Reference is a relation that obtains between a variety ofrepresentational tokens and objects or properties. For instance, whenI assert that “Barack Obama is a Democrat,” I use aparticular sort of representational token—i.e. the name‘Barack Obama’—which refers to a particularindividual—i.e. Barack Obama. While names and other referentialterms are hardly the only type of representational token capable ofreferring (consider, for instance, concepts, mental maps, andpictures), linguistic tokens like these have long stood at the centerof philosophical inquiries into the nature of reference. Accordingly,and to keep things to a reasonable length, this entry will focusprimarily on linguistic reference.^[1]

Assuming that at least some token linguistic expressions really dorefer, a number of interesting questions arise. How, for example, doeslinguistic reference relate to the act of referring—somethingthat we as speakersdo with referential terms? How exactly doreferential terms come to refer? That is, in virtue of what do theyrefer to what they do? Is there a single answer to this question, asingle mechanism of reference, or different answers depending on thesort of term in question—or even the circumstances in which asingle term is used? And what exactly is the relationship betweenreference and meaning? Answers to these various questions will turnout to be closely related, and the task of this entry will be to traceout some of the main clusters of answers.

• 1. Introduction
• 2. Proper Names
  • 2.1 Descriptivist Theories
  • 2.2 Millian Heirs
  • 2.3 Names with Multiple Bearers
• 3. Indexicals
  • 3.1 Pure Indexicals
  • 3.2 Impure Indexicals
• 4. Definite Descriptions
• 5. Four Models of Linguistic Reference
• 6. Nihilism, Particularism, and Pluralism
• Bibliography
• Academic Tools
• Other Internet Resources
• Related Entries

1. Introduction

We use language to talk about the world. Much of what we say about theworld appears to be meaningful; some of it is presumably even true.For instance, I seem to be saying something true when, in theappropriate sort of setting, I utter:

1. Barack Obama is a Democrat.

How do we manage to do such things? How, for instance, do I manage totalk about Barack Obama and thereby say meaningful and true thingsabout him? In a word: how do Irefer to Barack Obama by meansof the name ‘Barack Obama’? Metaphorically, we seem to becapable of using language to talk about the world because some of ourwords are themselves capable of ‘hooking onto’ things inthe world, things like Barack Obama. Proper names—that is,expressions like ‘Barack Obama’ and ‘MountKilimanjaro’—are widely regarded as paradigmatic referringexpressions. Although it may seem implausible to suppose thatall words refer, that all words somehow ‘hookonto’ bits of reality, certain types of words are widelypresumed to be of the referring sort. These include: proper names,pronouns, indexicals, demonstratives, plurals, natural kind terms, andvarious other sorts of property terms. Definite descriptions areanother, though highly controversial, candidate. Here, we’llfocus on just a subset of these—namely, what are often called‘singular’ referential terms. These are terms that,supposing they refer, refer to particular objects and individuals asopposed to groups or properties. Since singular referential terms havebeen subjected to intensive philosophical scrutiny over the pasthundred years, this will still leave us with plenty to discuss.

This article will focus on five closely related questions regardingreference: (i) How does the reference relation that obtains betweentoken uses of terms and objects relate to the act of reference? Inother words, do terms refer to what they do in virtue of our usingthem to do so, or do we use them to refer in virtue of their alreadydoing so? (ii) What are the meanings of referential terms? Are theyjust referents, or rather something more? (iii) What is the mechanismof reference? In other words, in virtue of what does a token use of areferential term attach to a particular object/individual? (iv) Isthere a single mechanism of reference common to all referring terms,or do different sorts of terms hook onto their referents in virtue ofdifferent sorts of things? And (v) to what extent, if at all, canreference depend on relatively private features of the speaker, suchas their intentions or other mental states?

Our goal here will be to map out how answers to these variousquestions cluster together to generate several distinct pictures ofthe nature and function of reference. To get there, we’ll startin Section 2 by looking at two of the main approaches to proper names.Then, in Section 3, we will turn to indexicals and demonstratives,which put pressure on the thought that these two approaches representan exhaustive set of options. Section 4 will focus on definitedescriptions, which will serve to highlight some potentialdeficiencies in the models of linguistic reference introduced inSections 2 and 3. Section 5 will step back to outline the mainclusters of positions we will have developed by that point. Finally,Section 6 wraps up by canvassing some more radical positions, likenihilism and pluralism about reference.

For the sake of both clarity and brevity, this entry will refrain fromventuring too deeply into the detailed debates regarding the meaning,syntactic form, and function of the various terms we will be lookingat. Rather, we will focus instead on what, if anything, can be saidabout the nature of linguistic referencein general. For amore detailed look at the idiosyncrasies of these various terms,please see the entries onnames,descriptions, andindexicals. See also the related entry onnatural kinds.

2. Proper Names

Proper names have long taken center stage in debates about linguisticreference. For present purposes, we’ll treat these as roughlyco-extensive with what ordinary (non-philosophically trained) speakersstandardly call ‘names’. So expressions like ‘BarackObama’, ‘Kyoto’, and ‘Mount Kilimanjaro’will all count as proper names for our purposes. What do theseexpressions have in common? In virtue of what do they constitute agenuine class of linguistic expressions? At least at first glance,these would appear to be syntactically simple expressions that refer,or at least purport to refer, to particular objects or individuals.Thus, ‘Barack Obama’ refers to a specific man,‘Kyoto’ refers to a specific city, and ‘MountKilimanjaro’ refers to a specific mountain. And, even though itis questionable whether expressions such as ‘Santa Claus’and ‘Sherlock Holmes’ actually refer to anything, therecan be no doubt that they at least purport to refer: to Santa Clausand Sherlock Holmes, respectively. They are thus to be counted asproper names as well for present purposes.

With respect to the reference of proper names, there are two basicorientations that have long captured philosophers’ attention:one that views names on the model of tags, and another that views themon the model of descriptions. We’ll briefly survey these twoviews, along with some complications that arise for each. Thenwe’ll turn to a problem that arises for both sorts of view:explaining how reference works for names with more than onebearer.

2.1 Descriptivist Theories

According to descriptivist theories of proper names, a particular useof a proper name refers when the descriptive content somehowassociated with that use suffices to pick out a specific object orindividual. On one standard way of working out this sort of view, oneassociated with both Gottlob Frege (1892) and Bertrand Russell (1911),the particular descriptive content associated with a given use of aname is so associated becausethe speaker associates thiscontent with the name in question. For the use of the name to refer,this descriptive content must uniquely determine the name’sreferent. So when a speaker uses the name ‘N’and, in so doing, successfully refers to a particular object orindividualx, this sort of descriptivist claims (i) that thespeaker must be thinking ofN as the (unique)F and(ii) thatx must in fact be the (unique)F. In otherwords, this sort of ‘classical’ descriptivist posits thatreferential success hinges on speakers attaching to each name in theirrepertoire some descriptive contentF which uniquely singlesout a specific object in the world. Conversely, when speakers fail toassociate a sufficiently precise description with a name, this sort ofdescriptivist predicts that reference fails.

Classical descriptivists, like Frege and Russell, were perfectlywilling to acknowledge that the descriptive content in question mightvary, sometimes quite markedly, from one speaker to the next. Indeed,according to Russell, such contents may vary across time for one andthe same speaker. Thus, while I might associate the name ‘DavidCameron’ with the descriptive contentthe U.K. PrimeMinister who called for a referendum on Brexit, Samantha Cameronmight associate the same name with the descriptive contentmyhusband. Had Cameron ultimately chosen not to call a referendumon Brexit, my identifying content associated with ‘DavidCameron’ would presumably have been different. If David andSamantha were to divorce, then Samantha’s identifyingdescription would no doubt change as well, perhaps to something likemy ex-husband. In all of these cases, the individual referredto by means of the name is determined (or, as it is often put, is‘picked out’ or ‘fixed’) by the descriptivecontent the speaker associates with that name. Because the descriptivecontent in question is typically characterized by means of a definitedescription (an expression of the formthe F), such theoriesare often known as ‘descriptivist theories’ of proper names.^[2]

To get the intuitive appeal of descriptivism more clearly in view,consider a case where we know two individuals named‘Boris’, one of whom is a mutual friend who has never heldany elected office and the other of whom is the former Prime Ministerof the United Kingdom. Suppose now that I utter:

2. Boris likes to party.

If for some reason you are confused about who I am talking about, thenatural thing for you to do is to ask me “Which Boris do youmean?” I might then respond “The former Prime Minister ofthe U.K.,” and this seems to be dispositive of the facts here.That is, supposing that this is the description I had in mind, then itwould seem that my utterance of (2) is genuinelyabout Boristhe former Prime Minister and not our mutual friend, thenon-politician Boris. Likewise, the truth or falsity of (2) would seemto hinge on how things are with the former Prime Minister and not onhow things are with our mutual friend of the same name.

Some other appealing aspects of descriptivism become evident when wepair the thesis considered above, a thesis about what determinesreference, with a natural companion thesis to the effect that thedescriptive contents associated with uses of names also provide theirtoken meanings. In other words, proper names may well refer, but theyonly do sovia their meanings—which are more likedefinite descriptions. By adopting this further thesis, thedescriptivist can now explain a range of philosophically interestingcases. So, consider:

3. Hesperus is Phosphorus.
4. Santa Claus lives at the North Pole.
5. Fred believes that Cicero, but not Tully, was Roman.

(3) is true, but it is not knowableapriori. That is,knowledge of (3) cannot be justified independent of experience. If themeanings of these two names is just whatever they refer to, however,then this looks puzzling. For then (3) would meanVenus isVenus, which is just an instance of the law of self-identity. Inother words, simply understanding the meaning of (3) would be enoughto know that it is true. According to meaning descriptivism, incontrast, what (3) means is that two distinct descriptions aresatisfied by the same object. Given what the relevant descriptions arelikely to be here, this should only be knowableaposteriori,or on the basis of experience. Similarly, in (4) the meaningdescriptivist can explain the meaningfulnes of the sentence withoutappeal to reference, by associating the name ‘Santa Claus’with a description likethe bearded, grandpa-like figure who runsa dodgy elf labor camp in the far north. And, finally, thepotential truth of (5) can be explained by noting that Fred might wellassociate different descriptions with ‘Cicero’ and‘Tully’, despite the fact that these names actuallyco-refer.

