Philosophy Physics
Physical Cosmology
Philosophy of Physics
Black Holes
The Big Bang
Anthropic Principle
Religion Atheism
Philosophy of Time
Philosophy of Language
Linguistic Metarules
Mind Consciousness
Philosophy of Science
Hist. of Analytic Phil.
Felt Meanings 1986
Books/Book Comments
Press Releases
Classical Music Lyricist


You can search this site:



Marcus, Kripke, and the Origin of The

New Theory of Reference

By Quentin Smith

Western Michigan University


Published in: Synthese, Volume 104, No. 2, August 1995, pp. 179-189.  Reprinted in (eds. James Fetzer and Paul Humphreys),  The New Theory of Reference: Kripke, Marcus and Its Origins, Kluwer Academic Publishers: Synthese Library Series 1998, pp. 3-12.)


P.W: Humphreys and J:H. Fetzer (eds.), The New Theory of Reference, 3-12. copyright 1998 Kluwer Academic Publishers. Printed in Great Britain.



The New Theory of Reference in the philosophy of language became widespread in the 1970s and is still flourishing today. The New Theory implies that r;- many locutions (e.g., proper names) refer directly to items, which contrasts with the traditional or old theory of reference, which implies that names and relevantly similar locutions express descriptive senses or are disguised descriptions. The New Theory encompasses such notions as direct reference, rigid designation, identity across possible worlds, the necessity of identity, a posteriori necessities, singular propositions, essentialism about natural kinds, the \2 i argument from the failure of substitutivity in modal contexts that proper names are not equivalent to contingent definite descriptions, and related ideas ~! and arguments. Some of the contributors to the development of this theory' include Kripke, Kaplan, Donnellan, Putnam, Perry, Salmon, Soames, Almog, Wettstein and a number of other contemporary philosophers.

The point of this paper is to correct a fundamental and widespread misunderstanding about the origins of the New Theory of Reference; the main misunderstanding is that it is widely believed that Kripke originated the major ideas of this theory, presented in his 1972 article on "Naming and Necessity" (Kripke, 1972) and his 1971 article on "Identity and Necessity" (Kripke, 1971 ). The fact of the matter is that the key ideas in the New Theory were developed by Ruth Barcan Marcus, in her writings in 1946-47 (1946; 1947) and especially in her 1961 article on "Modalities and Intensional Languages" (reprinted with small changes in (Marcus, 1993)). "Modalities and Intensional Languages" was presented in February, 1962 at the Boston Colloquium for the Philosophy of Science; Marcus' commentator was Quine and Kripke participated in the discussion which followed.


I shall begin by adducing a number of representative instances where this historical misunderstanding is evinced. The point of the following series of quotations is to illustrate the nature and extent of the misunderstanding of the origin of the New Theory of Reference.

                If one picks out at random any essay in the philosophy of language or essentialism in the past twenty-five years or so, then one will most likely find the New Theory of Reference attributed to Kripke (and some others, e.g., Kaplan and Donnellan) and that Marcus' name will only occasionally be mentioned. A recent example is Recanati's article in Philosophical Studies: "My starting point will be the ...notion or rigidity, introduced by Saul Kripke in the philosophical literature" (Recanati, 1988: 103). Paging through recent issues of Nous, one finds David Braun listing the proponents of the new "direct reference" theory of names: "Kripke (and) Donnellan's view strongly suggest direct reference. Almog, Kaplan, Salmon, Soames and Wettstein explicitly advocate versions of Direct Reference for proper names" (Braun, 1994: 465, n. I). It is notable that the originator of the new "direct reference" theory, Marcus, is not even mentioned.

Many misattributions of Marcus' ideas to Kripke and others can be found in the writings of contemporary philosophers, but what is even more surprising is that the leading developers of the New Theory do the same. Apart from Marcus, no one has done more to develop the New Theory of Reference than David Kaplan. David Kaplan, in some of his published works, attributes the New Theory of proper names to Kripke. He correctly notes that "the term 'rigid designator' was coined by Kripke to characterize those expressions which designate the same thing in every possible world in which that thing exists and which designate nothing elsewhere". But Kaplan proceeds to identify falsely the thesis or claim that proper names are rigid designators as Kripke's specific claim (rather than Marcus'); Kaplan writes: "He (Kripke) uses it in connection with his controversial, though, I believe, correct claim that proper names, as well as many common nouns, are rigid designators" (Kaplan, 1989: 492; my italics). Of course, Marcus did not make a claim for the rigidity of common nouns, and this idea is correctly credited to Kripke, but the same does not hold for the rigidity of proper names, as we shall shortly see.

