Saul Kripke (1970) revolutionised the philosophy of language with his ideas on proper names being ‘rigid designators’ as they overcame the problems of descriptivism. Descriptivism argued that the meaning of a proper name shares identity with how speakers describe it. This theory was strongly pushed by the likes of Russell (1905) and Frege (1980), however Kripke manages to dismantle these ideas in his 1970 lecture series, paving the way for his own thesis. His theory has strong counterarguments, especially regarding the mechanism that assigns rigid designation in his thesis (the causal account). Due to these counterarguments, asserting his thesis is successful appears foolish. However, Kripke himself numerous times refers to his ideas on rigid designation as a ‘picture’ (Kripke, 1970). This seems to imply that he himself did not believe his ideas were completely successful, but an initial step towards abandoning descriptivism with all its faults and searching for a different theory on proper names that could overcome the flaws of descriptivism. Lycan (2008) asserts that Kripke’s ideas are extremely successful at influencing the philosophical dialogue, as they carve out a gap for more ‘fleshed out’ theories (Lycan, 2008). Kripke also developed an original account of ‘a priori’ knowledge that shook the very foundations of epistemology and metaphysics. In this essay, I will argue that Kripke’s ideas as a thesis are not convincing, however they are successful in pushing the philosophy of language forward, more precisely the philosophical dialogue regarding proper names, and this I believe was his intention. I will start by outlining Kripke’s ideas about how proper names are ‘rigid designators’ alongside his dismantlement of descriptivism. This will be followed by some well-known counterarguments to his thesis and an issue I find regarding his idea of ‘transworld identity’ (Kripke, 1970) and eventually some concluding thoughts.
To defend his notion of proper names being rigid designators, Kripke argues for the invalidity of descriptivism by showing that descriptions are neither necessary nor sufficient for meaningful reference (ibid.). Descriptivists, such as Russell, claimed that proper names are definite descriptions in disguise and thus descriptions are sufficient and necessary for proper names (Russell, 1905). Using counterexamples, Kripke shows that there is a logical difference between names and descriptions. This disproves descriptivism as a theory in my mind. One of his counterarguments relates to hearing a name without knowing anything about its reference, yet still possessing the ability to correctly refer using that same name in a later conversation. Say someone who knows nothing about football hears a group of football fans discussing ‘Cristiano Ronaldo’ in a pub. Kripke would then argue that this person is able to successfully refer to Cristiano Ronaldo without having access to the descriptions we have about this man; a Portuguese footballer whom is arguably the best in the world. What these type of counterarguments show is that descriptions are not necessary for reference; we can still correctly refer to Cristiano Ronaldo without knowing anything about him.
He then uses another type of example to prove that descriptions are not even sufficient for reference. This type of example uses the notion of possible worlds within logic to show this. By postulating hypothetical examples, he claims we can change all possible descriptions and yet the reference remains intact. We may say that Sylvester Stallone is a Hollywood actor who plays the main character in the Rocky movie saga. However, we can clearly conceive of a possible world in which he did not get this part, if -say- the director did not like his audition Therefore this description of him is clearly not enough, as in this possible world the name ‘Sylvester Stallone’ is still attached to man it refers to, but this man does not fit his description any more. This thought experiment could be repeated enough times to completely strip any descriptions we may consider unique to a reference, thus showing that the name ‘Sylvester Stallone’ can’t be synonymous with any description we may consider unique to Sylvester Stallone, yet this name will successfully refer to Sylvester Stallone in all hypothetical cases- that is in all possible worlds. For this reason, names and descriptions are clearly different on a logical platform and thus, description is insufficient for reference. These examples evidently expose some key flaws of descriptivism, which is enough to refute it as a valid theory.
After providing this dismantlement of descriptivism, he proposes his alternative view that proper names are ‘rigid designators’, whereas descriptions are ‘non-rigid’ (Kripke, 1970). This idea is founded on the basis that proper names will always have the same referent in all possible worlds. What is necessary in order to refer is to be able to simply point at an object whilst calling it by its proper name (ibidem); we don’t need to actually know anything about the object. So how does a name receive its referent? Kripke explains there is a ‘baptismal event’ where a referent receives a name. This event can occur in a myriad of different ways, a perfect example is when parents name their child. This name is then preserved through what Kripke calls the causal chain, which means nothing more than through speaker to speaker. So to successfully refer, we must have a causal link to baptismal event of the referent’s name. This can be simply overhearing a name like ‘Cristiano Ronaldo’- that is sufficient for this causal chain to be preserved from when this man was given this name to the moment the person who knows nothing about him utters his name.