The central challenge to the descriptivist theory is that there isreason to suspect that proper names are not semantically equivalent todefinite descriptions. Saul Kripke (1972), in particular, argues thatnames pick out the same object even when embedded under modal termslike ‘might’, whereas definite descriptions typicallydon’t do this. Returning to our earlier example of ‘DavidCameron’ and supposing that the description I associate withCameron isthe U.K. Prime Minister who called for a referendum onBrexit, it seems that I can still truly assert:

6. David Cameron might not have called for a referendum onBrexit.

If descriptivism is right about both meaning and reference, however,then (6) should be equivalent to:

7. It might have been the case that: the U.K. Prime Minister whocalled for a referendum on Brexit did not call for a referendum onBrexit.

Granted, (7) is not really a claim of ordinary English. Still, holdingfixed that we are talking aboutmetaphysical rather thanepistemic possibility here, it should be clear enough that(7) is false: if there is someone who satisfies the description‘the U.K. Prime Minister who was called for a referendum onBrexit’ in whatever possible situation we are considering, thenthat individual called for a referendum on Brexit in that samepossible situation.

In more ordinary English, there is a way of hearing something alongthe lines of (7) as true: assuming that we are talking about the PrimeMinister whoactually called for a referendum on Brexit, ofcourse they might not have. But that is not how definite descriptionsare typically assumed to function; rather, they are typicallyunderstood to be flexible, ornon-rigid across the space ofpossibility, picking out whatever happens to satisfy them in whateverpossible situation we are considering. So, if the descriptivist is toavail themself of this sort of defensive maneuver, as some have beentempted to, then they will have to motivate the claim that names aresemantically equivalent not to definite descriptionsper se,but rather to ‘actualized’ or ‘rigidified’versions thereof.^[3]

This challenge can be avoided by a descriptivist who is willing togive up on the claim that the meaning of a proper name, on a givenoccasion, is equivalent to a definite description. In that case, anassociated description will fix reference relative to the actualworld, and then that referent is what is relevant to determining thetruth of modal statements. This response, however, requires giving upon the nice explanations of (3)–(5) that the meaningdescriptivist was able to provide. What’s more, even this moreminimal version of descriptivism runs into a different problem alsoraised by Kripke, what is often called the ‘semantic’objection to descriptivism.

This objection runs as follows: often, we don’t associate enoughinformation with a name to pick out any particular individual.Nonetheless, we seem to be capable of using that name to refer to aspecific individual. Kripke offers as an example the name‘Feynman’. Ordinary folks, Kripke claims, might know thatFeynman was a physicist, but they will not know anything besides thename that would serve to differentiate Feynman from any otherphysicist they have heard of. An indefinite description likeaphysicist will not suffice, however, to pick out any particularindividual in the world. Evena physicist named‘Feynman’ won’t do, at least in a world wheretwo physicists bear this name. At best, this sort of description willpick out an arbitrary member of a class of individuals, not the rightone consistently. And yet, as Kripke points out, it seems perfectlycoherent for someone who knows nothing about Feynman, who has onlyoverheard someone else using the name, to say to themself “Iwonder who Feynman is,” or to ask their friend “Who isFeynman?” In each of these cases, the natural thing to say isthat the speaker is using the name ‘Feynman’ to wonder orask about Feynman. How they manage to do so, however, looks to besomething that is going to be very difficult for the descriptivist toexplain—assuming (i) that they want to maintain the link betweenassociated descriptions and the information available to individualspeakers, and (ii) that they are unwilling to rely on descriptionslikewhoever the person I overheard this name from was using it totalk about.

Before moving on, it is worth briefly noting that some descriptivistshave indeed been tempted to defend the theory by giving up on the linkmentioned in (i). P.F. Strawson (1959), for instance, suggests thatspeakers may rely on others to provide the relevant descriptivecontent, the content that serves to hook a given use of a name onto anobject or individual in the world. In fact, Strawson allows thatgroups can effectively use names to refer so long as there is at leastone expert among them for whom for whom the following holds: when wepool expert opinion, a plurality of the descriptions they associatewith the relevant name are true of a single object. This willeffectively deal with Kripke’s Feynman case, but at an intuitivecost. For now the descriptivist has not only forfeited their abilityto explain (3)–(5), they have also risked making it the casethat speakers will have no special access to who or what they aretalking about when using a name. If speakers regularly lack sufficientinformation to identify the referent of their use of a proper name,then it is highly unclear why we would be justified in relying on themto answer questions like “WhichN were you talkingabout?” Some may be tempted to pay this price. Others may betempted to try to integrate bits and pieces of descriptivism intotheories which are otherwise anti-descriptivist. In fact, this latteroption has proven a popular one, and much of what follows can beviewed as a study in how this strategy has played out with respect todifferent sorts of referential terms.

2.2 Millian Heirs

The primary alternative to the descriptivist theory of names hastypically gone by the name of ‘Millianism’. According tothis view, which dates back to John Stuart Mill (1867), a name’smeaning is simply its referent. In its modern form, the view wasintroduced by Ruth Barcan Marcus (1961), who proposed that we ought toconceive of proper names as ‘tags’. To say that propernames are tags is, for Marcus, to say that they have no linguisticmeaning beyond their reference. Proper names do not, on this sort ofview, refer by way of the descriptions they allegedly stand for.Rather, they referdirectly, as it is sometimes put, to theirbearers. Important consequences of this theory include, as Marcusnotes, the necessity of identity statements between co-referringproper names—something which, though highly intuitive, is notguaranteed by most descriptivist theories of proper names.^[4] Other important consequences include the dissolution of puzzlesinvolving substitutivity in modal contexts (Marcus 1993).

Of course, saying that names function as tags—or that they are‘directly referential’—is not to provide a fulltheory of names. That requires, in addition, an explanation of whatmakes a name the particular tag that it is. In other words, we need tospecify what it is, if not an associated description, that fixes thereference of a name in a context. Put slightly differently, the claimthat names function as tags effectively furnishes us with asemantics for names. What remains, is to provide ametasemantics for names, a theory that tells us whichsemantic value should be associated with each name, or each name on agiven occasion of use, and why.

The most popular option has been to pair a Millian semantics with ametasemantic picture adumbrated at roughly the same time by PeterGeach (1969), Keith Donnellan (1970), and the aforementioned Kripke(1972). Typically called the ‘causal theory of reference’,the central idea developed in these works is that (the use of) a namerefers to whatever is linked to it in the appropriate way—a waythat does not require speakers to associate any identifyingdescriptive content whatsoever with the name. The causal theory isgenerally presented as having two components: one dealing withreference fixing, the other dealing with reference borrowing.Reference, on this sort of view, is fixed by a dubbing. In otherwords, a language usergives a name to an object by sayingsomething like “You are to be calledN.” Theparadigm case is one where the dubber is occurrently perceiving thetarget object when they utter this. After this initial act ofreference-fixing, the name gets passed on from speaker to speakerthrough communicative exchanges. Speakers succeed in referring tosomething by means of its name, on this sort of view, becauseunderlying their uses of the name are links in a causal chainstretching back to the initial dubbing of the object with that name.Subsequent speakers thus effectively ‘borrow’ theirreference from speakers earlier in the chain, though borrowersneedn’t be able to identify any of the lenders they are in factrelying on. All that is required is that borrowers are appropriatelylinked to their lenders through chains of communication.

As Kripke points out, complications arise due to the fact that we canapparently re-use names. So, I may have come across the name‘Napoleon’ via a chain of use leading back to the mostfamous of French generals. Having heard the name, I may now decide tocall the hedgehog who lives in my front garden, and who likes toimperiously survey her domain, ‘Napoleon’. When I use thename in this way, Kripke claims, that I have introduced a new name, orat least a new use of the name. This, Kripke claims, is due to myintentions—specifically, my intention to dub this hedgehog withthe name ‘Napoleon’ as opposed to using that name as Ialways have before. Still, there is obviously a real sense in which Iinherited the name from a historical tradition that traces back to18th century France. So not every sort of causal connection toprevious use suffices to preserve a chain of reference borrowing,according to the causal theorist. Some connections will suffice topass on reference, whereas others are irrelevant. Providing aconvincing account of the details here has proven to be a non-trivial task.^[5]

2.3 Names with Multiple Bearers

So far, we have generally proceeded as though names were univocal. Butthat hardly seems right. As we already saw, a name like‘Boris’ can refer equally well to the former British PrimeMinister and to anyone else whose parents decided to call them this,regardless of their politics. Likewise, once I’ve named thehedgehog in my garden ‘Napoleon’, it would seem that thename ‘Napoleon’ can be used equally well to talk abouteither the greatest of French generals or, alternatively, thisparticular hedgehog. Classical descriptivists have no real troubleaccounting for this, since each time a name is used, its reference isessentially fixed anew by whatever description the speaker happens tohave in mind. In other words, according to this sort of descriptivist,it is really only a matter of happenstance that uses of names everco-refer. That might seem a rather large bullet to bite. On the otherhand, once we move away from classical descriptivism, explaining hownames can refer to different individuals in different circumstancesbecomes significantly more challenging.

Consider a descriptivist in the Strawsonian mold, one for whom anutterance of the name ‘N’ refers in accord withthe plurality of the beliefs of the relevant experts. What are we todo with a name like ‘Boris’ then, which can refer toeither the former British Prime Minister or to our mutual friend? Wecannot simply aggregate all the relevant beliefs associated with thisname, considered as an orthographic or phonological form, and then seewho or what the plurality of those beliefs picks out. For that wouldleave us only ever able to use the name to refer to one or the otherof these Borises, and that hardly seems like the right thing to say.It seems, therefore, that a descriptivist of this sort will have tothink about names in some more fine-grained manner. Yet this will notprove easy. They cannot simply go back to appealing to the informationthe speaker happens to have to do the job, for we can easily imaginesomeone who seems capable of refering to either Boris and yet lackssufficient information to distinguish the one from the other.