Even in cases where we find philosophers listing a series of contributors to the New Theory of Reference, Marcus is typically not mentioned. John Perry, who, along with Kaplan, is the leading exponent of the New Theory of indexicals, writes that "Lessons learned from the works of Donnellan, Kaplan, Kripke, Putnam, Wettstein and other New Theorists of Reference have convinced me to accept two theses. ...First, the references of the singular terms do not depend on Fregean senses, or identifying descriptions in the mind of the speaker. ...Second, each of these utterances expresses what David Kaplan has called a 'singular proposition"' (Perry, 1988: 108).

The relevant lessons, however, were first taught by Marcus and repeated with elaboration by the named individuals. In Nathan Salmon's book on Reference and Essence, he writes that versions of the New Theory of Reference were "developed, to a considerable extent independently, by several contemporary philosophers of semantics, most notably Keith Donnellan, David Kaplan, Saul Kripke, and Hilary Putnam" (Salmon, 1981: 3). Another leading contributor to the New Theory, Howard Wettstein, writes such things as "New theorists like Keith Donnellan, David Kaplan, Saul Kripke, John Perry, and Hilary Putnam -and my own work falls into this tradition -proffer an account (of direct reference that is anti-Fregean)" (Wettstein, 1986: 185-186). Quotations such as these can be picked out at will from the current literature. I was once myself among the guilty, calling Marcus' theory of proper names "the Kripke- Donnellan theory of proper names" (Smith, 1987: 387).

Occasionally her work is alluded to in the literature on the New Theory[1] but her central role is overlooked. It seems to me that, from the point of view of the history of philosophy, correcting this misunderstanding is no less important than correcting the misunderstanding in a hypothetical situation where virtually all philosophers attributed the origin of the Theory of Forms to Plotinus. This correction is the aim of the following several sections. In these sections, I will outline some of the basic ideas in Marcus' work that also occur in Kripke's "Naming and Necessity" and "Identity and Necessity". As I shall explain later, I believe the two main causes of the subsequent failure of philosophers to mention her work is that attributions to Marcus did not appear in the relevant places in "Naming and Necessity" and "Identity and Necessity" and that many philosophers may have been insufficiently familiar with Marcus' earlier contributions.


I shall quote some passages from Marcus' 1961 article that reveal six main ideas she contributed to the New Theory.

First, let us start with the idea that proper names are directly referential and are not abbreviated or disguised definitions, as Frege or Russell and most philosophers up to the 1970s believed. Marcus writes:

But to give a thing a proper name is different from giving a unique description. ...(An) identifying tag is a proper name of the thing. ...This tag, a proper name, has no meaning. It simply tags. It is not strongly equatable with any of the singular descriptions of the thing. (1961: 309-310).

This is the basis of the contemporary "direct reference" theory of proper names, where proper names are argued not to be disguised descriptions. For example, "Scott" refers directly to Scott and does not express a sense expressible by such a definite description as "the author of Waverly".

A second idea that Marcus introduces is that we can single out a thing by a definite description, but this description serves only to single it out, not to be strongly equatable with a proper name of the thing. She says: "It would also appear to be a precondition of language [especially assigning names] that the singling out of an entity as a thing is accompanied by many ...unique descriptions, for otherwise how would it be singled out? But to give a thing a proper name is different from giving a unique description" (1961: 309). This idea later became widely disseminated through Kripke's discussion of how reference-fixing descriptions are sometimes used to single out a thing as a bearer of a name, but that the names are not disguised descriptions. Kripke writes: "It seems plausible to suppose that, in some cases, the reference of a name is indeed fixed via a description [but that the description is not "part of the meaning of the name"] (Kripke, 1972: 276). Kripke, however, added the novel idea (and this is one of the main original ideas in "Naming and Necessity") that in other cases names' reference may be secured by a historical causal chain stemming back to the original "baptism" (Kripke, 1972: 298-303).

