwithstanding his freedom of mind, man has a definite nature, that he is nothing but man, and that he is one.
In Rand’s fundamental Existence is identity, the identity of an existent includes its that/which and its what. Rand states her finer structure for the law of identity as follows:
Whatever you choose to consider, be it an object, an attribute, or an action, the law of identity remains the same. A leaf cannot be a stone at the same time, it cannot be all red and all green at the same time, it cannot freeze and burn at the same time. A is A.
. . .
A contradiction cannot exist. An atom is itself, and so is the universe; neither can contradict its own identity; nor can a part contradict the whole. (AS 1016)
Aristotle was the founder of logic, and his great contribution thereto was his theory of correct inference, which is largely his theory of the syllogism. Though he did not realize it, the formula “A is A” in the form “Every A is A” can be used to extend the kingdom of the syllogism. By about 1240, Robert Kilwardly was using “Every A is A” to show conversions such as the inference “No A is B” from the premise “No B is A” can be licensed by syllogism. Aristotle had taken these conversions, like the first-figure syllogistic inferences, to be obviously valid and not derivable. Aristotle takes first-figure syllogisms to be obviously valid and the paragons of necessary consequence. The mere statement of these syllogisms makes evident their necessary consequents. Using conversions as additional premises, Aristotle shows that all syllogisms not first-figure can be reduced to first-figure ones. Their validity is thereby established, by the obvious validity of the first-figure ones and by (what he took to be) the irreducible obvious validity of the conversions.
We find Leibniz, four centuries after Kilwardly, illustrating the utility of Some A is A for concluding Some A is B from All A is B via a syllogism, third mood of the first figure. “I offer these examples . . . to show that identities do indeed have a use and that no truth, however slight it may seem, is completely barren; on the contrary, . . . these identities contain the foundations for all the rest” (Leibniz 1679, 226; see also 1705, 362–63).
There are places in which Aristotle connects “A thing is something specifically” or “A thing is what it is” with the principle of noncontradiction: “The same attribute cannot at the same time belong and not belong to the same subject in the same respect” (Metaph. 1005b19–20). Though not given the pride of place given it by Rand, there is some recognition that Existence is identity in Aristotle: “If all contradictories are true of the same subject at the same time, evidently all things will be one . . . . And thus we get the doctrine of Anaxagoras, that all things are mixed together; so that nothing exists” (1007b19–26). Aristotle realized too that any existent not only is, but is a what.
Rand acknowledges the greatness of Aristotle particularly for his laws of logic, as they are called in elementary logic texts of today and the last few centuries: the laws of noncontradiction, excluded middle, and identity. Those are important principles of logic, though, as we have seen, Aristotle was not securely on board with that last one. It is not clear that Rand was cognizant of the even greater importance for logic of the theory of correct inference that Aristotle invented with his theory of syllogism.
The tremendous importance of the laws of noncontradiction, excluded middle, and identity, in Aristotle’s sense of them or in Rand’s, is importance for metaphysics, thence all appropriate thinking on account of that metaphysical structure. Rand praised Aristotle also for his identification of “the means of human knowledge.” That was in a postscript to her Atlas Shrugged. Presumably, this point of praise refers not only to the role of the three laws of what may be called metaphysical logic, but to Aristotle’s general picture of how we obtain knowledge by reasoning on sensory experience, not by mentally contacting a transcendent platonic realm of forms. Rand writes also that Aristotle’s “incomparable achievement lay in the fact that he defined the basic principles of a rational view of existence and of man’s consciousness: that there is only one reality, the one which man perceives—that it exists as an objective absolute . . .” (1961, 22).
This particle has the set of properties A.
Particles having the set of properties A are electrons.
Therefore, this particle is an electron.
That is a case of a first-figure syllogism, or anyway, excluding singular terms from syllogisms, it is a close relative of that first-figure, third-mood syllogism: Some a’s are b, and all b’s are c; therefore, some a’s are c. On account of a syllogistic inference such as my electron one, Rand could sensibly say “the process of subsuming new instances under a known concept is, in essence, a process of deduction” (ITOE 28). This deduction looks every bit like a case of identification, based on Rand’s kind-sense of identity, though without reliance on the principle of noncontradiction. So I’d say Rand got the genus right, though the differentia wrong, when she defined logic as the art of noncontradictory identification. Perhaps “the art of perfectly truth-preserving inferential identification” would be a better definition of deductive logic devolving from Rand’s conception that logic is slave of existence, that existence is identity, and that consciousness is identification.
The inferences of first-figure syllogisms are, I maintain, licensed directly by identity alone, in Rand’s full sense of identity, and without recourse to noncontradiciton. Nathaniel Branden and Leonard Peikoff erred in trying to support Rand’s definition of logic, with its differentia of the noncontradictory, by appeal to noncontradiction rather than directly to identity as basis of the inference in a certain first-figure syllogism. It is the inference-form of my electron example, but in the familiar case: Socrates is a man, all men are mortal, and therefore, Socrates is mortal. They rightly point out that denial of this inference would lead to contradiction, but that is not to the point of first basis: One already knows that these first-figure inferences are valid, that their conclusions necessarily follow, just as Aristotle observed.
Rand took thinking and logical inference to be volitional cognitions. “To think is an act of choice. . . . The connections of logic are not made by instinct” (AS 1012). Logical inference is consciously directed, in Rand’s view, and that seems right to me in consideration of the process of bringing forward and latching onto logically relevant reasons for some target proposition and the process of finding implications of some target proposition.
Rand wrote further: “The pre-conceptual level of consciousness is nonvolitional; volition begins with the first syllogism” (1961, 15). I rather think volitional thinking, with action- and image-schemata, is in the repertoire before attaining first uttered word (at about one year), which word is co-referential and incorporated into schemata (and later into sentences). But the thing of present interest is Rand’s notion that volition begins with the first syllogism. That would be a deductive inference, whereas the abstractive process of getting one’s first worded concepts would really be, in her view, an induction. “The process of observing the facts of reality and integrating them into concepts is, in essence, a process of induction” (ITOE 28).