Similar issues arise for the Millian. Here, however, some concreteproposals have been floated. For instance, David Kaplan (1990) hasargued that, against all appearances, names really do have only onereferent apiece. In fact, Kaplan suggests that we should think ofnames as individuated, in part, by their referents. In other words, onKaplan’s view, one cannot simply ‘read off’ whichname has been used from its overt phonological or orthographic form.Rather, Kaplan posits that there are any number of names all writtenand pronounced ‘Boris’, or even ‘BorisJohnson’. While each of these names is written and pronouncedidentically, each refers to a different person. The question nowbecomes: what determineswhich name a speaker has used in agiven context? Kaplan suggests that the answer to this question has todo with the speaker’s mental states, and in particular with thespeaker’s intention to talk about one or another individual. Ineffect, Kaplan takes it that hearing a name leaves us with a mentaltrace, a mental analogue of a linguistic name, that directly refers tosome particular individual. By looking to see which trace shows up inthe speaker’s intentions, we come to know the referentialproperties of the utterance or inscription being used to express thismental trace.

Kaplan’s theory has some rather odd consequences, however. Forinstance, it turns out that no two people actually ever share aname—for names come complete with their referential features.^[6] Kaplan suggests that what people share are not names but rather somecommon aspects of how their names tend to be externalized. In otherwords, for Kaplan, what it is to share a name with someone else is forboth of your names to belong to the same class of phonological and/ororthographic forms. Kaplan calls these ‘generic names’, asopposed to the ‘common currency names’ which, for him, arethe bearers of semantic properties like reference.

Finding themselves dissatisfied with Kaplan’s proposal here,other direct reference theorists have opted for different responses tothe problem of names with multiple bearers. For instance, bothFrancois Recanati (1997) and Michael Pelczar and Joe Rainsbury (1998)have suggested that names ought to be treated on the model ofindexicals, a sort of term that we will consider in more detailshortly. Briefly though, the suggestion runs: there is one and onlyone name ‘Boris’, but context makes clear to whichindividual a particular utterance of this name refers. In effect, theproposal preserves something of the causal theory—by allowingthat baptisms and passings-on of names are the right sort of thing todetermine the set of possible referents associated with a name at anygiven time—while appealing to some further feature of thecontext to do the work of selecting an individual from this set. Incontrast to the pure indexicals like ‘I’ or‘here’, however, there is no obvious candidate for whichaspect of the context serves to make this selection. Things get worsestill once we consider contexts in which multiple people bearing thesame name are all present and being talked about in the course of asingle utterance (e.g. “I never would have expected Boris andBoris to get on so well.”).^[7] Alternatively, one might posit that names somehow function more likethe true demonstratives ‘this’ and ‘that’,though obviously the details of such a suggestion would need to beworked out.

To better understand these last two proposals, we first need to getclearer on how reference works for in the case of indexicals anddemonstratives. In the next section, we will work to fill in thisgap.

3. Indexicals

We have now seen two basic models of how words refer to things. On thedescriptivist model, words refer by being associated, somehow, with adescription that serves to isolate a particular object as thereferent. By varying how we associate the relevant descriptions withparticular uses of words, we derive different versions ofdescriptivism. On the causal-perceptual model, in contrast, words areassociated with chains of use leading back to some original act ofdubbing. That act itself then serves to bridge the gap between wordand world. While both these models were developed with names in mind,we should ask ourselves “Can either serve to explain linguisticreferencein general? That is, can either model plausiblyextend to other sorts of referential terms, beyond justnames?”

To answer this question, and ultimately to introduce a third distinctmodel of linguistic reference, we turn now to the indexicals: wordslike ‘I’, ‘you’, ‘here’,‘now’, ‘he’, ‘she’,‘this’, and ‘that’. As we will soon see, it ishardly clear that all indexicals refer in the same way. In particular,a distinction has often been drawn between what are called‘pure’ and ‘impure’ indexicals, with ratherdifferent theories of reference being offered for each. The challengesthat arise in trying to offer accounts of the impure indexicals willbring us back to one of the questions with which we began: namely, dowords refer because we use them to do so, or rather do we use them todo so because they already refer?

3.1 Pure Indexicals

What are pure indexicals? Roughly, these are expressions the referenceof which appears to co-vary with certain very regular aspects of thecontexts in which they are used. Here, ‘context’ should beunderstood as including,inter alia, a speaker, hearer, time,and place. In contrast, the reference of ‘impure’indexicals is supposed to be more difficult to characterize in termsof picking out some distinct, repeatable feature of a context. Boththe existence and the significance of this distinction arecontroversial (see the entry onindexicals for further discussion). That said, assuming that there is such adistinction, it is widely held that ‘I’,‘here’, ‘now’ are pure indexicals, whereas‘this’, ‘that’, ‘he’,‘she’, and ‘it’ are impure. ‘You’is perhaps the most active locus of dispute (see Radulescu(2018)).

The traditional approach to indexicals, dating back once more to Fregeand Russell, has it that the reference of such expressions is fixed bysome sort of descriptive content associated by the speaker with theexpression. This reference-fixing description is the meaning of agiven utterance of the expression. The motivation for such a view islargely intuitive. Indexicals must mean something, and their meaningspresumably have something to do with how these expressions refer. Soit is not too much of a stretch to think that ‘I’ mightmean something likethe speaker of this utterance and referto that individual. Similarly, ‘now’ might meanthetime of this utterance and refer to that time. And so on.

One obvious objection to this view is that what the term‘I’ refers to does not appear to be sensitive to whateversort of descriptive content a speaker happens to associate with thatterm. For instance, the fact that I might happen to associate thedescriptionthe 44th President of the United States with theterm ‘I’ does not mean that I can somehow succeed in usingthe term ‘I’ to refer to Barack Obama. Another problemwith this view, discussed extensively in Kaplan (1989b), is thattaking these reference-determining descriptions to be a part of whatwe assert when we use indexicals can lead to our making some ratherodd predictions. Consider an assertive utterance of:

8. I like to ride my bicycle.

Suppose, first, that I am the speaker. I utter (8). Now, suppose thatyou are the speaker. You utter (8). While there may be a loose sensein which we ‘said the same thing’ with our utterances,there is another, stricter sense in which we clearly did not. I saidsomething about myself, whereas you said something about yourself.Yet, according to the classical descriptivist, we’ve both saidthe very same thing. Namely, we’ve both asserted that thespeaker likes to ride their bicycle. All that differs is the contextin which we asserted this.

Now, to be clear, there are ways of tinkering with the view to avoidboth these objections. Some, like Hans Reichenbach (1947) or morerecently Manuel García-Carpintero (1998) and John Perry (2001),have argued that indexicals are ‘token reflexives’,meaning that the descriptions that should be associated with them willneed to involve explicit reference to the utterance of that very tokenuse of the term. So, for instance, the description for ‘I’might be: the speaker of this very token of ‘I’. Sincedifferent tokens will be involved when each of us utters (8),we’ll no longer be asserting the very same thing. If we addthat, for token indexicals at least, the relevant descriptionsaren’t under the speaker’s control, but are ratherassociated with particular terms via the rules of language, then wecan avoid our earlier objection as well.

A different sort of approach was developed by Reichenbach’sstudent, the aforementioned David Kaplan. Kaplan (1989b) took many ofthe same elements that Reichenbach was working with, but put them intoa framework where the meaning of an indexical in context—thatis, its contribution to what is said or asserted—is just anobject. According to Kaplan, we need to distinguish between two typesof meaning, which he called ‘character’ and‘content’. Content is basically what we have been callingmeaning to this point; it is what the utterance of an individual termcontributes to what is said or asserted by an utterance of thecomplete sentence of which it is a part. Character, on the other hand,is more akin to a rule of use; a character tells us, for any givencontext, what the content of a given expression is. Names, onKaplan’s way of thinking about them, have constant characters:in any two arbitrary contexts, uses of the same name will be mapped tothe same referent (recall that Kaplan thinks of names in afine-grained way, such that they are never shared). The rule for anyname ‘N’ thus turns out to be: in any contextwhatsoever, returnN as the referent. Not so for terms like‘I’ or ‘here’, which Kaplan takes to beassociated with rules likereturn the speaker andreturnthe location of the utterance. So in a context where I amspeaking in Foyle’s, my utterances of ‘I’ and‘here’ will refer to me and to Foyle’s,respectively. When Nat is speaking at the American Bar, his utterancesof ‘I’ and ‘here’ will refer to Nat and to theAmerican Bar.

Importantly, the characters of the pure indexicals are supposed to beinsensitive to speakers’ mental states. That rules out anypossibility of my using the term ‘I’ to refer to BarackObama. Nor are your and my utterances of (8) predicted to say the samething. What’s more, as Kaplan points out, the view allows us toproductively distinguish between ‘metaphysical’ necessityand what Kaplan calls ‘logical’ necessity. The sentence“I am here now,” Kaplan claims, represents a logicalnecessity: in virtue of what the indexicals ‘I’,‘here’, and ‘now’ all mean, this sentencecannot be uttered falsely. Yet clearly it is not metaphysicallynecessary that a particular speaker be wherever they happen to be atthe time of utterance; they could just as easily have been somewhere else.^[8]

Now we have three basic models of reference on the table: thedescriptivist model, the causal-perceptual model, and the charactermodel. With the last of these in view, we can clarify an importantaspect of the indexical theory of names that was discussed at the endof the last section. Essentially, that theory proposed to hybridizeour second and third models. Instead of names having constantcharacters, as Kaplan would have it, the proposal is to assign themmore interesting ones. For instance, we might try stipulating that thecontext, in addition to containing a speaker, place, time, etc., mustalso include a most salient individual bearing a given name. Names,the idea runs, always refer to their most salient bearer at a context.Since names no longer have constant characters, they are more likeindexicals than we might initially have thought. On the other hand, wecan preserve a good deal of the causal theory by re-characterizing itas a theory about who counts as a name-bearer in a given context.^[9]

This way of developing the indexical theory is not without itsdownsides, however. For instance, it hardly seems to be true that wealways use names to refer to the most salient bearer of that name in acontext, particularly once we consider utterances of names prefixedwith phrases like ‘the other’ (e.g. “The other Davidis not here yet, though I can hear him down the hall.”).What’s more, there is reason to worry that embracing this sortof indexical theory entails giving up on one of the purportedadvantages of the causal theory: its ability to explain how it is thatwe can pass on the capacity to think about objects merely by passingon names for those objects. Since names are causally anchored to theobjects they name, according to the causal theory, acquiring a newname should suffice to put us in causal contact with the object named.On the indexical theory of names, on the other hand, we do not pass onnames for things; rather, we use names to refer to the most salientbearer of those names in a given context. Perhaps this suffices topass on ause of a name—loaded, as it were, with areferent—where the listener doesn’t already have that usein their repertoire. But if the listener fails to have this use intheir repertoire, then it becomes hard to see how the referent of thatuse could count as the most salient bearer of the relevant name in thecontext.