A third component of the New Theory of Reference introduced by Marcus is the famous modal argument for the thesis that proper names are directly referential rather than disguised contingent descriptions. Contrary to Nathan Salmon's claim that "The modal arguments are chiefly due to Kripke" (Salmon, 1981: 24), they are chiefly due to Marcus and presented by Kripke ("Naming and Necessity" (1972: 269f1) without reference to Marcus' earlier statement of them. Let us begin with this passage, where Marcus is discovering (10) and (15) (in her notation). (10) and (15) are

(10) The evening star eq the morning star

(15) Scott is the author of Waverly

The symbol "eq" stands for some equivalence relation. Types of equivalence relation include identity, indiscernibility, congruence, strict equivalence, mate- rial equivalence and other~. Marcus wants to argue that the equivalence relations to be unpacked in (10) and (15) are not strong enough to support the relevant theses of the "disguised contingent description" theory of proper names. She writes (1961: 308-309):

If we decide that "the evening star" and "the morning star" are [proper] names for the same thing, and that "Scott" and "the author of Waverly" are [proper] names for the same thing, then they must be intersubstitutable in every context. In fact it often happens, in a growing, changing language, that a descriptive phrase comes to be used as a proper name -an identifying tag -and the descriptive meaning is lost or ignored.

Marcus will find that not all of the relevant expressions are names for the same thing. They are not intersubstitutable in modal contexts; Marcus writes:

Let us now return to (10) and (15). If they express a true identity, then 'Scott' ought to be anywhere intersubstitutable for 'the author of Waverly' in modal contexts, and similarly for 'the morning star' and 'the evening star'. If they are not so universally intersubstitutable -that is, if our decision is that they are not simply proper names for the same thing; that they express an equivalence which is possibly false, e.g., someone else might have written Waverly, the star first seen in the evening might have been different from the star first seen in the morning -then they are not identities. (1961: 311)

Marcus' modal argument shows why the 'disguised contingent description" theory of proper names is false. Since (10) and (15) do not express identities, the expressions flanking "is" are not proper names for the same thing. In (10) and (15) a weaker equivalence relation should be unpacked, for example, by a theory of descriptions. (By contrast, the sentence "Hesperus is Phosphorus" evinces an identity sign flanked by the two expressions; thus, it passes Marcus' modal testlor containing two proper names of the same thing.)

This modal argument goes back to Marcus' formal proof of the necessity of identity in her extension of S4 (Barcan, 1946; 1947), which is a fourth component she introduced into the New Theory of Reference. She showed that

(T) (xIy) �?(xIy)

is a theorem of QS4, QS4 being her quantificational extension of Lewis' S4. The quadruple bar here means strict equivalence. Since identities are necessary, a

i failure of intersubstitutivity in modal contexts will show that a proper name does not express the relevant descriptive sense. If "Scott" is not intersubstitutable with "the author of Waverly", "Scott" does not express the sense expressed by this definite description. This opens the door to the theory that proper names do not express descriptive senses but instead are directy referential.

Of course, this argument does not prove that proper names do not express senses, merely that they do not express senses of contingent definite descrip- tions. Marcus' modal argument is consistent with the idea of L. Linsky (1977) and A. Plantinga (1978) that proper names express senses expressible by necessary definite descriptions, which are definite descriptions that express modally stable senses. For example, "Scott" may express the modally stable sense of "the actual author of Waverly".