At times Rand seems to use syllogism in a super-broad, rather emblematic way to mean simply any logical inference, deductive or inductive. Pellegrin points out that at 92a28 of Post An. “the term syllogism is taken in a broad and non-technical sense” (2010, 131n15). In the Barnes translation of Post. An., the term syllogism does not appear in this line. Rather in Barnes, Aristotle’s allusion reads “if you produce a demonstration in this way.” In a translation appearing in Code 2010, the allusion reads “for this manner of syllogism” (78). Taking demonstration in an ordinary, nontechnical way, I suggest that the places in which Rand uses syllogism in a broad and nontechnical sense, she means demonstration (or validation) in a broad and nontechnical sense. That is, in a sense broad enough to include demonstrations by deductive inference (say, mortality of man from mortality of animals) as well as demonstrations by induction (say, from various sets of evidence to the mortality of all animals, to the roundness of the earth, or to the cause of the tides).
Rand’s broad use of syllogism is a bit grating in speaking of the right way of changing adult minds, authentically changing them, coordinate with their autonomy and with objective facts. Logic would have been better, reason better still. Her use of syllogism in connection with acquisition of one’s first concepts is grating to the point of a jam. Some sort of abstractive induction is prize principle of that day, and even demonstration (or validation) would seem out of order as characterization of what is happening in that day of the child’s development: (i) personal grasp and naming of some class of items grouped by similarity facts and (ii) boost in personal power of communication by that acquisition.
Notwithstanding such fumbling, I incline to think Rand has proposed worthwhile extension and reform of Aristotle in philosophy of logic; with her conception of logic and all cognition as identification; with existence (ever with identity) supplanting being; with entity supplanting substance; with identity supplanting form; and with essential characteristic(s), as relative to a context of knowledge, supplanting absolute essence. I expect all these shifts imply further differences, yet to be discerned, between logic in Aristotle and logic as it should stand in the metaphysics and epistemology of Rand.
 Cf. Avicenna 1027: “It is evident that each thing has a reality proper to it—namely, its quiddity” (I.5.10). I shall use what or whatness in place of the traditional quiddity (quidditas); see e.g. Gilson 1939, 199.
 Leibniz 1678; Baumgarten 1757 , §11; Kant 1755, 1:389; 1764, 2:294. Rand, in the “About the Author” postscript to AS, and N. Branden, in Basic Principles of Objectivism, erroneously thought Aristotle held the tight bond of identity and noncontradiction that had actually come to be recognized only with Leibniz and his wake.
 Aristotle, Metaph. 1030a20–24, 1041a10–24.
 Intellectual truths such as “Nothing is greater or less than itself.”
 Leibniz 1684, 293.
 AS 1015–16.
 Cf. Peikoff:
Aristotelians seem committed, in spite of themselves, to the view that particulars qua particulars are unknowable by man. Every determinate characteristic of a particular, and thus everything knowable about it, is placed ultimately on the side of Form; Matter in itself is the unorganized, the indeterminate, the nothing-in-particular; it is, as . . . Aristotle put it, “unknowable.” But if all we can ultimately know of a particular is Form; if the individualizing element, the principle of individuation, is in itself unknowable; does this not suggest that the individuality of things is in itself unknowable, i.e., that particulars qua particulars are unknowable? (1964, 214)
 First mood of the second figure; Kneale and Kneale 1962, 235–36; see also Kant 1800, §44n2.
 Lear 1980, 3–5.
 Lear 1980, 1–14.
 See also Aristotle, Metaph. 1006b26–27, 1007a26–27. Let EI designate Rand’s “Existence is Identity.”Aristotle, Avicenna, Henry of Ghent, John Duns Scotus, Francis Suárez, Spinoza, Leibniz, Baumgarten, Kant, and Bolzano also reached principles close to (EI), though not the Randian rank of (EI) or near-(EI) among other metaphysical principles. A Thomist text Rand read had included: “What exists is that which it is” (Gilson 1937, 253). That is a neighbor of Rand’s “Existence is identity.” Neighbor Baumgarten: “Whatever is entirely undetermined does not exist” (1757, §53).
 Metaph. 1030a20–24; Post. An. 83a25–34.
 De Int. 17a33–35; Metaph. 1011b26–27; Plato, Rep. 436b.
 De Int. 17b27–29; Metaph. 996b26–30.
 Cf. Bolzano: “The name of laws of thought is justified to a certain extent because laws of things as such are also laws of our thinking of those things” (1837, §45).
 Cf. Rödl 2012, 22, 39–43.
 Aristotle, Post. An. 99b35–100b5; Salmieri 2008; 2010.
 AS 1016.
 “Logic rests on the axiom that existence exists” (AS 1016). That does not imply that logic is not a tool of right inference concerning propositions containing only terms referring only to nonexistent things (specified by mention of existent things).
 Branden c.1968, 67; Peikoff 1991, 119. Leibniz errs in this way as well; 1678, 187. But on another occasion, Leibniz writes, after listing some “Propositions true of themselves” (such as A is A), writes “Consequentia true of itself: A is B and B is C, therefore A is C” (quoted in Kneale and Kneale 1962, 338).
 See further Buridan 1335, 119–20.
 The first definition of syllogism in my American Heritage Dictionary is as in any elementary logic text. The second definition is: “Reasoning from the general to the specific; deduction.” The definitions are very like these in my Webster’s Unabridged.
 AS 1022–23.
Aristotle c. 348–322 B.C. The Complete Works of Aristotle. J. Barnes, editor (1984). Princeton.
Avicenna 1027. The Metaphysics of The Healing. M. E. Marmura, translator (2005). Brigham Young.
Baumgarten, A. 1757 . Metaphysics. 4th ed. C. D. Fugate and J. Hymers, translators (2013). Bloomsbury.
Bolzano, B. 1837. Wissenschaftslehre. P. Rusnock and R. George, translators (2014).
Branden, N. c. 1968. The Basic Principles of Objectivism Lectures. Transcribed in The Vision of Ayn Rand (2009). Cobden.
Buridan, J. 1335. Treatise on Consequences. S. Read, translator (2015). Fordam.
Code, A. 2010. An Aristotelian Puzzle about Definition: Metaphysics Z.12. In Lennox and Bolton 2010.
Gilson, E. 1937. The Unity of Philosophical Experience. Ignatius.
——. 1939. Thomist Realism and the Critique of Knowledge. M. A. Wauk, translator (1986). Ignatius.
Kant, I. 1755. A New Elucidation of the First Principles of Metaphysical Cognition. In Walford and Meerbote 1992 (WM).
——. 1764. Inquiry Concerning the Distinctness of the Principles of Natural Theology and Morality. WM.
——. 1800. The Jäsche Logic. J. M. Young, translator. 1992. In Immanuel Kant – Lectures on Logic. Cambridge.
Kneale, W., and M. Kneale 1962. The Development of Logic. Oxford: Clarendon Press.
