There are two central theories of names or, perhaps more accurately, two general ideas of how names work that have developed in the philosophy of language. These two perspectives on names are well explicated in the works of Saul Kripke's lectures from Naming and Necessity and John Searle's Intentionality. Kripke's view on language, which has its roots in the writing of John Stuart Mill, is “that proper names are rigid designators” (Kripke 276). This is the position that I hope to ultimately discredit, in favor of a position similar to Searle's, in which names are set by internal conditions in a way that is consistent “with our general account of” Intentional content (308). My own view will  expand on that of Searle's in explaining how it is that names are bound to a particular set of descriptions, and how these labels persist through changes in descriptions. This will ultimately invite us to view names as much more flexible placeholders for the groups of properties that description theorists generally hold to be what a name is shorthand for. My intention is to suggest that names are not rigid designators nor a specific or essential set of descriptions, but a rough placeholder for a group of properties that can be altered.

Before expounding on my own version of description theory, it will be beneficial to review the positions of Kripke and Searle so that we can understand how description theory has dealt in the past with Kripke's causal chain theory. To be fair, Kripke has acknowledged that his views are not what could be considered a complete theory of names, and that his characterization of naming is “far less specific than a real set of necessary and sufficient conditions for reference would be” (Kripke 281). Regardless of whether or not Kripke's ideas on naming are complete enough for a comprehensive theory of names, central to Kripke's view is the idea that names rigidly designate their referents. That is, names pick out the same referent regardless of the properties of the individual to whom we are referring.[1]

An example that elucidates the differences between the rigid designator view and the descriptionist view is introduced by Kripke and effectively refuted by Searle; the story, entirely fictional, imagines that Kurt Gödel “was not in fact the author” of the famous theorem that proves the incompleteness of arithmetic (Kripke 280). The theorem was instead authored by a man named Schmidt, but fell into Gödel's hands through some circumstances of questionable scruples. As many people recognize Gödel only as the mathematician who discovered this incompleteness theorem, Kripke questions who the layperson would be referring to when using the name “Gödel” in his thought experiment. The implication is that, with the descriptionist view, the individual would have to be referring to Schmidt. Kripke's reaction to the suggestion that this might be so is one of simple incredulity, taking it as obvious that the person using the name “Gödel” must be picking out, in some way, the actual Gödel as he exists in the world. Kripke's incredulity is certainly understandable; it seems intuitive that, even if his story were true, we would not be picking out Schmidt when we describe the author of the incompleteness theorem. This thought experiment is actually not implausible, as Kripke points out himself; there are likely many people who know of Columbus only as “the first man to realize that the earth was round” (Kripke 280). Are these individuals referring to the same Columbus as an historian more educated in such matters?

The example mentioned is only one of several points that Kripke brings up, although arguments of its nature seem to be the focal point of most of Kripke's motivation for viewing names as rigid designators. Searle has a poignant and, I believe, entirely effective argument to deal with Kripke's critique, which I will explain before discussing descriptivist theory in detail for the sake of continuity. He answers Kripke by noting that intentional content is missing from the story, but shows that its omission makes quite a difference. At the very least, the individual in question almost certainly does not only interpret “Gödel” as merely the man who proved the incompleteness of arithmetic, but, though usually unnecessary to state explicitly, as the person called “Gödel” within a particular linguistic community, or at least called Gödel from whomever he “got the name” (Searle 318). Furthermore, Searle goes on, the isolated events in which the name “Gödel” is used are not irrelevant. For example, if we imagine reading the incompleteness proof itself and claiming that “Gödel made an error here,” we would intuitively, I think, want to say that Schmidt was the referent in such a scenario; this seems to be untrue if we view names strictly as rigid designators. In contrast, if an individual, knowing Gödel only as the author of the incompleteness theorem, “says 'Kurt Gödel lived in Princeton,'” we might well claim, as interpreted by descriptivist theory, that the person meant by “Gödel” something more along the lines of the person called “Gödel” within their linguistic community. Searle stresses that merely examining the speech acts is not enough, but that “the speaker's Intentional content that” is attached to the name must be considered when determining reference (318). In other words, we cannot merely examine what a speaker says, we must examine what their mental content is about.