3.2 Impure Indexicals

In contrast to the pure indexicals, the reference of deictic uses ofpronouns like ‘he’, ‘she’, and‘it’ and true demonstratives like ‘this’ and‘that’ looks to be far less amenable to an analysis interms of character. After all, what regular feature of a speechcontext might any of these terms serve to pick out? One option wouldbe to say that tokens of each of these simply refer tothedemonstratum of the context. But not only does that threaten toforce us to say that all of these terms effectively mean the samething—something which seems rather counterintuitive—now wealso need to know what exactly serves to make something thedemonstratum in a context. And what are we to do when multiple suchterms, or multiple instances of the same term, are used within thespan of a single utterance? Are we really to believe that contextsinvolve not just a demonstratum role, but also afirstdemonstratum role, asecond demonstratum role, etc.?

A number of possible responses arise at this point. First, followingKaplan (1978) and Colin McGinn (1981), we might try to take advantageof the fact that uses of these terms are often accompanied byostensive gestures. The picture this suggests is: the reference of ause of a demonstrative (and, presumably, the other impure indexicalsas well) is determined by the accompanying ostensive gesture.Formally, we might treat these gestures as determining the firstdemonstratum, second demonstratum, etc. But we might equally wellchoose not to. Unfortunately, this approach faces two rather obviousobjections: first, even fairly directed pointing gestures can usuallybe disambiguated in a number of different ways; and, second, manyperfectly good uses of demonstratives simply aren’t accompaniedby any sort of ostensive gesture. The first of these problems invitesa simple solution: appeal to the speaker’s intentions todisambiguate the gesture. That move, in turn, invites the question:why not just appeal directly to these ‘directingintentions’ and leave the physical aspect of the gesture to theside? We will return to that option shortly.

The latter objection invites a rather different response: perhaps itis not the ostensive gesture itself that makes an object thedemonstratum, but rather what that gesture does. Gestures, we mighttake it, serve to make objects salient in a context. This leads us toa second sort of view about what makes an object the demonstratum of aparticular utterance of an impure indexical: whatever object is themost salient in the context at the time of utterance, that is thedemonstratum. Howard Wettstein (1984) and Allyson Mount (2008) haveeach developed views along these lines, the basic idea being that anutterance of a term like ‘she’ refers to whoever counts asthe most salient woman in the context and with similar rulesapplying to the other impure indexicals. More problematic are bareuses of ‘this’ and ‘that’, which don’tseem to contain any substantive information about the relevant sortal.This means that we need to posit that contexts in which these termsrefer contain not just a most salient woman, but a most salient objectfull stop. What might serve to make an object maximally salientsimpliciter? The natural suggestion would seem to be: ourmutual interests and perspective.

Dialectically, this suggestion is problematic. For, as EliotMichaelson and Ethan Nowak (2022) point out, the interests andperspective of the speaker and listener can easily diverge. One optionhere would be to claim that, in cases of divergence, one of theseperspectives (i.e. the speaker’s or listener’s)systematically trumps the other. Opting for either of these, however,invites the question of whether we can dispense with the notion ofsalience entirely and make do instead with how the privileged agentconstrues the relevant term, regardless of whatever happens to besalient to them at the moment. In essence, the worry is that the viewmight well collapse into either a speaker intention-type view or alistener-construal type one. The natural way of trying to avoid such acollapse would be to claim that reference simply fails when no oneobject is mutually maximally salient. This is the route that Mountadvocates taking. As she notes, this way of going entails that a greatmany seemingly meaningful and felicitous uses of ‘this’and ‘that’ fail to refer. For instance, the view entailsthat reference fails whenever the speaker uses the term‘this’ but the listener can’t hear them, isdistracted, etc. While Mount is prepared to accept this result, fewothers have been enthusiastic about this prospect.

For those not so inclined, a third possible response presents itself.According to Kaplan (1989a), token uses of the impure indexicals, likedemonstratives and deictic uses of pronouns, refer to whomever orwhatever the speaker intends for them to refer to—at least for‘perceptual’ uses of such terms, or uses where the speakeris occurrently perceiving their target. Let us leave this hedge aside,however, and allow that such intentions might be grounded in memorytraces, the conversational record, or what have you. Withperceptually-grounded uses of demonstratives, this directing intentionmight well be thought of as the internal analogue of an externalizedpointing. But the move inside makes this picture more durable: noexplicit gesture is necessary for reference to succeed, and no longerare we limited to using the impure indexicals just to refer to objectsin our physical and temporal vicinity. Since we do indeed use impureindexicals to refer to physically and temporally distant objects, thislooks like a significant advantage for a generalized‘intentionalist’ approach to demonstrative reference.

None of this is to say that Kaplan’s view has been universallyembraced. On the contrary, the view is generally thought to face arather serious problem. We noted above that terms like the baredemonstratives ‘this’ and ‘that’ look to berather flexible in their application. Yet even these terms arepresumably notinfinitely flexible. For instance, youpresumably cannot point directly and deliberately at a picture ofBoris Johnson, utter (9), and succeed in using the term‘that’ to refer to the picture of David Kaplan that youhave tucked away in your desk drawer for just such occasions:

9. That is a picture of one of the most Hawaiian shirt-lovingphilosophers of the twentieth century.

This is an instance of what has sometimes been called the‘Humpty Dumpty Problem’.^[10] Basically, any time we might be tempted to suggest that what tokenuses of some particular term mean or refer to depends on thespeaker’s intentions, the following sort of problem inevitablyarises: unless we impose some constraints on what those intentions canlook like and still prove effective, we end up having to say thatspeakers with sufficiently-bent intentions can mean or refer to somehighly unexpected things by means of their utterances, things that,intuitively, it does not seem possible to mean or refer to withutterances of the relevant terms.

A closely related case, oddly enough from an earlier time-slice ofKaplan (i.e. Kaplan 1978) and used there to help motivate theostention theory, can be used to help bring out the issue here.Imagine a scenario where the speaker is sitting at their desk andtheir prized picture of Rudolf Carnap, which typically hangs on thewall behind them, has been switched for a picture of Spiro Agnew.Having failed to notice the switch, and intending to use the term‘that’ to refer to their picture of Carnap, the speakerpoints behind themself, directly at the picture of Agnew, and utters(10):

10. That is a picture of one of the greatest philosophers of thetwentieth century.

According to Kaplan, (10) is false not in virtue of the utterance of‘that’ failing to refer, but rather because the speakerhas asserted something false about the actual picture hanging behindthem, the picture of Agnew.^[11] Note that this case differs from (9) in an important respect: in(10), it doesn’t seem altogether implausible that the speakerhere mightalso have intended for their use of‘that’ to refer to whatever picture is hanging behindthem. After all, they are somehow confused about which picture thatis. This confusion would seem to offer the intentionalist additionalroom to maneuver.

In fact, intentionalists have gone several different ways here: AlexRadulescu (2019) has embraced the claim that the token uses of‘that’ in (9) and (10) do refer to the pictures of Kaplanand Carnap, respectively. Most intentionalists, however, have not beenso bold. Instead, they have either tried to restrict the range ofintentions that serve to fix reference, or else they have tried tolimit when speakers’ intentions in fact determine reference.This first strategy has generally proceeded by appealing to certainconsiderations drawn from Paul Grice’s (1957) theory of meaning.The second, in contrast, has tended to try to derive the relevantconstraints from the standing meanings of, or rules of use for, theimpure indexicals and demonstratives. More recently, a fewphilosophers have also tried combining these two approaches.

The Gricean approach to reference was first clearly articulated byGail Stine (1978), though it had been alluded to by both Grice himselfand Keith Donnellan (1968).^[12] The basic idea is that the relevant sorts of intentions for fixingthe reference of impure indexicals are, properly speaking, intentionsaimed at getting the listener to identify a particular object as thereferent. In keeping with Grice’s broader theory of meaning,these intentions are posited to be even more complex than this: notonly are they aimed at getting the listener to identify some object asthe referent, they are also meant to be fulfilled in part on the basisof the listener’s recognizing the speaker’s intention forthem to do so.

Having followed the Gricean line this far, now we must ask: can theGricean thesis regarding the complexity of referential intentionssomehow cut off the threat of a collapse into Humpty Dumpty-ism? Theanswer, according to the Gricean, is “Yes.” Griceans arelikely to claim that the speaker lacks a genuinely referentialintention in a case like (9), since they should have no expectationthat their listener will be in a position to recover their intendedreferent. In a case like (10), however, it looks like the speakerdoes have an intention for their listener to recover thepicture of Carnap as the referent. What then is the Gricean to do?Kent Bach (1992) helpfully suggests the following refinement to theview: while genuinely referential, the speaker’s intention torefer to the picture of Carnap runsvia an intention to referto whatever picture happens to be behind them. That latter intentionis meant to be recognized directly, whereas the former can berecognized only indirectly, via the recognition of both this latterintention and (in the good case) the fact that the picture behind thespeaker actually is the picture of Carnap. A better way of formulatingthe Gricean thesis with regards to referential intentions, Bachclaims, reflects this sort of asymmetry: when there is a conflictbetween a direct and an indirect referential intention, and both aregenuinely referential, it is the direct intention that serves to fix reference.^[13]

So far, so good for the Gricean. But just how general is theirsolution to the Humpty Dumpty Problem? Can they, for instance, accountfor a speaker who appears to be fully competent in a language likeEnglish, but who mistakenly believes that everyone else can simplyread off their referential intentions, directly? In other words,consider a speaker who believes that the listener can read their mindregarding their referential intentions, though nothing else, and onlywhen they utter a demonstrative. Unless we regard such a speaker asincompetent with the English language, it would seem that they can nowsucceed, according to the Gricean, in making a token of‘this’ or ‘that’ refer to whatever they like,on any occasion whatsoever. For this speaker can coherently intend fortheir listener to recover that object as the referent of their use ofthe demonstrative, partly in virtue of directly recognizing this veryintention, regardless of what that object is. In other words, givensome sufficiently strange background beliefs, the Gricean predictsthat a speaker really can make a use of ‘that’ in anutterance like (9) refer to their picture of DavidKaplan—regardless of whether anyone short of a real clairvoyantwould be capable of recovering this referent. So the Gricean strategyfor blocking intentionalism’s collapse into Humpty Dumpty-ismdoesn’t fully generalize.