Lear, J. 1980. Aristotle and Logical Theory. Cambridge.
Leibniz, G. W. 1678. Letter to Herman Conring – March 19. In Loemker (L) 1969.
——. 1679. On the General Characteristic (L).
——. 1684. Meditations on Knowledge, Truth, and Ideas (L).
——. 1705. New Essays on Human Understanding. P. Remnant and J. Bennett, translators (1996). Cambridge.
Lennox, J. G., and R. Bolton, editors, 2010. Being, Nature, and Life in Aristotle – Essays in Honor of Allan Gotthelf. Cambridge.
Loemker, L. E., translator, 1969 . Gottfried Wilhelm Leibniz: Philosophical Papers and Letters. 2nd edition. Kluwer.
Peikoff, L. 1964. The Status of the Law of Contradiction in Classical Logical Ontologism. Ph.D. dissertation.
——. 1991. Objectivism: The Philosophy of Ayn Rand. Dutton.
Pellegrin, P. 2010. Definition in Aristotle’s Posterior Analytics. In Lennox and Bolton 2010.
Plato c. 428–348 B.C. Plato: Complete Works. J. M. Cooper, editor (1997). Hackett.
Rand, A. 1957. Atlas Shrugged. Random House.
——. 1961. For the New Intellectual. Signet.
——. 1990 [1966–67]. Introduction to Objectivist Epistemology. Expanded second edition. H. Binswanger and L. Peikoff, editors. Meridian.
Rödl, S. 2012. Categories of the Temporal – An Inquiry into the Forms of the Finite Intellect. S. Salewski, translator. Harvard.
Salmieri, G. 2008. Aristotle and the Problem of Concepts. Ph.D. dissertation.
——. 2010. Perception, Experience and the Advent of Universals in Posterior Analytics II.19. In From Inquiry to Demonstrative Knowledge – New Essays on Aristotle’s Posterior Analytics. J. H. Lesher, editor. Academic.
Walford, D., and R. Meerbote, translators, 1992. Immanuel Kant – Theoretical Philosophy, 1755–1770. Cambridge.
© Stephen C. Boydstun 2015
Hi Stephen. Clearly, this was a lot of work! I have some comments, which I’ll group into three points.
First, your distinctions between Aristotle’s, Leibniz’s, and Rand’s conceptions of identity are a bit confusing to me. Here’s what it sounds to me like you’re saying. For Aristotle, “A is A” asserts little more than that a thing is itself, that it is one. This is the mere fact of singularity of reference. For Leibniz, “A is A” states “the identity of sameness,” which you explain as some sort of sameness between subject and predicate. You seem to mean that the content of the predicate is a subset of the content of subject (whether necessarily a proper subset, I can’t tell.) For Rand, identity is identity of nature/character. Every entity has a specific nature, which is its identity, and which is not other than it is.
Whether or not these are the definitions you have in mind, I think you should be a little more explicit about them. I feel I am having to guess about your meaning.
The Leibniz definition is the one that gives me the most trouble. It seems to require that every statement is analytic, and you even say, “In Leibniz’s view, the law of identity entails that predicates of affirmative propositions are contained in their subjects.” Now, as I write this, it comes back to me that that is right about Leibniz! For Leibniz, every true proposition is indeed analytic. But this is a consequence of his peculiar (screwball) metaphysics, which neither Aristotle nor Rand (nor, as far as I’m concerned, any sensible person) would accept. Propositions like, “my car is in the garage,” “Obama is President,” and “the rain in Spain falls mainly in the plain,” can be true without being analytic. So for that matter can identities like “the morning star is the evening star.” Yet you say, “Identity-containments by subjects of their predicates … look rather like Rand’s identity of character, or nature.” This makes Rand and Leibniz seem much closer, in their metaphysics and philosophy of language, than they really are.
You then contrast Leibniz and Rand and seem to qualify the statement I just quoted when you say, “In Rand’s metaphysics, identity as to nature is tied at the most basic level to identity as self-sameness. For Leibniz identity as self-sameness is the deeper reality of the two. Not so for Rand.” I think your meaning is that for Rand, a thing’s character explains—is the basis for—the sameness of content between subject and predicate. A thing, which is the subject of a proposition, has a set of attributes, F, G, H, …, and if the proposition is true, then the predicate names some of these attributes. Thus there is sameness of content, and the reason is that a thing’s nature consists in a specific set of attributes. But for Leibniz, you say, it’s not like that. He makes sameness of propositional content basic.
If this is what you’re saying, I’m not sure it’s true. There are two problems. First, Leibniz thinks the “concept” of anything is an exhaustive list of its attributes, including accidents and external relations (e.g., being at such-and-such a place at such-and-such a time), covering the whole history of its existence. That is very like the view of nature and its relation to the truth of propositions I just attributed to Rand. The difference is that Leibniz rolls the entire nature (the total set of attributes) of a thing into its concept; Rand doesn’t.
Second, Rand is the opposite extreme on this point. The “file folder” view of concepts anticipates Kripke, it seems to me. It rejects reference fixation by definitions in terms of any set of attributes. Thus, I don’t think she would accept “self-sameness” at all, if this means that the predicate of a true proposition is contained in its subject. A thing metaphysically has definite nature (a set of attributes), to be sure, but our concept of a thing is ultimately just its reference (the thing itself), not a set of attributes. It is not part of one’s concept of Obama that he is President; that Obama is President is not a conceptual truth. So to speak of sameness of content between subject and predicate in this proposition is incorrect.
All this might just reflect my confusion about what you’re trying to say. If so, please disregard.
Second, on the definition of logic, you claim that demonstrative inference does not necessarily depend on the law of contradiction. Your argument appears to be that first figure Aristotelian syllogisms are self-evident and not derivable. That is, they are not based on any further argument, and in particular they are not based on any argument to the effect that to deny their conclusions would entail a contradiction. So they represent a form of reasoning that does not depend on the law of contradiction. Therefore, Ayn Rand is wrong to define logic as noncontradictory identification. It should be called “the art of perfectly truth-preserving inferential identification” instead.
I think this is your most substantive and interesting point and deserves to be developed more than you do here. How would someone who disagrees argue against you? You have quoted Peikoff to the effect that the law of contradiction is not a distinct principle from the law of identity. Maybe he would say that therefore it’s not wrong to call logic noncontradictory identification, and perhaps for some reason it’s more helpful or useful to emphasize avoidance of contradictions. Maybe there’s some sort of heuristic value to thinking in terms of avoiding contradictions?