In an admittedly reversed order, I'll move from Searle's refutation of Kripke and give a more structured account the descriptivist view of proper names. The genesis of the view comes from Gottlob Frege, who believed that “the meaning of a name is given by a single associated definite description” (Searle 313). This idea was promising, though perhaps too strict; the properties of referents inevitably change, and with them the single descriptions the names are a shorthand for. As a result, we might be more tempted to view names as the labels for something more like a bundle of descriptions, rather than a definite set. This appears to be headed in the right direction, allowing names to be more flexible types of things that can persist through a referents changing descriptions. I will come back to how this leads to a more satisfying theory of names, but to do so I must first discuss more of what Searle means by intentional content. Toward the end his essay on proper names, Searle seems to become increasingly more clear on how it is that intentional content allows names to refer to objects. He introduces four principles to be mindful of when accounting for names in philosophy of language.

The first of these principles states that, for names to even exist in the way that they do, that is, referring to things in the world, some sufficient amount of Intentional content must exist to pick out the object that a name is bound to. This seems obvious, for it is clear that we need some sort of minimal internal representation of what we are picking out with names; this must also be established merely to refer to names in any sort of context. The second principle Searle states is that, once “the connection between name and object has been set up,” speakers are able to make use of the connection of the name with the object, without knowing anything specific about the object (Searle 322). In other words, we can set a name for an object, and have the name connected purely to an internal representation of the object even if we are at a loss for words on describing the object. Searle's third principle is that all reference is accomplished by means of intentional content; this may be done “by way of names, descriptions, indexicals, tags, labels, pictures, or whatever” (Searle 322). I can think of no exceptions to this. The fourth and final principle that Searle introduces is, I believe, counterintuitive to most, but also the most critical blow to Kripke's view of names as rigid designators. Searle notes that objects, as targets for names and reference are “always determined relative to a system of representation” (Searle 322). What Searle means by this, though seems reluctant to make explicit here, is that objects only exist as distinct units with separate numerical identities insofar as we model them as such. Searle discusses this at the very beginning of his essay on proper names, though the reader may be forgiven for failing to notice just how this ties in with his final principle. Searle mentions at the start of his article on proper names that:

objects are not given to us prior to our system of representation; what counts as one object or the same object is a function of how we divide up the world. The world does not come to us already divided up into objects; we have to divide it; and how we divide it is up to our system of representation, and in that sense is up to us, even though the system is biologically, culturally, and linguistically shaped (Searle 308).

This point is insightful and a very important objection that those in the causal chain camp such as Kripke must outright deny. While Searle introduces this point, some of its implications are missed or intentionally left out. By looking at how we divide up the world, we can achieve a much clearer understanding of what it is that proper names are doing. In this vein, Searle is right in focusing on intentionality, but omits the fascinating and entirely relevant discussion of its origins and evolutionary function. Proper names, as I will attempt to persuade, are not rigidly designating something in the external world, but are the balancing points for sets of descriptions; they are abstractions which I will now baptize as, echoing one popular American philosopher, centers of nomenclatural gravity.