Partly in response to worries like these, we might think that theright sort of response to the Humpty Dumpty Problem is not to up thecomplexity of referential intentions themselves, but rather to imposelimits on how particular sorts of terms—even the truedemonstratives—can be used to refer. Marga Reimer (1991, 1992)offers a suggestion along just these lines:when uses of thedemonstratives ‘this’ and ‘that’ areaccompanied by ostensive gestures, then the referent itself must liein the general direction indicated by that gesture.^[14] When there is no gesture, then there is no such constraint. In otherwords, Reimer re-emphasizes the apparently special connection betweendemonstratives and gestures that earlier theorists tried to leverageinto a full theory of demonstrative-reference. This time, however,Reimer suggests that we use this connection not to generate a completetheory of demonstrative reference, but rather to impose someconstraints on the intentionalist theory. Those constraints, in turn,are what allows Reimer to make the intuitively correct predictionsregarding (9) and (10): with respect to (9), she will say that onecannot succeed at pointing to one picture and referring to anotherpicture, unless perhaps the first is a photograph of the second; withrespect to (10), she will say that the speaker intendsbothto refer to the picture behind herself and to that of Carnap, but onlyone of these satisfies the constraint associated with the gesture.Thus, only one of these intentions is in a position to determinereference.

To be clear, Reimer’s position is by no means incompatible withthe Gricean theory of referential intentions. One might, in fact,accept both that there are substantive constraints on reference thatderive from the meanings of the impure indexicals themselves, and alsothat referential intentions are as the Gricean describes them to be.Bach (2017) has recently embraced such a view, for instance. Thatsaid, one should demand an independent reason for adopting the Griceanthesis. For the Gricean thesis has often been advertized as a responseto the Humpty Dumpty Problem, but Reimer-esque constraints allow theintentionalist to avoid such a collapse independent of adopting theGricean thesis and, as we have seen, the Gricean thesis alone is notsufficient. What’s more, as Michaelson (2022) has argued, theGricean picture of referential intentions comes at a cost: it makes itdifficult, if not impossible, to explain how speakers can obfuscatewith referential terms. That is, it makes it difficult to explain howspeakers can seemingly refer to one thing with the use of areferential term while simultaneously intending for their listener tomisinterpret their utterance. Still, given the enduring appeal ofGricean theories of meaning, it is undoubtedly too soon to write offthe Gricean theory of reference.

The fourth and final view of demonstrative reference that we willconsider starts with the contention that we took a wrong turn at theoutset by giving up on the idea that what a demonstrative refers to isjust the demonstratum of the context. According to Una Stojnić,Matthew Stone, and Ernie Lepore (2013, 2017), demonstratives refer towhatever is at the ‘center of attention’ in the discourse.This might sound a bit like the salience theories we considered above,but for Stojnić et al., what is at the center of attention isdetermined purely linguistically. Sometimes this determination is dueto earlier discourse and, in particular, the sorts of discourserelations that these authors take to be grammatically encoded in ourutterances. At other times, the object at the center of attention willbe determined by a gesture, which Stojnić et al. take to have alogical form of their own. As Nowak and Michaelson (2020) point out,however, this latter claim leaves Stojnić et al. needing tosomehow disambiguate pointing gestures, just as the ostention theoristdid.

What’s more, the view faces a different sort of challenge aswell. Consider a discourse-initial, gestureless use of demonstrativeslike:

11. That was unexpected.

If I utter (11) to Clayton, sitting in stunned silence the morningafter the election of Donald Trump, it seems plausible that I willhave both said something true about the election result and managed tocommunicate this to Clayton. Yet since the utterance isdiscourse-initial, there is nothing in the prior conversation to setthe center of attention. Likewise, since there is no gesture, thatcannot be what determines the center of attention here. Instead,Stojnić et al. propose to treat utterances like (11) as,essentially, complex quantified statements; what (11) says issomething likethere is some pair of events such that the onesummarizes the other and has something unexpected as its center.In effect, Stojnić et al. propose a theory on which most uses ofdemonstratives succeed in referring as a matter of grammar, butdiscourse-initial, gestureless uses of demonstratives simply do notrefer at all. And likewise for the other impure indexicals.

4. Definite Descriptions

In contrast to the sorts of terms we have considered so far, theprimary question regarding definite descriptions—complexexpressions like ‘the biscuit’ or ‘theoff-license’—is not how they refer, but whether they referat all.

Both Alexius Meinong (1904) and Frege (1892) thought that they did.That left them with a problem, however, since the overt descriptivematerial found in definite descriptions is typically insufficient toidentify a single object as the referent. For Frege, at least, therewas a fairly obvious solution: treat that overt descriptive materialas just a part of the sense of the definite description, allowing morematerial to be added by whatever fuller description the speakerhappens to have in mind. Faced with this same set of issues, Russell(1905) took the opposite tack: he posited that descriptions simply donot refer. Rather, what he offered was a translation procedure forsentences of the form ‘The A is B’, which he posited tomean: there is one and only one A, and that A is also B.^[15]

This dispute becomes even more interesting when one considers thatboth Russell and Fregeagreed that names were to beunderstood as, in some sense, akin to definite descriptions. Now weare in a better position to see the deep differences between them inspite of this superficial agreement. For Frege, that claim in no wayprevented names from being genuine devices of reference. For Russell,it did. For Frege, a name’s contribution to the truth or falsityof a sentence was, intensional contexts to the side, its referent. ForRussell, it was the associated description.

Why suppose that definite descriptions do not function to refer? Afterall, we certainly seem to use them to talk about particular objects.When I say “I want the sloop to the right of that one,” Iseem to be expressing a desire to possess a particular sloop, not justa desire to be relieved from slooplessness in some complex fashion.What’s more, we can substitute definite descriptions for namesand indexicals in most linguistic contexts while preserving truth.Naively then, it might seem reasonable to posit that, if names andindexicals refer, then so too do definite descriptions. Russell,however, did not lack for motivation in his claim that definitedescriptions do not function to refer. Consider assertive utterancesof the following sentences:

12. The King of France is bald.
13. The author ofMiddlemarch was the third child of Robertand Christiana Evans.

(12) is meaningful, but certainly not true. Russell takes it to beplainly false, though that is controversial (see Strawson (1950) and,more recently, Von Fintel (2004)). Regardless, if one thinks thatdefinite descriptions are devices of reference, then it’sunclear how one will be justified in claiming that sentences like (12)are meaningful if the description fails to refer. With (13), theproblem facing the referentialist is different: since both thesedescriptions should refer to the same individual, Mary Ann Evans (or‘George Eliot’), it becomes highly unclear why (13)shouldn’t be knowableapriori. As with identitystatements involving co-referring names, this looks to be merely aninstance of the law of self-identity. And, presumably, any instance ofthat law is knowableapriori.

Note that Russell’s theory, in contrast, has no difficultyexplaining any of this. (12) just makes a claim about there being aunique king of France, a claim which is both perfectly meaningful andstraightforwardly false. (13), on the other hand, makes a claim aboutthe unique satisfier of one set of properties also uniquely satisfyinganother set of properties. That should hardly ever count as knowableapriori.

Still, there are also reasons to question Russell’s claim thatdefinite descriptions don’t function to refer. For instance, asStrawson (1950) points out, many assertive utterances involvingdefinite descriptions will seem true even when the relevantdescription fails to have a unique satisfier. So, consider:

14. The table is covered with books.

Assertively uttered in a context where there is a single table stackedhigh with books, (14) would seem to be true. Yet Russell predicts theopposite, so long as there is at least one additional table somewhereor other in the universe.

Strawson’s alternative was to claim that there are genuinelyreferentialuses of definite descriptions. Used in this way,definite descriptions will, like names and indexicals, contribute anobject or individual to the content asserted by the utterance—solong as that object satisfies the relevant descriptive material. If itdoesn’t, or if there is no such object, these token useswon’t refer and, according to Strawson, the utterances of whichthey are a part will fail to be either true or false. Still, they willcount as meaningful so long as the description in question isassociated with a coherent rule of use. Meaningfulness, according toStrawson, hinges on its being clear what a given expression can beused to do; reference, on the other hand, requires being in the rightkind of situation, a situation where the relevant rule of use issatisfied. Strawson doesn’t seem to object to there beingattributive uses of definite descriptions, which functionmore or less as Russell suggested, in addition to referential ones. Hejust doesn’t focus on those. Indeed, Strawson seems tocountenance there being a number of different distinct uses ofdefinite descriptions, including a kind or generic use (e.g.“The whale is a mammal”) which has only come in forserious philosophical scrutiny much more recently.

Keith Donnellan (1966) goes even further than Strawson, claiming thatdefinite descriptions can be used to refer even when the putativereferent fails to satisfy the explicit descriptive material. In whatis probably his most famous example, Donnellan asks us to consider anutterance of the following:

15. Who is the man drinking the martini?

Suppose that we are both looking at a man who appears to be drinking amartini, but who is in fact drinking water, when you utter (15). OnDonnellan’s telling, this is irrelevant; your utterance of thedescription ‘the man drinking the martini’ can refer tothis man regardless of whether he is in fact drinking a martini. Whatmatters, according to Donnellan, is just that you have this man‘in mind’ when using the relevant description, that youuse the description with this man as your intended target.