I think you could be right about logic. It has long seemed to me that the fundamental basis of all thought is identification. Ultimately, thought comes down to what we “see.” We see facts, objects, attributes, reasons, relations, and so forth. And not all reasons, even all demonstrative reasons, are seen by seeing that to deny them entails a contradiction.
By the way, I wouldn’t worry about the singular “this electron” or “Socrates” in your syllogisms, and I don’t think the cure should be to convert those to particular (“some”) propositions. The standard way to treat singular propositions in Aristotelian logic is as universal (“all”) propositions, not particular propositions. See McCall’s old Basic Logic, 2nd ed., Barnes & Noble, 1952, p. 52, and Baronett’s Logic, 2nd ed., Oxford U.P., 2013, p. 203.
Also, endnote 19 makes no sense to me, where you say, “only to nonexistent things.” I simply don’t understand this. Why nonexistent?
Third, you write:
You then worry that if volition begins with the first syllogism, and syllogisms presuppose concepts and other abstract achievements, Rand would appear to be saying that, for example, concept formation is not a process of volitional thought. You resolve the problem by interpreting her use of “syllogism” broadly to include any logical inference or indeed any process of reason. I’m sure this is right. In context, it’s clear that she means to distinguish “pre-conceptual” from conceptual thought. Her use of “syllogism” is just a colorful way of expressing herself.
I also agree that it’s unlikely that volition is specially connected with conceptual thought somehow. She seems to be invoking an idea I find first in Locke, also in Rousseau, and most particularly in Kant, that freedom lies in the power of our minds to act in the light of reasons rather than from the compulsion of efficient causation. Kant especially seems to conceive of the mind as split into two realms, an animal realm of impulses, sense-impressions, and associationistic “reasoning,” and a human realm of concepts, rational will, and judgment. The former is passive and heteronomous; the latter is active and autonomous (free). This division is particularly clear in Neo-Kantians like Wilfred Sellars and John McDowell, with their talk of “the space of causes” versus “the space of reasons” (the latter being accessible to humans but not other animals).
It is odd that she should seem to endorse such a picture of the mind. For one thing, there is really no clear connection between conceptual or rational thought and volition. To act in the light of reasons may be different from acting from efficient causes, but it seems no less deterministic. It may be—I hope it is—that we agree that 2+2=4 because we see the reason in it, not because we are efficiently caused to do so. But it’s not as though there’s any choice about it; the reason is determining. And once you have seen the reason, you cannot choose not to accept it. We do not choose our beliefs, and we do not choose our reasons. (Actually, we do not choose our beliefs because we do not choose our reasons.) But whether one is determined by reasons or by causes, determination is still determination, not volition.
Where we seem to have some choice is in where we direct our attention, not what we see when we direct it there. Ironically, this is Rand’s view most of the time. But this capacity to direct our attention is not internally related to conceptual or rational thought, and I can’t see that they are empirically related either. In that case, volition and conceptual thought are separate faculties.
LikeLiked by 1 person
Thank you, David. Your comments, impressions, and leads are helpful, as always.
I’ve not thought of Leibniz’ “A is A” or identity of sameness as something to be seen by him or us as explainable by sameness between subject and predicate. I’ll need to renew acquaintance with his thought to see if such a picture could be square. His uses of identity in mathematics and equations needs to be part of that picture also. For Rand, I’ve always seen her conception of identity as encompassing particular identity, not only specific identity. Not only identity of species or kind, but identity of that/which, when, and where. Clearly, I need to make all this explicit to head off misinterpretation.
In saying “Identity-containments by subjects of their predicates . . . look rather like Rand’s identity of character, or nature,” I meant only that he has some sort of awareness of specific identity, not only particular identity, and that he has got them both working together in subject-predicate. The idea that for Rand, “a thing’s character explains—is the basis for—the sameness of content between subject and predicate” is interesting, but I don’t think so, and it’s never occurred to me. (My model of predication suited to Rand’s metaphysics, in the final section of my 1991 “Induction on Identity” [[i]Obj.[/i] V1N3] resolved predication into two uses of particular identity together with one of specific identity.) I don’t see particular identities as more or less primitive than specific identity in Rand, and as in the ’91 essay, I still see them (and Rand as seeing them) as both present in some way in every concrete.
You mentioned the old Objectivist idea of mental “file folders.” In the book, I do hope to assimilate what Rand/Branden were saying about that with what I learn from Recanati’s [i]Mental Files[/i] (2012) on my shelf.
I’ll have a chapter completing, reforming, and providing systematic basis of my online series of articles “Objective Analyticity.” That has to include determining relations between analyticity and logical necessity, expected to be considerable opposition to the Rand/Peikoff attempt. That is a whole shelf of books here in my library. Naming and Necessity is part of that.
Thanks for the references on singular subject and syllogism.
I’ll definitely want to develop rejoinders to counterarguments to my reasoning away from Rand’s definition of logic. It should treat the response you made in behalf of the vista of Peikoff, which is also the vista on non-contradiction set out in Branden’s lectures of the ‘60’s. (The case against Rand’s definition of logic will need to move beyond syllogism in the end and include mathematics’ tool of deductive logic [including mathematical induction], which seems to rely a lot on direct identity.) The idea of noncontradiction being an epistemological norm due to the metaphysical fact of identity, an Objectivist idea, is fine and can remain standing throughout repair of Rand’s definition of logic. I’d decline the counterargument, to my critique, that conceives of appeal to contradicition under denial of the inferences as a checking that first-figure inferences are indeed valid, however obvious their validity may seem. The obvious correctness and necessity of the principle of noncontradiction are no greater than the obviousness and necessity of the first-figure inferences themselves.
I should mention an error of detail I made in the paper. It is not only conversion that Aristotle uses to reduce the other syllogisms to first-figure. In those reductions, he also uses first-figure inferences themselves. That’s not a problem, and by these reductions, he has assurance of the validity of all the syllogisms not obviously valid (again, a nod to the esteemed Jonathan Lear). I expect we could also show their validity by the noncontradiction technique, but it remains that much in logic is identity-direct, not only the identity-indirect that is noncontradicition, which latter is no more profound in logic than recognitions of certain identities directly.
Just a couple of quick comments, Stephen.