The Center of Nomenclatural Gravity (CONG)

To fully understand what I mean when I say that names are centers of nomenclatural gravity, I must introduce the concept of a center of narrative gravity. The center of narrative gravity, first introduced by Daniel Dennett in Consciousness Explained, is what gives us the illusion of a unified self persisting through time. This center of narrative gravity, which I will henceforth simply refer to as the “self,” is constructed when we cognitively separate properties of ourselves from the rest of the world. This separation is not anything unique; many types of boundaries between the internal and the alien are found throughout the biological world. Indeed, the need to draw such borders between what is and is not a part of an organism is paramount in any evolved entity. This fact is easily recognizable when we consider what would happen to an amoeba that was unable to differentiate between things that “belong” in the organism and foreign matter. The bacteria would be not only be unable to direct repair unto itself, but would likely run the risk of attempting to feed on parts of itself. Another amusing example, perhaps one that is easier to visualize, is the example of an octopus mentally trying move a piece of kelp to grab at its own tentacle in the same manner as it would do the opposite; the absurdity of the idea clearly illustrates why  organisms must be able to discriminate between things that are part of themselves and parts of the external world.

Human beings, of course, are no exception to the rule, and we separate ourselves from things that are foreign in a variety of ways including biochemically, immunologically, and cognitively. While the former methods are certainly interesting in themselves, the manner in which we separate ourselves cognitively is of concern at the moment. Nearly every human being alive performs the instinctual act of telling stories about what is going on in the world; this act is no more conscious than a spider spinning its web.[2] The evolutionary advantage of spinning a self is especially clear for social animals that must keep track of social politics and use language effectively. The self is “spun” in effort to convince others “to posit a unified agent” (Dennett 418). These tools over here belong to me; it was I who gathered food for the party; the swelling in my leg seems to have subsided. All of the previous clauses project[3] a unified agent that is separate from the rest of the word. In this manner, the internal I is not a physical part of the brain dependent upon any essential properties or medium, but is rather more of an abstraction constructed by the process of a machinery that projects a unified agent.

This projection of a self, while unique in that the properties it groups insofar as such properties are attributed to the machinery doing the projecting, is not an exceptional activity. Human beings, along with, almost certainly, most members in the animal kingdom, project all kinds individual entities onto the world. Each projection, as with Dennett's center of narrative gravity, exists to present properties of perception as unified particulars, each with a numerical identity that is equally as abstract and internal as the self. I make it sound like this activity leads to a gross misrepresentation of how the universe exists externally, but modeling the world in such a way is a needed means of getting around. By separating all of the properties of what I project as the cup of tea in front of me and all of the other properties I observe in my environment, I am able to make a mental model of and use language to describe a variety of acts that can be performed with a closely knit set of particles occupying a general area of my desk. I say general area because the hard demarcations perceived become gradually more fuzzy as we zoom in on a much wilder microscopic world where particles of all sorts fly in and out of the region of whatever material it is that composes china. Water molecules evaporate into the surrounding area, and what looks like matter occupying a definite region in space is really moving around constantly. The boundaries drawn, however, are certainly useful; perceiving my cup of tea and my desk as a single object might result in some amusing behavior. Likewise, although the particles of tea inside my cup will readily separate from the particles of china of my cup, I can recognize what might otherwise be projected as two separate entities as being functionally bound (barring any unfortunate spills) and refer to a “cup of tea” as a single object on my desk.

It is by labeling projections, that names function, and we must therefore abandon the idea that names are “rigid designators” of sorts that pick out some external referent. Referents cannot be picked out by names because referents do not exist externally, but are projected as a means of modeling the world more clearly. As such, the names themselves are not pinned down to anything specific, but are like the “I” used to focus a set of properties to a center of narrative gravity. Centers of nomenclatural gravity (CONGS) focus groups of properties in the same way, and are just as malleable in their ability to persist through an objects changing of properties.[4] CONGS are, usually very flexible, collections of properties grouped under a single label; in this regard, I present my own theory of names as a version of Searle's descriptionist theory, with a few meaningful alterations that will attempt to offer novel solutions on some supposed problems associated with Searle's idea of names.