One might worry, as does Alfred MacKay (1968), that this leavesDonnellan open to the charge of offering a Humpty Dumpty view ofdefinite descriptions—a view on which there are no substantiveconstraints on what token descriptions can be used to refer to.Interestingly, Donnellan (1968)’s response is to appeal to theGricean theory of meaning, thus raising the possibility what he meansby ‘having in mind’ is something akin to theGricean’s notion of having a referential intention. As we sawabove, however, there is reason to think that even this sort of movewon’t suffice to head off the collapse of intentionalism intoHumpty Dumpty-ism in a fully general sense. So there may still bereason to think that Donnellan’s willingness to jettison theexplicit descriptive material as a constraint on referential successis perhaps a bridge too far.

But what of the intuitions behind Donnellan’s example? For, evenif we reject his claim that a description like ‘theF’ can be used to refer to something that isn’tanF, it still seems as though communication is possible in acase like (15). That is, I should be able to tell who you’reasking about regardless of whether this man is in fact drinking waterrather than a martini. Kripke (1977) suggests that the way to dealwith cases like this one is to view the sort of reference involved asmerely pragmatic, as having to do with communicative efficacy ratherthan with the sort of strict aboutness relevant to determining truthor falsity in a context. According to Kripke, there is no need toposit that definite descriptions ever refer. Instead, we can explainaway their apparent referential properties by noting that, even ifthey do not refer, speakers will typically still have targets in mindfor them to latch onto. So long as we are reasonably good atrecovering those targets, we will then be in a position to direct eachother’s thought to particular objects and individuals by meansof utterances of definite descriptions without needing to posit thattoken instances of these expressions themselves ever function to lockonto those objects or individuals. In other words, we are free toaccept the Russellian theory of descriptions, supplemented with thisclaim that descriptions are very often used to direct listeners’attention to the ‘speaker’s referent’, as Kripkecalls what he takes to be the referential analogue of Gricean‘speaker’s meaning’ (or, roughly, whatever thespeaker is trying to get across to the listener, literally orotherwise).

Kripke seems to have in mind that genuine linguisticreference—or what he calls ‘semanticreference’—needs to run purely via convention. Hisparadigm example is names, which he takes to be associated with aconvention to the effect that their reference is fixed via an appealto causal chains. Alternatively, he might have pointed toKaplan’s character-based approach to the pure indexicals, a viewwhich is equally reliant on conventions to establish reference. As wesaw above, however, there are plausibly limits to this sort ofapproach: a conventionalist approach to the reference of impureindexicals, like demonstratives and pronouns, has proven challengingto offer. Likewise, once we take seriously the problem of names withmultiple bearers, it becomes far less clear that the reference oftoken uses of names can be fixed by anything so simple as purelinguistic convention. Rather, in both cases, it looks like it isgoing to be rather tempting to appeal to facts about the speaker inorder to fix linguistic reference. But if such an appeal is acceptablein that instance, the case for ruling out definite descriptions asnon-referring starts to look weaker.

If, unlike Kripke, we start with the idea that linguistic referenceis, at root, the result of something thatwe do rather thanjust something thatwords themselves do, then definitedescriptions start to look more like the normal case and the trueindexicals like the outlier. This is, in fact, an old idea, one thatdates back at least to Susan Stebbing (1943) and the later work ofLudwig Wittgenstein (1958), and which we already saw running throughthe ideas of Grice, Strawson, and Donnellan.^[16] The enduring worry, of course, is that without some restrictions onhow we can use terms to successfully refer, we will end up divorcingthe theory of reference from both our intuitions on cases and from anyrole in an overall theory of communication. But if we embrace therebeing limits on referential success—imposed, perhaps, by Griceanreflexivity, conventionalized constraints, or both—these worriescan plausibly be avoided. Indeed, with respect to definitedescriptions the case for there being a conventionalized constraint onreferential success, as Strawson took there to be, can lookparticularly appealing.

More recently, a rather different sort of challenge has emerged forthose claiming that token definite descriptions sometimes refer:namely, the difference in meaning which is standardly assumed to markdefinite descriptions off from their indefinite counterparts (that is,descriptions like ‘a philosopher’ as opposed to ‘thephilosopher’) has itself been called into question. Two mainreasons have been offered for rejecting a difference in meaningbetween these sorts of phrases. First, pinning down exactly what thisdistinction is supposed to amount to has proven remarkably difficult;every aspect of Russell’s original analysis (existence,uniqueness, and satisfaction of the relevant descriptive material),for instance, has subsequently been called into question. Second, asPeter Ludlow and Gabriel Segal (2004) point out, a great manylanguages turn out to lack anything like the definite/indefinitedistinction. This raises the possibility of this distinction being farless significant, and less well-defined, than it is usually taken to be.^[17]

How would rejecting the definite/indefinite distinction affect debateson whether token uses of definite descriptions can refer? Well, if itwere clear that indefinite descriptions could not be used to refer,and if these were indeed semantically equivalent to definitedescriptions, then we would have an additional reason to reject theidea that definite descriptions can be used to refer. On the otherhand, while it is certainly a minority position, Charles Chastain(1975) and Janet Dean Fodor and Ivan Sag (1982), among others, haveargued that indefinite descriptions toocan be used to refer.Even if we were to reject this view, advocates of the unitary theorystill typically grant that there is some communicative or pragmaticdifference between the use of ‘the’ as opposed to‘a’. So it would be open to the stalwart defender of theview that token definite descriptions can refer to try and argue thatsuch pragmatic markers can play a significant role in regulating theuse of descriptions to refer—despite these markers being nowherereflected in meaning at the level of what is said or asserted.Obviously, none of this will be settled here.

5. Four Models of Linguistic Reference

Our original goal was to lay out some different models of the natureand function of linguistic reference. By looking at how philosophershave attempted to account for the referential features of variousdifferent sorts of linguistic terms, we have effectively introducedfour distinct ways of understanding how referential terms come torefer to particular objects and individuals. In other words, we haveintroduced four distinct models for the what is often called the‘metasemantics’ of referential terms:

1. On thedescriptivist model, words refer in virtue of beingassociated with a specific descriptive contents that serve to identifyparticular objects or individuals as their referents. This descriptivecontent might be associated with a given term on a particular occasionin a number of different ways: in virtue of its being something thespeaker associates with the term, in virtue of its being what thecommunity associates with the term, that the relevant experts in thecommunity associate with the term, etc.

2. On thecausal model, words refer in virtue of beingassociated with chains of use leading back to an initiating use or‘baptism’ of the referent. Extending this model beyondnames has proven difficult, but one option is to insist that it isreally the perceptual connection that underlies most baptismal eventsthat runs the show. In that case, perceptually-grounded uses ofdemonstratives, deictic pronouns, and definite descriptions can befolded into the picture relatively easily, with anaphoric uses treatedas something akin to links in a chain of reference-borrowing.^[18]

3. On thecharacter model, words refer in virtue of beingassociated with regular rules of reference. Paradigm rules of thissort involve only repeatable, publicly-identifiable elements of thecontext and serve to connect particular terms to particular elementsof the context.

4. On theintentionalist model, words refer in virtue ofbeing used, intentionally, to refer to particular objects. In otherwords, words refer in virtue of their being uttered as part of complexintentional acts which somehow target particular objects orindividuals.

As should be clear by now, these four models aren’t strictlyexclusive of each other. One might, for instance, conceive of theclassical descriptivism—which appeals to the speaker’smental state in order to associate a particular description with aparticular token referential term—as a particular kind ofintentionalism. Presumably, speakers will only manifest the right sortof mental state when they are intentionally using a referential termto refer. Likewise, one might view the causal theory as a special caseof the character theory: in the case of terms like names, theinvariant rule of use associated with a particular name is itself tobe explained in terms of a chain of uses leading back to a specificobject or individual. In fact, if one is willing to allow thecharacters of certain sorts of terms, like the demonstratives, toallude to the speaker’s referential intentions, then charactertheorists can countenance a certain sort of intentionalism within thescope of their theory as well. In doing so, however, it is unclearwhat remains of the original character model, with its reliance onpublicly-identifiable aspects of the context as the primary driver ofthe rules of reference.

It’s worth noting some clear divisions among these options.First, there is a divide between views according to which referentialterms themselves refer, and that we thus use them to refer becausethey already do so, and views according to which it isus whodo the referring, with token linguistic reference following onlyderivatively. The causal theory and classical, intention-free versionsof the character model are paradigm instances of the former sort ofview, whereas the intentionalist model is a paradigm instance of thelatter. If we extend the character model to include characters thatappeal to the speaker’s intentions, then the character modelwould seem to straddle this divide, with some terms being used incontext because of what they refer to, independently of us, and othersreferring because we use them to.

A second important divide has to do with the extent to which referenceis allowed to depend on potentially idiosyncratic and relativelyinaccessible aspects of the speaker’s mental state. As we sawabove, both classical descriptivists and intentionalists allow for astrong dependence on the speaker’s mental state, and this canlead to some odd predictions regarding reference. Griceans andconstraint-type theorists like Reimer have offered ways of trying tominimize these results, but one might also worry that allowingreference to depend on the speaker’s mental state at allundercuts our ability to explain how we use referential terms tocoordinate our object-directed thoughts with each other—in otherwords, to communicate about specific things in the world. On thecausal model, for instance, we literally give each other ways ofthinking about the same object by passing on a name. That makesexplaining communication rather straightforward, supposing, that is,that we are consistently able to recognize which name has been utteredon a given occasion. The problem of names with multiple bearers meansthat this might turn out to be non-trivial; but the payoff, should webe able to make good on this sort of story, looks significant.Alternatively, on the Kaplanian theory of true indexicals, we canexplain our ability to coordinate our thought on particular places,times, and individuals in virtue of the fact that the place, time, andspeaker of the utterance are things that are typically publicknowledge, equally available to every member of the conversation. Thequestion is whether such an analysis can possibly be extended toinclude terms like the impure indexicals.