In order to rule out this modally stable descriptive theory of proper names, one needs further argumentation, such as the epistemic argument that proper names are directly referential. If the descriptive theory of proper names is true, i.e., proper names are defined by descriptions, then "Venus is the evening star" should express a truth knowable a priori, i.e., knowable merely by reflection upon the concepts involved. But it cannot be known a priori that Venus is the evening star; this is known a posteriori, through observation of the empirical facts. As Marcus writes:

You may describe Venus as the evening star, and I may describe Venus as the morning star, and we may both be surprised that, as an empirical fact, the same thing is being described. But it is not an .empirical fact that


(17) Venus I Venus. (1961: 310)

Here "I" is the identity symbol. If "Venus" expresses the modally stable sense expressible by "whatever is actually the evening star and morning star", then the persons designated by "you" and "I" in the passage quoted from Marcus' article should be able to know a priori, simply by reflection upon the semantic content of the expressions "Venus", "the morning star" and "the evening star" that Venus is both the morning star and the evening star. The fact that they cannot know this indicates that "Venus" does not express the modally stable sense expressed by "whatever is actually the evening star and morning star". Thus we have the irony that Plantinga's and Linsky's theory of proper names was refuted years before they formulated it, unbeknowst to themselves (and unbeknownst to later New Theorists)!

Marcus' arguments for the "direct reference theory" make manifest her discovery of a fifth crucial component of the New Theory of Reference, the concept of rigid designation (although the name of this concept, "rigid designation" was first coined by Kripke). "Hesperus" is intersubstitutable salva veritate with either occurrence of "Phosphorus" in "Necessarily, Phosphorus is Phosphorus". Each of these two names actually designates Venus in respect of every possible world in which Venus exists and does not actually designate anything in respect of worlds in which Venus does not exist. If these two names were instead equivalent to contingent descriptions (e.g., "the morning star" and "the evening star"), they would not be intersubstitutable salva veritate in this modal context and thus would be non-rigid designators. Marcus notes in her 1970 APA paper, "Essential Attribution", presented at a symposium at which Kripke was one of the symposiasts, that "individual names don't alter their reference, except to the extent that in (respect of) some worlds they may not refer at all" (1971: 194).

Although I have used the "rigid designation" terminology, Marcus does not use it, since Kripke's introduction of this expression in his "Identity and Necessity" (1971) assimilated proper names to some descriptions (viz., mod- ally stable descriptions), which obscure their different semantic properties. Marcus' points can be accommodated, consistently with the continued use of "rigid designators", if we make the following classification, which is familiar to those working with the New Theory of Reference. Adopting the genus/species terminology, we may say that the genus is rigid designators, and the different species are (a) proper names, (b) referentially used definite descriptions (in Donnellan's sense), (c) attributively used definite descriptions that express a modally stable sense, (d) uses of indexicals, (e) natural kind terms, and certain other expressions. We avoid assimilating proper names to some modally stable descriptions, since proper names refer directly, whereas attributively used definite descriptions that express modally stable senses refer indirectly, via the expressed sense.

A sixth idea introduced into the New Theory of Reference by Marcus is the idea of a posteriori necessity. Recall our earlier quotation of Marcus' remarks about Venus and the evening star:

You may describe Venus as the evening star, and I may describe Venus as the morning star, and we may both be surprised that, as an empirical fact, the same thing is being described. But it is not an empirical fact that

(17) Venus I Venus. (1961: 85)

Consider the expression "Hesperus is Phosphorus". We do not know this to be true a priori. It is not an analytic assertion whose truth value is known by analysis of the concepts involved. Nonetheless, it is necessarily true, since both names directly refer to the same thing, Venus. It is true that


(a) Hesperus I Phosphorus,

whereas, as before, "I" is the sign of identity. Given Marcus' theorem of the necessity of identity, it follows that

(b) Necessarily, Hesperus I Phosphorus.

Thus "Hesperus is Phosphorus" can be viewed as a synthetic a posteriori necessary truth. This belies Salmon's historical comment about the sentence ]"Hesperus, if it exists, is Phosphorous". He writes of Kripke's three 1970 talks at Princeton, published in 1972 as "Naming and Necessity", that "In 1970 Saul J Kripke astonished the analytic philosophical community with his claim - supported by the rich theoretical apparatus of possible-world semantics and his new 'picture' of reference - that [the mentioned sentences], though synthetic and a posteriori contain necessary truths, propositions true in every possible world" (Salmon, 1986: 2).[2] A more accurate statement would be that Kripke eloquently elaborated upon Marcus' idea and extended it to new sorts of items, such as "water is H2O".