My knowledge of Leibniz is not deep. My account was intended to invoke only something “everyone knows” about him, namely that he held a “bundle theory” of particulars according to which the identity of a particular included every property of that particular, including external relations. Thus, in Section 8 of the Discourse on Metaphysics we learn that whether Alexander the Great died by poison is knowable a priori by anyone who has the concept of Alexander. This fact about Alexander is actually known a priori only by God, but it is so known by Him because He has the relevant concept. It is an analytic truth that Alexander died by poison, if it is true at all. This doctrine is a consequence of the metaphysical doctrine that particulars are individuated by their universal properties; there is no further basis of particularity. Every particular is and must be a unique set of universals.
Doesn’t identity apply to everything? Whatever one thinks exists—particulars, universals, relations, events, states, concrete or abstract objects—is subject to identity, isn’t it? So I’m not sure what you mean here.
Your distinction between particular identity and specific identity reminds me that Rand is, by my lights, a nominalist. She thinks only particulars exist. “Universals” are concepts based on primitive similarity relations (or something like that). This makes me think you’re right to criticize me here. I was too quick to attribute to Rand the view that the character of a thing consists of a set of properties. That could still be a useful way to think about her philosophy of language, but I don’t know that and don’t feel comfortable offering an opinion about it.
LikeLiked by 1 person
David, you mentioned in your original comment something to which I forgot to respond. You didn’t understand why I was speaking of nonexistent things in note 19:
“‘Logic rests on the axiom that existence exists’ (AS 1016). That does not imply that logic is not a tool of right inference concerning propositions containing only terms referring only to nonexistent things (specified by mention of existent things).”
It would ordinarily be thought that deduction applies to nonexistent things as well as to existent things. The Rand point which I have adopted might suggest that logic (right logic) is inapplicable to nonexistent things. I want to deny such a conception as that in what Rand was putting forth in the statement “Logic rests on the axiom that existence exists.” There are other denials I should make by way of precising this conception of Rand’s, at least as I cash it out for my own view. Part of the cashing out will be more positive and more definitely extension beyond what Rand made of her proposition. Those would be in constraints on right implications (such as the Bolzano denial that contradictory premises imply ANYTHING, as we commonly hold; rather they imply NOTHING) and likely some selection among various modern modal logics.
In a later comment, you mentioned that Rand could be regarded as a nominalist on the ground (much like Abelard) that only particulars exist. It is concrete particulars, I think, but perhaps you already meant that. Anyway my usual spiel, which I still think right, is that Rand’s view that only concrete particulars exist does not entail that only entities are concrete particulars (ITOE 241, 278). And indeed in her view my height and your height are concrete particulars, and so is the ratio of our heights. Such magnitude dimensions are really out there, independent of our scaling of them. (This is a point in an exchange between Rand and Peikoff in the ITOE Appendix, 199–200, though what I term ‘magnitude’, they term ‘quantity’.) They are out there with all the structure we show in a synthetic, unscaled geometry such as we do in Euclid. Some similarity relations and comparative similarity relations are out there, and in Rand’s view those similarity relations as well as our concepts are explicable in terms of magnitude dimensions in the world. The magnitude dimensions and their various ratios or degrees are primitives. So I would not class hers as a nominalism, not resemblance nominalism in particular. So far I think of her theory of universals as a mensural theory; perhaps that is a new variety of moderate realism.
There is a further complication of classifying Rand’s theory of universals and concepts, I’d say. That is that her theory really won’t do as a theory of how we compose all our concepts, only as a theory of how (perhaps) all our concepts could best be composed. (All concepts excepting those that are logically presupposed by the technical concepts of measurements, in my view.) The other theories seem not to be normative, but then maybe we should try to look at them anew in that way.
Concerning logical inferences that involve propositions whose terms refer to nonexistent things, yes I see now what you meant. Actually, I think I should have seen it before.
As to what she says exactly and what it means, I’ll leave that to you. But I’m a little confused by your talk of a person’s height, for example, being a concrete particular. From what you seem to mean, your terminology I think is nonstandard. Surely a concrete is a complete particular (whatever a particular exactly is) with all its properties. I am a concrete particular, and so are you. When we talk about my height, however, by which I assume you mean my height as instantiated by me but distinct from the rest of me, then surely that is an abstract object, not a concrete. Of course—if I understand you correctly—my height is still a particular. It is not just my height we talking about, but my-height-as-instantiated-by-me. This is what we might call a property instance.
An abstract particular is abstract only in being abstracted from the remainder of the concrete particular. It is not abstract in the sense that the property (the height) subsumes a range of others. Having a height between five and six feet is abstract in the latter sense, as is being a cat, being a car, and being a subatomic particle. These are all abstract in the sense that they have many ways of being instantiated. Being some precise height is not abstract in this sense, but it is still abstract. It doesn’t exist by itself, and we can only conceive it by abstracting it from the other properties that compose a concrete particular.
Is this what you meant by my height being a “concrete particular,” that it is a property instance in the sense I described?
If so, your account makes Rand sound like a trope theorist. Trope theorists don’t believe in universals, but they don’t only believe in concrete particulars either. They take property instances as primitive. These are particulars, as I said. So there are only particulars. If you and I are the same height, then my height is one existent and your height is another. The heights themselves (independent of instantiation) are not numerically identical, as a realist would have it. But they are qualitatively identical; that is, they are exactly similar. And these similarity relations are the basis for our talk of universals. This sounds like the view you are attributing to Rand.
The trouble with trope theory, it seems to me, is that it cannot specify a basis for the qualitative identity between tropes. The negative charge of this electron is said to be not numerically the same as the negative charge of that one, but qualitatively the same. But what is a quality? It can’t be what they have in common. They have nothing in common; that’s the whole point of trope theory. They are “exactly similar.” But what does that mean? Not that they share a common property! Really, it seems that similarity has to be taken as primitive, and it is the only real primitive. Tropes don’t explain similarity relations, similarity relations explain tropes. It really is just a version of resemblance nominalism. And now you have all the problems to which that view is subject, which can only be escaped (as far as I can see) by adopting a Lewisian metaphysics of possible worlds, or something equally ornate and implausible. Better to just accept universals, I say, and simplify life.
LikeLiked by 1 person
Re: Potts post below of 11/22/15 (I don’t find how to post this downstream; hope it is alright to add here.)
Thanks for the reflections on tropes. I have a recent book on that by Ehring and the somewhat earlier book on resemblance nominalism by Rodriguez-Pereyra. But I haven’t yet come round again to concepts and universals in the course of my own writings and their studies to dig into these yet.