Third, and finally, there is an important divide between models ofreference which areexpansive as opposed tonarrow.Both intentionalism and descriptivism are clearly expansive. They aretheories of linguistic referenceper se, purporting to applyequally well to any sort of referential term whatsoever. The causalmodel, as it is traditionally understood, is a narrow model ofreference; it only purports to apply to the reference of names. Theextended version of the view is more expansive, but even that view hasnever been applied to the pure indexicals. Once more, the charactermodel proves difficult to place; just how broad or narrow it proveswill depend on just how far we are willing to stray in our theorizingfrom the paradigm case of the true indexicals.

Expansive theories hold out the promise of a substantive account ofthe reference relation. That is, expansive theorists can claim thatthe reason that token uses of referential terms play a certain sort ofrole in explaining the truth or falsity of an utterance is that acertain sort of relation obtains between that token use and someobject in the world. Narrow theorists, in contrast, will have to allowthat there are different sorts of relations that serve to fixreference for different sorts of referential terms, and even perhapsdifferent uses of the same term. Thus, they will have to say that whatunifies referential terms is not that, when they succeed in referring,a particular sort of relation obtains between that token use and someobject or individual, but rather that they play some specificfunctional role in determining the truth conditions of an utterance,its assertive content, or what have you. In other words, standingbehind debates over which of these models we should adopt is a muchlarger, and as yet unresolved, question: just what are our explanatoryambitions in offering a theory of reference in the first place? Are wetrying to understand some unified natural kind, which might then, inturn, be used to explain a range of further linguistic phenomena? Oris it rather that there is a range of further linguistic phenomenathat we have some grip on, and positing something like reference ismerely a convenient, or perhaps even necessary, way of explainingthese things? Whichever way we go here is clearly going to have somesignificant downstream effects on the related question of what, ifanything, relates linguistic reference to its apparent analogues inart, cartography, mentation, and so on.

6. Nihilism, Particularism, and Pluralism

So far, this article has mostly taken it for granted that there is aphenomenon of linguistic reference which is worth investigating. Butthere are some who have doubted that there is any such relationbetween words and world and others who have doubted that there is justone such relation worthy of investigation.

Concerns of this sort are hardly new. W.V.O. Quine (1960), forinstance, famously claims that reference is ‘inscrutable’,or that there is simply no matter of fact what a given referentialterm refers to. His arguments, however, depend on certainmethodological constraints that many would now be inclined to reject.Not so with the related ‘problem of the many’, popularizedat roughly the same time by Peter Unger (1980) and Geach (1980). Tosee the issue, consider someone using an utterance of‘that’ to refer to a cloud in the sky. What exactly isthis cloud? The obvious answer would seem to be: a set of waterdroplets suspended in the air in such-and-such region. But what aboutsome droplet right on the edge of that region? Should it be counted ornot? In fact, there will be innumerable such droplets, and we seem tohave no systematic way of answering this question: if we say‘yes’, we face a continuous march outwards; if we say‘no’, we face a continuous retreat. Neither option seemseven remotely satisfactory, and yet if we cannot provide an answer tothe question of what exactly the cloud is then it might well seem thatwe have equally well failed to answer the question of what therelevant use of ‘that’ refers to.^[19]

More recently, pressure on the traditional assumption that referentialterms typically refer to particular objects has come from otherdirections. For instance, Richard Kimberly Heck (2014) has suggestedthat, once we distinguish what the speaker takes a given term to referto from what the listener does, there is no further work left for someother, ‘objective’ notion of reference to do. And MarioGómez-Torrente (2019) has argued that the best response to thesorts of enduring disagreements we see with respect to many complexcases of reference, like Kaplan’s Carnap/Agnew case, is toaccept that reference is simply indeterminate in such cases.

Responses to such challenges have been varied. Heck can be read eitheras a nihilist about linguistic reference, suggesting that there isnothing worthy of the name, or as a limited sort of pluralist,suggesting that there are two reasonable claimants to that title andthat neither of them takes priority over the other.Gómez-Torrente, in contrast, suggests that reference isconventional but that the conventions governing many of ourreferential terms are incomplete. We can, he suggests, sketch a set ofsufficient conditions for referential success and failure. But thesewon’t cover every case, and nor do they offer any guarantee thattheir outputs will not be in conflict with each other. None of thesesuggestions does much to address the problem raised by Unger andGeach, though I think it is safe to assume that both Heck andGómez-Torrente might want to say that this is a problemprimarily for either metaphysicians or philosophers of perception, andthat philosophers of language should feel free to adopt whateversolutions eventually emanate from those quarters.^[20]

A more radical option is mooted in Christopher Gauker (2008), whosuggests that there is simply no possibility of offering a systematictheory of demonstrative reference. Rather, on Gauker’s view, thebest one can do is point to the sorts of factors that might fit intoall things considered judgments about reference. While Gauker stopsshort of this, a natural extension of his view would apply it to thefull panoply of referential terms. In that case, there would turn outto be facts of the matter about reference even in cases where we tendto disagree—though a full explanation of those facts willforever prove evasive. One might even go so far as to suggest thatthis sort of particularistic approach to linguistic reference canserve to resolve the problem of the many as it applies to linguisticreference: on any occasion of successful reference to an object, therewill be an answer to the question of what the material constitution ofthat object is. It’s just that we will hardly ever, if ever, bein a position to know the answer. Whether we should consider this asatisfying response to that problem is, of course, another matterentirely.

Finally, Nowak and Michaelson (2022) have suggested that some of theseissues can be overcome by embracing a pluralistic approach toreference. Like Heck, they take it that sometimes we will beinterested in the speaker’s construal or the listener’sconstrual of the relevant term. But unlike Heck, they argue that thereis a robust sense in which both the speaker and listener can be wrongabout what a given utterance of a referential term referred to. Forinstance, consider a version of Kaplan’s Carnap/Agnew case wherethe speaker is unaware of the switch and the listener, only halfpaying attention, takes the speaker to be talking about a picture ofBoris Johnson that they saw in the paper earlier. Heck’sappraisal of this case will be that the speaker and listener’sconstruals are enough for us to say that communication has notsucceeded. But Nowak and Michaelson contend that there are otherthings we might be interested in, like: how would we settle a bet if Ihad wagered that the speaker would say something true and you hadwagered that they would say something false? If we do indeed haveconsistent judgments on these sorts of questions, and if thosejudgments depend on a sort of reference that isn’t reducible toeither the speaker or the listener’s construals, then it lookslike a more extensive sort of pluralism about reference might bewarranted—one that takes reference to be relative to be both acontext and a question we are interested in investigating, rather thanjust relative to a context. Of course, even a more thoroughgoingpluralism of this sort won’t do anything to address thechallenge raised by Unger and Geach.

At this point, it is natural to wonder whether the same sorts ofconsiderations which have motivated the exploration of nihilism,particularism, and pluralism about linguistic reference will carryover to other kinds of reference as well. It is fairly easy to imaginehow the issues raised by Unger and Geach might well arise with respectto pictorial or perceptual reference, but harder to see howHeck’s challenge is going to apply in the case of the latter. Itmight thus turn out that there are reasons to be a pluralist about,say, linguistic reference but not about perceptual reference. Gettingclearer on this, however, is likely to require making headway on thequestion of what (if anything) unifies e.g. linguistic, perceptual,and pictorial reference, what (if anything) makes them species of asingle genus. So far at least, that question remains largelyunexplored.