What explains the wide misunderstanding of the historical origins of the New Theory of Reference, a major movement in the history of analytic philosophy? I have already suggested that two reasons may be that many philosophers were insufficiently familiar with Marcus' earlier work and that Kripke did not attribute the relevant ideas to her in "Naming and Necessity" and "Identity and Necessity" (despite the fact that he was present when she presented her seminal work in 1962 and was undoubtedly familiar with her earlier formal papers on modal logic). There seems to me a plausible explanation of why Kripke did not make these attributions.

He writes in the Preface to the 1980 edition of Naming and Necessity that "The ideas in Naming and Necessity evolved in the early sixties -most of the views were formulated in about 1963-64" (1980: 3). Of course some of the ideas in "Naming and Necessity" are genuinely new, such as the causal chain picture of the reference of names, the idea that natural kinds are rigid designators and the theory of the necessity of origins. But since most of the views in "Naming and Necessity" occur in 1961 with Marcus' article "Modalities and Intensional Language" and in her formal work on modal logic in 1946-47, we need to look at Kripke's remark about the evolution of his views from this perspective. First, recall that Marcus presented her paper "Modalities and Intensional Languages" in February 1962 at the Boston Colloquium for the Philosophy of Science. Her paper was published, in conjunction with the colloquium, in the 1961 volume of Synthese.

Kripke was present and participated in the discussion, which was subsequently published. Kripke, it seems, did not wholly grasp Marcus' ideas at this time. During the discussion of the paper at the 1962 colloquium, Kripke made

the following remark about Marcus' theory:

The tags are the "essential" denoting phrases for individuals, but empirical descriptions are not, and thus we look to statements containing "tags", not descriptions, to ascertain the essential properties of individuals. Thus the assumption of a distinction between "names" and "descriptions" is equivalent to essentialism. (Marcus et al., 1962: 142)

This is mistaken, since Marcus clearly did not claim in her article that things have their names essentially. As Marcus later explained, "that was not my claim. Socrates might have been named Euthyphro; he would not thereby be Euthyphro" (1993: 226-227). To the contrary, her claim was that names are what later came to be described as directly referential. They are not denoting phrases, essential or otherwise. ) This suggests how we may understand the statement in the 1980 Preface to . )Naming and Necessity that "most of the views were formulated in about 1963- 64". We should interpret this as suggesting that Kripke first correctly understood Marcus' theory in 1963-64 and that before this time, he did not grasp what she conveyed in the presentation he attended. In particular, it was Marcus' theory of the necessity of identities, where names flank the identity sign, and the associated ideas of direct and rigid reference that became clear to Kripke in subsequent years. We should perhaps interpret this 1980 passage from Kripke in this light: He says "Eventually I came to realize -this realization inaugurated the aforementioned work of 1963-64 -that the received presuppositions against the necessity of identities between ordinary names were incorrect, that the natural intuition that the names of ordinary language are rigid designators can in fact be upheld" (1980: 5).

But why does Kripke not say instead that at this time he first came to understand Marcus' arguments for the necessity of identity and the directly referential and rigid character of proper names? In the 1972 essay, he attributes one idea to Marcus: "Marcus says that identities between names are necessary" (1972: 305). But instead of explaining how this idea and Marcus' other ideas formed the theoretical basis of "Naming and Necessity", Kripke goes on to criticize a minor aside made by Marcus, viz., that a good dictionary should be able to tell one if "Hesperus" and "Phosphorus" have the same reference (1972: ) 305). (But as Marcus 1ater explained, she should have made it clear that what she .\ had in mind was a "dictionary" that functions as an encyclopedia, where co- .j. referring names are listed as in a biographical dictionary (Marcus, 1993: 34, n. 1 ).) I believe a reasonable explanation of why Kripke did not attribute the central features of the "New Theory" to Marcus is that he originally misunderstood Marcus' New Theory of Reference. When he eventually understood it, after a year or two, the insight that came made it seem that the ideas were new. I suspect that such instances occur fairly frequently in the history of thought and art.[3]

Philosophy Department

Western Michigan University




Almog, Joseph: 1984, 'Semantical Anthropology'. in P. French et al., eds., Midwest Studies in Philosophy, 9,479-90.

Almog, Joseph: 1986, 'Naming Without Necessity', The Journal of Philosophy, 83,210-42. Braun, David: 1993, 'Empty Names', Nous, 27(4),443-69.