Thanks for the flag on what is a concrete. Anything physical I take for concrete, and leave open so far what more, if anything, should be allowed in a more general concept of the concrete subsuming the physical. In saying our heights and their ratios are concretes, their attachment to us or other objects is not the thing I’ve had in mind. Those lengths and their ratios are concretes wherever they are, including in regions of empty space. (With Newton on this.) We distinguish in our era between physical spatial nature and mathematical spaces and seek which of the latter are the case for the former (in which physical settings; in this room, say, or in the larger scales of the universe). Physical space and its relations are concretes.
Rand’s take on the sameness of universals in different individuals came down to sameness of certain groups of traits, each trait having same magnitude dimension(s) in different individuals in the class, with different (or same) measure value of each such dimension being had by members of the class. She got a ways beyond just saying “same crucial traits shared by different individuals” or “similarities (comparative similarities) of a class of individuals” by conjecture of magnitude structure of all traits and similarities, then mentally lifting particular distinct measure values of the individuals (within some reasonable range of values) to contract the cases into a concept. It is a big proposal, one she could not execute very far, and about which she had quite a few confusions. She did not understand, for important example, that different classes of measurement scales are objectively appropriate (or inappropriate) to different sorts of magnitude structures in the world; she assumed an impoverished scale is only an indication of our ignorance about how to measure something. But that and other confusions can be cleared up, and her central fix on suspension of particular measure values (whether on an ordinal scale or higher) as universal key for concept-class abstraction can remain a conjecture for further development and challenge.
It is the reality of recurrent mind-independent traits, dimensions, and magnitude structures, and the reality of mind-independent dependencies among things that incline me at times to take Rand’s scheme as some sort of moderate realism. One’s walking and laughing and height are concrete things. Those can be perceived without abstraction. We can start thinking of them as depending on entities or matter or substance, and as arriving again at walking, laughing, and height by mental prescinding, but I can’t see that that changes their concreteness. Similarly, there are other concretes we can only discern through abstraction, such as elementary particles, electromagnetic fields, or spacetime, but what they are is concrete.
In Rand’s thinking and my own the concrete cuts across all her categories: entity, attribute, action, or relation. (I realize the last three are overlapping, but its OK by me to just lift them up for salience.) It seems to me the completeness you associate with a concrete is something R&B might consign to entity and only it. B would be a bit wary of that, considering the way locomotion types can enter perceptions as wholes and considering that metaphysics has to square with our best physics. It’s true as Nozick quipped, and not regrettable in my view, that the physical tail wags the metaphysical dog.
Hi Stephen. Just a couple of remarks, suggested as food for thought.
What does this exclude? I mean, isn’t well nigh everything physical? And on the other hand, if mind were a separate substance, would a given pain sensation, say, not be concrete?
I think of the universal–particular and the abstract–concrete distinctions as separate though related. A universal is a one-in-many, what constitutes the unity of things insofar as they are of the same kind. A particular is, roughly, a bearer of universals. Whatever else they are, particulars are spatiotemporally unique. I myself think that particularity just is spatiotemporality. And I think of particulars as space–time worms. On the other hand, an abstraction is a mental construction. We can separate and recombine aspects of reality in thought and thereby form abstract ideas. The concrete is reality as we encounter it in sense experience. As such, it is already subject to some cognitive processing, true, but not (or not necessarily) to higher level, conscious processing. Thus, the abstract–concrete distinction pertains to cognition; the universal–particular distinction to metaphysics.
Conceptualists like Locke and Rand seem to want to replace the universal–particular distinction with the abstract–concrete distinction. This seems like a mistake to me. For one thing, we really need both distinctions. Having both enables us to talk about abstract particulars, as I did in a previous comment, and also concrete universals. An abstract particular is a particular instance of one or more universals; for example, the electric charge of a certain electron. A concrete universal is all the universals comprised by a certain kind of particular; for example, all the properties of an electron. The universal being an electron might be thought of as a concrete universal. Again, any concrete object can in principle be specified by the totality of the universals it instantiates. This totality is potentially repeatable.
For another thing, not all abstractions are universals. The concepts of chair, furniture, footrace, automobile, and tree are all abstractions, but they are not universals. Abstract ideas have borderline cases; universals don’t. Universals are the true building blocks of the universe. There is presumably a relatively small number of elementary (non-compound) universals. As you say, it is the business of science to identify them. (A valuable paper in regard to these points is David Lewis’s “New Work for a Theory of Universals.”)
LikeLiked by 1 person
Thanks for the Lewis title. I’ve just gotten hold of it thanks to this site:
PAPERS OF DAVID LEWIS
I take difference in spatiotemporal character in concrete particulars and abstractions as fundamental to their distinction. Your distinction of concrete and particular puts in mind that perhaps there is a fine difference in spatiotemporal character between them.
You mention the universal-concrete distinction by Locke and Rand, but I’m finding it put in those terms also in Aristotle’s Metaphysics at III.4:
“If we admit in the fullest sense that something exists apart from the concrete thing, whenever something is predicated of the matter, must there, if there is something apart, be something corresponding to each set of individuals, or to some and not to others, or to none?”
“And since this is impossible there must be something besides the concrete thing, viz. the shape or form.”
Do you think Aristotle runs with Rand and Locke, as rather loose between ‘concrete’ and ‘particular’ (‘individual’), or is this translation (Ross) deforming Aristotle here?
I’m sensitive to the circumstance that not all abstractions are universals. From the early work in Objectivity, I always cite Macnamara’s work on proper names and the logical resources required for competency in them.
On the question of whether all concretes are physical, yes, I incline to think so, but in presentation prior to taking up whether occasions of consciousness are only physical, I think it’s better to leave open the possibility of concretes being a wider class.
Thanks again for all the feedback, David.
I see now another point of Aristotle that needs incorporation into the text above (which is from the draft second chapter of the book [Part 1 of 2, historical settings with Rand and with my own alternative]—one, Nietzsche; two, Aristotle; three, Descartes; four, Kant, Reinhold, Fichte; five, Hegel, et al.; six, Bolzano, Schelling, Peirce.). “For all things that we know, we know in so far as they have some unity and identity” (999a28).
I don’t know the answer to that, and I’m in Houston at the moment and so don’t have the Greek text to look at. I don’t know if David Riesbeck, who also reads Greek, could be induced to weigh in.
I will say that what strikes me most about the first passage you cite is the remark, “whenever something is predicated of the matter [there should be a universal].” Notice that predication, language, the fact that we group things under the same label, is regarded as the criterion of universals. (And I hope I’m getting even this right, by the way. As I recall now, Aristotle’s actual term, which gets translated “is predicated of,” more literally means “belongs to,” so what Ross translates as “A is predicated of B” more literally reads “B belongs to A.” If this is right, then Aristotle’s discussion is less linguistically oriented than I am saying. I can’t trust my memory on this, however, and will have to check it when I get home.)