Bibliography

• Abbott, B., 2008. ‘Issues in the Semantics and Pragmatics ofDefinite Descriptions in English,’ in J. Gundel & N. Hedberg(eds.),Reference: Interdisciplinary Perspectives, Oxford:Oxford University Press.
• Bach, K., 1992. ‘Intentions and Demonstrations,’Analysis, 52(3): 140–146.
• –––, 2017. ‘Reference, Intention, andContext: Do Demonstratives Really Refer?,’ in M. de Ponte and K.Korta (eds.),Reference and Representation in Thought andLanguage, Oxford: Oxford University Press.
• Burge, T., 1973. ‘Reference and Proper Names,’Journal of Philosophy, 70(14): 425–439.
• Chastain, C., 1975. ‘Reference and Context,’, in K.Gunderson (ed.),Language, Mind and Knowledge (MinnesotaStudies in the Philosophy of Science: Volume VII), Minnesota:University of Minnesota Press.
• Cohen, J. and E. Michaelson, 2013. ‘Indexicality and theAnswering Machine Paradox,’Philosophy Compass, 8(6):580–592.
• Devitt, M., 1997. ‘Meaning and Psychology: A Response toRichard,’Noûs, 31(1): 115–131.
• –––, 2022. ‘The Irrelevance of Intentionsto Refer: Demonstratives and Demonstrations,’PhilosophicalStudies, 179(3): 995–1004.
• Dickie, I., 2015.Fixing Reference, Oxford: OxfordUniversity Press.
• Donnellan , K., 1966. ‘Reference and DefiniteDescriptions,’Philosophical Review, 75(3):281–304.
• –––, 1968. ‘Putting Humpty Dumpty TogetherAgain,’Philosophical Review, 77(2):203–215.
• –––, 1970. ‘Proper Names and IdentifyingDescriptions,’Synthese, 21(3–4):335–358.
• Evans, G., 1973. ‘The Causal Theory of Names,’Proceedings of the Aristotelian Society, Supplementary Volume47: 187–208.
• –––, 1982.The Varieties of Reference,Oxford: Oxford University Press.
• Fara, D. Graff, 2015. ‘Names are Predicates,’Philosophical Review, 124(1): 59–127.
• von Fintel, K., 2004. ‘Would you Believe It? The King ofFrance is Back! (Presuppositions and Truth-Value Intuitions),’in A. Bezuidenhout and M. Reimer (eds.),Descriptions and Beyond:An Interdisciplinary Collection of Essays on Definite and IndefiniteDescriptions, Oxford: Oxford University Press.
• Fodor, J.D., and I. Sag, 1982. ‘Referential andQuantificational Indefinites,’Linguistics andPhilosophy, 5(3): 355–398.
• Frege, G., 1892. ‘On Sense and Reference,’ in P. Geachand M. Black (eds.)Translations from the Philosophical Writingsof Gottlob Frege, Oxford: Blackwell (1952).
• García-Carpintero, M., 1998. ‘Indexicals asToken-Reflexives,’Mind, 107(427): 529–564.
• Gauker, C., 2008, ‘Zero Tolerance for Pragmatics,’Synthese, 165(3): 359–371.
• Geach, P., 1969. ‘The Perils of Pauline,’Reviewof Metaphysics, 23(2): 287–300.
• –––, 1980.Reference and Generality,Ithaca: Cornell University Press.
• Gray, A., 2014. ‘Name-Bearing, Reference, andCircularity,’Philosophical Studies, 171(2):207–231.
• –––, 2015. ‘Lexical Individuation andPredicativism About Names,’Thought, 4(2):113–123.
• Gómez-Torrente, M., 2019.Roads to Reference,Oxford: Oxford University Press.
• Grice, P., 1957. ‘Meaning,’PhilosophicalReview, 66(3): 377–388.
• Heck, R.K., 2014, ‘Semantics and Context-Dependence: Towardsa Strawsonian Account,’in A. Burgess and B. Sherman (eds.),Metasemantics: New Essays on the Foundations of Meaning,Oxford: Oxford University Press (originally published under the name‘R.G. Heck Jr.’).
• Jeshion, R., 2015. ‘Names Not Predicates,’ in A.Bianchi (ed.),On Reference, Oxford: Oxford UniversityPress.
• Kaplan, D., 1978. ‘Dthat,’ in P. French, T. Uehling,and H. Wettstein (eds.),Contemporary Perspectives in thePhilosophy of Language, Minneapolis: University of MinnesotaPress.
• –––, 1989a. ‘Afterthoughts,’ in J.Almog, J. Perry, and H. Wettstein (eds.),Themes from Kaplan,Oxford: Oxford University Press.
• –––, 1989b. ‘Demonstratives: An Essay onthe Semantics, Logic, Metaphysics, and Epistemology of Demonstrativesand Other Indexicals,’ in J. Almog, J. Perry, and H. Wettstein(eds.),Themes from Kaplan, Oxford: Oxford UniversityPress.
• –––, 1990. ‘Words,’Proceedingsof the Aristotelian Society, 64: 93–119.
• Keiser, J., 2023.Non-Ideal Foundations of Language, NewYork: Routledge.
• King, J., 2013. ‘Supplementives, the Coordination Account,and Conflicting Intentions,’PhilosophicalPerspectives, 27(1): 288–311.
• –––, 2014. ‘Speaker Intentions inContext,’Noûs, 48(2): 219–237.
• Kripke, S., 1972.Naming and Necessity, Cambridge, MA:Harvard University Press.
• –––, 1977. ‘Speaker’s Reference andSemantic Reference,’Midwest Studies in Philosophy, 2:255–76.
• Lewis, D., 1993. ‘Many, but Almost One,’ in J. Bacon(ed.),Ontology, Causality and Mind: Essays in Honour of D MArmstrong, Cambridge: Cambridge University Press.
• Ludlow, P. and G. Segal, 2004. ‘On a Unitary SemanticalAnalysis for Definite and Indefinite Descriptions,’ in A.Bezuidenhout and M. Reimer (eds.),Descriptions and Beyond: AnInterdisciplinary Collection of Essays on Definite and IndefiniteDescriptions, Oxford: Oxford University Press.
• MacKay, A., 1968. ‘Mr. Donnellan and Humpty Dumpty onReferring,’Philosophical Review, 77(2):197–202.
• Marcus, R. Barcan, 1947. “The Identity of Individuals in aStrict Functional Calculus of Second Order,”Journal ofSymbolic Logic, 12(1): 12–15.
• –––, 1961. ‘Modalities and IntensionalLanguages,’Synthese, 13(4): 303–322.
• –––, 1993.Modalities, Oxford: OxfordUniversity Press.
• McGinn, C., 1981. ‘The Mechanism of Reference,’Synthese, 49(2): 157–186.
• Meinong, A., 1904. ‘The Theory of Objects,’ in Meinong(ed.)Untersuchungen zur Gegenstandstheorie und Psychologie,Barth: Leipzig.
• Mill, J. S., 1867.A System of Logic, London:Longmans.
• Michaelson, E., 2022. ‘Speaker’s Reference, SemanticReference, Sneaky Reference,’Mind and Language, 37(5):856–875.
• –––, 2023. ‘The Vagaries ofReference,’Ergo, 9(52): 1433–1448.
• –––, forthcoming. ‘UnspeakableNames,’Synthese, first online 13 February 2023.doi:10.1007/s11229-023-04038-0
• Michaelson, E. and E. Nowak, 2022. ‘On Salience-BasedTheories of Demonstratives,’ in S. Archer (ed.)Salience: APhilosophical Inquiry, New York: Routledge.
• Mount, A., 2008. ‘Intentions, Gestures, and Salience inOrdinary and Deferred Demonstrative Reference,’Mind andLanguage, 23(2): 145–164.
• Neale, S., 2004. ‘This, That, and The Other,’ in A.Bezuidenhout and M. Reimer (eds.),Descriptions and Beyond: AnInterdisciplinary Collection of Essays on Definite and IndefiniteDescriptions, Oxford: Oxford University Press.
• Nelson, M., 2002. ‘Descriptivism Defended,’Noûs, 36(3): 408-435.
• Nowak, E., 2020, ‘No Context, No Content, No Problem,’Mind and Language, 36(2): 189–220.
• Nowak, E. and E. Michaelson, 2020. ‘Discourse andMethod,’Linguistics and Philosophy, 43(2):119–138.
• –––, 2021. ‘Who’s Your IdealListener?’Australasian Journal of Philosophy, 99(2):257–270.
• –––, 2022. ‘Meta-Metasemantics, or theQuest for the One True Metasemantics,’PhilosophicalQuarterly, 72(1): 135–154.
• Perry, J., 2001.Reference and Reflexivity, Stanford:CSLI Publications.
• Pelczar, M. and Rainsbury, J., 1998. ‘The IndexicalCharacter of Names,’Synthese, 114(2):293–317.
• Predelli, S., 2002. ‘Intentions, Indexicals, andCommunication,’Analysis, 62(4): 310–316.
• Quine, W.V.O., 1960.Word and Object, Cambridge MA: MITPress.
• Radulescu, A., 2018. ‘The Difference Between Indexicals andDemonstratives,’Synthese, 195(7):3173–3196.
• –––, 2019. ‘A Defense of Intentionalismabout Demonstratives,’Australasian Journal ofPhilosophy, 97(4): 775–791.
• Rami, D., 2014. ‘The Use-Conditional Conception of ProperNames,’Philosophical Studies, 168(1):119–150.
• Recanati, F., 1997.Direct Reference: From Language toThought, Oxford: Blackwell.
• Reichenbach, H., 1947.Elements of Symbolic Logic, NewYork: Dover.
• Reimer, M., 1991. ‘Demonstratives, Demonstrations, andDemonstrata,’Philosophical Studies, 63(2):187–202.
• –––, 1992. ‘Three Views of DemonstrativeReference,’Synthese, 93(3): 373–402.
• Russell, B., 1905. ‘On Denoting,’Mind,14(56): 479–93.
• –––, 1911. ‘Knowledge by Acquaintance andKnowledge by Description,’Proceedings of the AristotelianSociety, 11: 108–128.
• Schiffer, S., 1981. ‘Indexicals and the Theory ofReference,’Synthese, 49: 43–100.
• Schoubye, A., 2011.On Describing, Ph.D. Thesis,Department of Philosophy, St. Andrews University.
• Searle, J., 1983.Intentionality, Cambridge: CambridgeUniversity Press.
• Soames, S., 2002.Beyond Rigidity, Oxford: OxfordUniversity Press.
• Speaks, J., 2016. ‘The Role of Speaker and Hearer in theCharacter of Demonstratives,’Mind, 125(498):301–229.
• –––, 2017. ‘A Puzzle About Demonstrativesand Semantic Competence,’Philosophical Studies,174(3): 709–734.
• Stebbing, S., 1943.A Modern Elementary Logic, London:Methuen.
• Stine, G., 1978. ‘Meaning Other Than What We Say andReferring,’Philosophical Studies, 33(4):319–337.
• Stójnic, U., Stone, M., and E. Lepore, 2013. ‘Deixis(Even Without Pointing),’Philosophical Perspectives,27(1): 502–525.
• –––, 2017. ‘Discourse and Logical Form:Pronouns, Attention, and Coherence,’Linguistics andPhilosophy, 40(5): 519–547.
• Stokke, A., 2010. ‘Intention-Sensitive Semantics,’Synthese, 175(3): 383–404.
• Strawson, P., 1950. ‘On Referring,’Mind,59(235): 320–44.
• –––, 1959.Individuals: An Essay inDescriptive Metaphysics, London: Methuen.
• Unger, P., 1980. ‘The Problem of the Many,’Midwest Studies in Philosophy, 5(1): 411–468.
• Unnsteinsson, E., 2022.Talking About: An IntensionalistTheory of Reference, Oxford: Oxford University Press.
• Wittgenstein, L., 1958.The Blue and the Brown Books,Oxford: Basil Blackwell.
• Wettstein, H., 1984. ‘How to Bridge the Gap between Meaningand Reference,’Synthese, 58(1): 63–84.

Academic Tools

	How to cite this entry.
	Preview the PDF version of this entry at theFriends of the SEP Society.
	Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO).
	Enhanced bibliography for this entryatPhilPapers, with links to its database.

Other Internet Resources

[Please contact the author with suggestions.]

Related Entries

descriptions |externalism about the mind |Goodman, Nelson: aesthetics |indexicals |many, problem of |mental content: causal theories of |mental content: teleological theories of |mental representation |names |natural kinds |nonexistent objects |rigid designators |skepticism: and content externalism

Acknowledgments

Editor’s note: In 2014, Eliot Michaelson took overresponsibility for updating the entry originally written by MargaReimer. As of the 2024 update, Eliot Michaelson is listed as soleauthor because no substantive content remains from Reimer’sversion.