Donnellan, Keith: 1966, 'Speaking of Nothing', The Philosophical Review, 83,3-32.

Kaplan, David: 1985, 'Dthat', in A. Martinich, ed., The Philosophy of Language. Oxford University Press, New York.

Kripke, Saul: 1980, Naming and Necessity. Harvard University Press, Cambridge, Mass.

Kripke, Saul: 1972, 'Naming and Necessity', in D.Davidson and G. Harman, eds., Semantics of Natural Language. Reidel, Dordrecht, pp. 253-355.

Kripke, Saul: 1971, 'Identity and Necessity', in M. Munitz, ed., Identity and Individuation. New :} York University Press, New York.

Linsky, Leonard: 1977, Names and Descriptions. University of Chicago Press, Chicago. ; Marcus, Ruth Barcan: 1993, Modalities. Oxford University Press, New York.

Marcus, Ruth Barcan: 1971, 'Essential Attribution', Journal of Philosophy, 68, 187-202.

Marcus, Ruth Barcan: 1961, 'Modalities and Intensional Languages', Synthese, 13,303-22. [Marcus] Barcan, R.: 1946, 'A Functional Calculus of First Order Based on Strict Implication',

Journal of Symbolic Logic, II, 1-16.

[Marcus] Barcan, R.: 1946, 'The Identity of Individuals in a Strict Functional Calculus of First Order', Journal of Symbolic Lagic, 12, 12-15.

Marcus, Ruth Barcan et al.: 1962, 'Discussion of the Paper of Ruth B. Marcus', Synthese, 14, 132- 43.

McDowell, John: 1994, Mind and World. Cambridge, Mass.

Munitz, Milton: 1981, Contemporary Analytic Philosophy. Macmillan Pub. Co., New York. Perry, John: 1977, 'Frege on Demonstratives', The Philosophical Review, 86,474-97.

Perry, John: 1988, 'Cognitive Significance and New Theories of Reference', Nous, 22, 1-18.




[1] SSome philosophers have noted in passing some of Marcus' contributions,

but have not recognized their full extent. J. Almog (1986: 220, n. 8) goes further

 than most in recognizing some of Marcus' contributions (e.g., the idea of direct

 reference), but confines this remark to a footnote in an article devoted to the

 theory that proper names are rigid designators, which he believes was

originated by Kripke rather than Marcus. Another example is Alan Sidelle's

recognition that Marcus, not Kripke, first formulated the idea of the necessity

of identity (1989: 25), but Sidelle notes this in a book devoted to the idea of a

posteriori necessity, which he believes was originated by Kripke rather than

Marcus. John McDowel1 (1994: 105, n. 28) mentions the trend towards

construing some expressions as directly referential and, after referring to

Kripke and Donnellan as "early proponents of this trend", writes: "See also,

predating the trend, Ruth Barcan Marcus, 'Modalities and Intensional

Languages', Synthese, 27, (1962), 303-22". Material added in May

1995: It should be added that in publications subsequent to the ones I

quoted from Kaplan, N. Salmon, and H. Wettstein, each of them credits

Marcus for developing at least one of the ideas for the New Theory of

Reference. These appear in Salmon's Frege's Puzzle, Wettstein's Has

Semantics Rested on a Mistake? and in Kaplan's contribution to Modality,

Morality and Belief" Essays in Honor of Ruth Barcan Marcus, Walter

Simmot-Armstrong et al. (eds.). }


[2] Of course Salmon is expressing here the nearly universal misunderstanding of the origin of this idea. For another example, consider Sidelle's comment in his book devoted to the idea of a posteriori necessity: "Enter Kripke and his Naming and Necessity. Kripke made it very plausible that there are necessary truths that are synthetic and knowable onlya posteriori. Some of the more '. familiar examples are 'Hesperus and Phosphorus"' (Sidelle, 1989: 2).

[3] Even official histories, such as Munitz's Contemporary Analytic Philosophy, get it wrong. In the chapter where the ideas developed by Marcus are explained, we find the chapter title "Referential Opacity, Modality, and Essentialism: Saul Kripke".


Back to Philosophy of Language