In my view, perhaps the most significant breakthrough in the theory of universals since Aristotle himself took place only about 40 years ago with the clear separation of metaphysical questions about universals from linguistic questions about the semantics of kind terms. I attribute this mainly to David Armstrong (A Theory of Universals), though I think credit is also due to Hilary Putnam (“On Properties”). Before them, it was usual to assume that wherever there was a common name, there was a kind, and it was the job of the theory of universals to explain the nature of that kind and of our knowledge of it. Therefore, a realist in the theory of universals was obliged to find a real, metaphysical universal to be the referent of the term. Thus, “cat” implied a real universal of cathood, “blue” implied a real universal of blueness, “car” implied a real universal of automobilehood, “justice” implied a real universal of justice, and so forth. It is obvious how this doctrine could make nominalism seem like a pretty good idea! Ditto for conceptualism.
Armstrong’s innovation was to make science the arbiter of universals, not language. So for Armstrong, there is no real universal of cathood, much less of justice, but there will be (probably) real universals of having a mass of 1 gram, having an electric charge of -1, being a neutrino, etc. In this way most of the problems that embarrassed realism, especially the problem of borderline cases, but also the problems of claiming that we have some sort of special cognitive contact with a metaphysical realm of universals and of claiming that unruly “kinds” like table and chair and river and even horse and cat corresponded to real universals, were swept away, which enabled philosophers to focus on the more purely metaphysical question of what grounds the uniformities in nature.
This is a sweet quote. Notice that unity and identity are only being asserted to be prerequisites of knowability, not metaphysically necessary. They are necessary conditions of cognition. The way I read him, this is also what Aristotle says about the law of contradiction. To the extent that Leonard Peikoff makes stronger claims about Aristotle’s “logical ontologism” (in his dissertation), I’m not so sure he’s right.
I’m still mulling over the philosophy of language/logic part of this discussion, but I have some thoughts on the last part, the connection between volition and cognition. It seems to me that some issues are falling between the cracks.
Rand thinks that there’s a strong connection between volition and conceptual-level cognition. In a broad way, I agree with her, but I think she’s sloppy and ambiguous about how she puts the point, and I think Stephen has overstated it some. I take her view to involve two (or three) basic claims:
1. Higher-order cognitive functioning is something that needs to be enacted by the knowing agent; it doesn’t just happen on its own, without the agent’s initiative. In fact, it requires a distinctive sort of metaphysically free act which involves the agent’s resolution to know the world, to conform his thinking to what is the case independently of that thinking, and not to be distracted by things that impede thought. Put in slightly Freudian terms, adherence to the reality principle requires a volitional act. I actually like the way Peikoff puts this point in OPAR: “objectivity [is] volitional adherence to reality by the method of logic” (p. 116), understanding “logic” in a very broad way.
2. Once the basic resolution involving (1) is in place, the agent has to monitor and regulate his thinking so as to maintain the resolution involved in (1). That requires a certain kind of internal control, and the control involved entails metaphysical libertarianism, a fact that can be confirmed by introspection and inference from it.
Here’s what Stephen says:
I think that way of putting Rand’s thesis is unnecessarily strong: in other words, you can preserve what she’s saying without going quite as far as this gloss implies.
I take my (1) to map onto the statement “To think is an act of choice,” and my (2) to map onto “The connections of logic are not made by instinct.” But it’s an overstatement to say that “logical inference is consciously directed,” where that’s taken to mean that all inferences are consciously directed, hence volitional. Some inferences are automatized, hence not consciously directed, and not (at the time) volitional. Other inferences are consciously directed, monitored, controlled–hence volitional. How we divide the terrain will differ knower to knower, but for any knower, it’ll be the case that the inferences that are not automatized will have to be enacted, monitored, etc. in a volitional way.
Something similar is true, I think, of concept formation. I don’t think Rand is committed to the thesis (or needs to be committed to the thesis) that literally all concept-formation is volitional. It’s possible that all first-level concept formation is driven by deterministic factors. All she needs is the claim that there is some level of abstraction–call it “abstraction from abstractions” or a proper subset of such abstractions–such that that level requires volitional initiative. Only an agent who’s resolved on (1) and committed herself to (2) will properly form concepts at that level, whatever it is.
When Rand and Stephen say this, then, I think they’re overstating a legitimate point:
There’s no reason to draw the distinction in this strong way, so that pre-conceptual=deterministic, and conceptual=volitional. More plausibly, Rand should have said that the pre-conceptual and swatches of the conceptual can both be deterministic; the point is that some higher-order functions of conceptual thought are volitional. For the same reason, there’s no (strong) reason to attribute metaphysical freedom to one-year-olds (and no strong reason not to). That’s a difficult (possibly irresolvable) empirical question, not one settled simply by observing that they’re engaged in conceptual thought.
That said, apart from his very first paragraph (the one that starts “You then worry that if…”), I don’t agree with David, either. Here’s David:
On the very first sentence: I wasn’t sure that Stephen did regard it as unlikely that volition was specifically connected with conceptual thought. So perhaps Stephen could clarify that.
But I don’t think Rand is making the claim that David finds in Locke-Rousseau-Kant (assuming that L-R-K are asserting the same claim, which I doubt). The idea of “volitional adherence to reality” that I described in (my) claim (1) above is not a case of the mind’s acting in light of reasons rather than compulsion by efficient causation. It’s essential to Rand’s view that there’s no antecedent reason for “volitional adherence to reality”; you just resolve to adhere or not, on confronting reality itself. So the contrast with deterministic efficient causation is there, but not the affinity with neo-Kantianism a la Sellars and McDowell.
I don’t think it helps to compare Rand with Kant on this point. Kant does conceive of the mind as split into two realms, but Rand doesn’t. There’s no analogue in Rand for Kant’s “animal realm of impulses,” except when she regresses into such talk to attack primitive or immoral people; I find that sort of talk unfortunate and confused, but even there, what she’s saying is a far cry from Kant. There’s not, for Rand as there is for Kant, an ineradicable source of animal impulses for all humans qua humans. In any case, when she alludes to animal-like impulses, the source of the impulses isn’t a separate “realm.”