Searle and Kripke Revisited

With a new idea of what names are, I will revisit Kripke and Searle and provide a slightly different solution to the problems addressed to description theory by Kripke. Let us readdress Kripke's fictional tale of the less than ethical means by which Gödel obtained his proof of the incompleteness of arithmetic. Adopting the CONG view, and drawing on some of Searle's earlier insights, we can envision our layperson having a specific Intentional content about Gödel, with the name “Gödel” being the CONG for all of the properties attributed to Gödel. Although Kripke has limited us to giving the individual only the knowledge that Gödel found the system of mathematics to be incomplete, we can reasonably assume that such a person would have rational prejudices as far as the additional properties Gödel would be likely to have. For instance, the layperson of interest would likely assume Gödel to be human, intelligent, and all of the properties that these two categories entail. There would be less obvious properties as well, such as those stated by Searle, (e.g. “the man called 'Gödel' in my linguistic community,”) that most people would be unlikely to state explicitly (318). It would not be trivial to say that an assumed property of Gödel would be that others refer to him as “Gödel,” that other individuals knew him personally, or that, apart from his renowned feats in mathematics, there are several other, yet unknown, properties to separate the one known as “Gödel” from any other entity in existence.

When we finally arrive at the story of what is going on when this person uses the name “Gödel,” the answer seems obvious. No object in the external world is being directly selected in the way that Kripke would want to assert, rather the world is being modeled in a way that associates the discoverer of the incompleteness of arithmetic with the name “Gödel,” along with all the other properties described. We have a CONG that has combined properties that simply would not be grouped together if the person grouping them had more information at hand. In effect, “the one called 'Gödel'” and “the author of the incompleteness theorem” are linked to a single CONG, but would be no longer when our layperson discovers that Schmidt is, in fact, the true author of the theorem. It is by grouping together properties in a way that effectively models the way the world actually is that makes it appear as though we are picking out, with names, absolute things with numerical identities in the external world. In effect, we are not picking out actual entities at all, and names do not designate things such as tea cups or mathematicians because these things do not exist as objects outside of our Intentional content. The world outside of Intentional content, undivided, contains no referent; the referent, instead, is constructed by the mind by grouping properties together and as such is giving a name to these groups. This is what allows objects to persist through changing properties, as the brains modeling of the external world would be of little use if a new object were projected with every changing property. As a result of this need for seeing objects persisting through time, we often are puzzled by how objects may radically change their properties, yet still seem to be the same entities.[5] These objects persist with the same names as the properties grouped together are altered; should we feel that a group of properties has changed sufficiently, or that a perceived referent has lost one or more of what we consider essential properties, a new CONG can be constructed, as distinct from the old.[6]

Before concluding, I would like to note that this view that posits a name as a CONG might contribute to our perceptions of fictional characters. The name “Sherlock Homes,” for instance, can be viewed, just like “Kurt Gödel,” as a CONG associated with a group of properties. Such properties would include pipe smoker, detective, Englishman, and, though I fear I may be biting off a bit too much here, fictional character. A fictional character is simply a collection of properties grouped in the Intentional content of individuals. Sherlock Holmes, in this regard, is a projection just as much as Kurt Gödel, but Gödel is model of a (albeit ephemeral) collection of animated matter that exists in time and space.[7]

In conclusion, we can see that the view of names as rigid designators is inadequate, as the objects these names are believed to designate do not exist externally. Instead, recognizing the importance of Intentional content in naming, we can see that the way the brain models the world relies on the grouping of properties together to form such objects with distinct numerical identities. These groups of properties work in a similar way as Dennett's “self,” in which we set a boundary between the properties of ourselves and the rest of the world. As a result, all names can be viewed as centers of nomenclatural gravity, which are constructed by the separation of a particular group of properties from all other properties in existence.

Dennett, Daniel C. Consciousness Explained. 1st ed. Boston: Little, Brown and Company, 1991. 101-430.

Kripke, Saul. "Naming and Necessity." The Philosophy of Language. Ed. A P. Martinich. New York: Oxford UP, 2001. 272-287.

Searle, John R. "Proper Names and Intentionality." The Philosophy of Language. Ed. A P. Martinich. New York: Oxford UP, 2001. 308-324.