That said, there is a clear distinction to be drawn between whims, impulses, perception, associative reasoning (etc.) on the one hand, and the control we exert over our thoughts when we engage in higher-order inquiry and inference on the other. The former is relatively passive, and the latter is active. Rand makes that distinction, as do a lot of philosophers, but I don’t see how it saddles her with the Kantian-type commitments David discusses. I think it’s an introspectively obvious set of facts leading to a basically uncontroversial distinction.
To continue with David’s comment:
The connection between rational thought and volition is supplied by the resolution to conform one’s thought’s to reality and the control required to effectuate that resolution during inquiry. But that isn’t really acting “in the light of reasons,” but a precondition for doing so.
It’s true that as adults simple mathematical operations like 2+2=4 are now automatized, hence no longer a matter of conscious inference, monitoring, control, etc., but Rand wasn’t denying that. Her point was that ab initio, one comes to enact one’s mathematical capacities volitionally. You can’t learn addition or subtraction without volitionally engaging those capacities. Once you do, however, what happens next may well be (probably is) entirely deterministic. Once you’ve learned math, the 2+2=4 inference is clearly deterministic. (The whole point of learning one’s multiplication tables is to make multiplication a deterministic process.) Once you have the reason, you can’t choose not to accept it. But the legitimate point she’s making is that, at least for a proper subset of inferences, the agent has to make a more primitive choice to resolve to know the world and enact the relevant world-tracking capacities in order to be in a position to grasp rational connections in the first place.
I think the Randian response to David is that he’s focusing on cognitive processes that are operating downstream from the locus of volition. He’s abstracting from what the agent volitionally has to have done in order to be in a position to be determined by reasons. You can only be determined by reasons after you’ve decided to operate by reason. Rand’s point is that the latter decision is volitional. Unfortunately, she’s not sufficiently clear about how much of the rest of our cognitive life is deterministic. But Bayer and Salmieri’s 2014 paper on this topic seems to me a start in the right direction.
As I see it, the real problem with Rand’s view is not that she draws a distinction between active/passive which maps in some way onto a distinction between higher-order cognition and lower-order consciousness. That’s what she gets right. The problem arises from her (inconsistent) insistence that consciousness is “active” through and through.
The problem is that this account of consciousness is incompatible with her account of our conscious experience of passively-received states like pain and pleasure.
So one set of claims here is: Consciousness of pain is a sensation, and consciousness of pain is an automatic response. Further, every automatic response is a passive state. Another set of claims is: consciousness is not a passive state, but an active “process.” The problem is, consciousness of pain is an instance of consciousness, but it’s unclear how it can be active in the relevant sense.
So I agree with the spirit but not the letter of David’s criticism. There’s an active/passive problem in Rand, but the problem is that she seems to want to associate activity with consciousness as such, so that consciousness is never passive. But she also has to admit that consciousness has a passive dimension. As stated, I’d conclude that her “theory” of consciousness is not coherent.
LikeLiked by 1 person
Hi Irfan. I suspect we don’t really disagree about anything important here as regards volition. But let’s find out.
The two main questions are, first, what is going on when Rand says, “The pre-conceptual level of consciousness is nonvolitional; volition begins with the first syllogism,” and second, what is it exactly that volition controls? In dealing with both, you invoke the distinction between automatic and controlled processing in psychology. You take it that automatic processing is deterministic and controlled processing is volitional. I can agree with that. But you also take it that automatic processing governs lower-level (pre-conceptual) processes (“whims, impulses, perception, associative reasoning, etc.”), as well as automatized conceptual processes such as simple arithmetic, whereas controlled processing governs (non-automatized) higher-level, conceptual processes. For example:
Here is where I see a problem. This is not the right mapping. Lots of lower-level, pre-conceptual processes are controlled. These include search of memory to remember details of an event you experienced in the past, or to remember which actors were in a certain movie, etc.; the process of keeping a list of six items in working memory; scanning a group of people for a woman with white hair (or any conjunction of features [“Where’s Waldo”]); catching a football on the run (though perhaps not for a professional player); maintaining a faster walking pace than is natural for you. Even a shift of visual attention from left to right, if it is the result of a conscious choice, is a controlled process, and volitional. (Some of these examples have come from Daniel Kahneman’s Thinking, Fast and Slow, Farrar, Strauss and Giroux, 2011, and I could supply many other references for how psychologists make the distinction between automatic and controlled processing.)
This is why it is very strange to read that volition begins with syllogisms. It seems so obvious that it doesn’t. I’m not sure why anybody would believe such a thing unless they were in the grip of the sort of Locke-Rousseau-Kant view of freedom I mentioned, and that is why I suggested that she might (when she wrote the sentence in question, at least) have been under the influence of that idea.
Your own statement of her view of the connection between volition and conceptual-level cognition, which you broadly agree with, is:
One thing that strikes me about this is that it seems to restrict volition to a choice to know the world, to adhere to the reality principle. This can’t be quite right, can it? Surely evasion of reality is a choice too. We can choose to think or not and to direct our attention and resources where we will. That is my own view. I suspect it’s really yours also.
If volition consists in our ability to consciously direct our attention and resources where we will, this can be guided by our knowledge and experience. As we learn principles of logic, for example, we learn fallacies we should be on the lookout for, and we can direct our attention in the direction of spotting them—or neglect to do so. We also learn what policies make for more accurate judgments, and we can direct our attention toward following them—or not. It is in making these kinds of choices that we volitionally adhere to the reality principle. I see this as consistent with your own view.
The final question is whether recognizing reasons is volitional. A key point I was trying to make is that it is not. Although you have to direct your attention in the right direction to see a reason, and you have to expend some mental energy, the recognition itself is not a choice. I am saying that, for example, if you acknowledge the thought that “if the sun is shining, then it is day,” and if you also recognize visually that the sun is shining, and if you think these two thoughts in conjunction, then you will recognize that it must be day. Other than putting the effort into thinking about it, this is not volitional. You do not arbitrarily choose whether to see the logical connection between these statements. You can choose whether to look, but if you do choose to look, you cannot choose what you will see. And having seen it, you cannot take it back. You cannot arbitrarily, consciously choose not to recognize a logical connection once you have seen it.
It seems like you accept this, although it’s not crystal clear. You say:
I don’t disagree with any of this. And it seems to endorse the point I was making about the recognition of reasons. If so, then we are in agreement about the points of greatest importance to me.
I agree with everything you say in the second half of the post, the part that begins, “Your own statement of her view…” I want to look through Kahneman before I comment on the first half; I’m not quite sure yet of what I want to say.
LikeLiked by 1 person