This may all seem quite convincing, as it overcomes the previously discussed issues that descriptivism faces. However, it faces some issues that are difficult to overlook. A major one is raised by Gareth Evans (1973), and it attacks the causal chain Kripke’s thesis highly depends one. In order to do this, he uses an example of the following form: two children are born, one is names ‘Tim’ and the other ‘Tom’. Due to a hospital error, these two children are accidentally changed in their cots, and so the parents receive a different child. Both set of parents refer to their new baby by the name of their old baby. Evans here argues that it is ludicrous to assume that both babies still conserve their original referents, however Kripke’s causal chain doesn’t seem to allow for a change of referent or second baptismal event (Evans, 1973).
Another issue is dubbed the ‘qua-problem’ (Sterelny & Devitt, 1987) and it concerns cases where people have certain beliefs about any given referent that are, ultimately, incorrect. In these cases, it seems some sort of description is necessary, as these erroneous uses of a name clearly don’t properly refer. There are many different types of examples of this; an easy one to grasp is the use of the name ‘Bill Bryson’ in Durham. This is the name of the university library (obviously named after a person, however this is irrelevant in this example). If someone overhears some Durham students talking about the ‘Bill Bryson’ (or as it is more commonly known: the ‘Billy B’), they may assume they are referring to a person. If they then later on use this name under this assumption, this is surely a case of erroneous referral. This seems to imply we do actually need a bit of context, or description, to make proper references. In fact, Kripke himself spots a similar issue with his own work Santa Claus example (Kripke, 1970). This dwells on the fact that it is unreasonable to claim children who say ‘Santa Claus’ are referring to saint this name originally stems from- they are clearly referring to a plump, festive bearded man who climbs down a chimney every Christmas to bring them presents.
Kripke’s constant use of the concept of ‘transworld identity’ is problematic in my eyes. He uses it to claim a name will pick out the same referent regardless of it having varied descriptions in different possible worlds (ibid.). This seems to imply that the referent can completely vary in different possible worlds; for example, maybe ‘Sylvester Stallone’ refers to something we would describe as a successful actor in one possible world, and in another refers to something we would describe as a sock. This sounds ridiculous, however the possible worlds thought experiment allows us to postulate this. According to Kripke, the successful actor and the sock are both being correctly referred to; since names are rigid designators the name is picking out the same reference in both these worlds. This seems to imply they share some sort of identity, which is clearly not based on any sort of description, as the successful actor is as different from a sock as possible. Kripke would describe this identity as a vague concept and of no real concern, as the possible worlds thought process does not entail we can consider overlaps between worlds and so we shouldn’t compare the descriptively different referents. He describes the thought experiment as a process that can be likened to looking at another reality through a telescope (ibid.). However, it seems too convenient for his thesis that this identity is vague. On the one hand he uses it to show descriptions are insufficient and unnecessary for referring. Yet, on the other hand, he implies there is an identity shared between referents of different realities that share the same name. It seems that by saying the concept of identity unclear, he is trying to justify the fact that he has no way of explaining it. Taking descriptions completely away from the definition of identity creates a whole new philosophical debate that Kripke does not account for.
These contradictions are too strong for Kripke to account for. However, as implied in the introduction, Kripke refers to his ideas as a ‘picture’ (ibid.), not a thesis. He employs words like ‘often’ and ‘usually’ when talking about proper names. This implies his awareness that his ‘picture’ is not strong enough to account for all cases of possible names and thus not a fully formed thesis. This, however, should not serve to disregard his ideas, as they form a platform for the philosophy of language to develop a new string of ideas and they also heavily enrich the philosophical dialogue regarding proper names. This was Kripke’s intention all along, to produce a ‘picture’ that, though flawed, could be picked up by other philosophers to develop their own ideas. This was the case for Michael Devitt (1981) who produced a more intricate form of Kripke’s picture to overcome Evans’ counterargument. Kripke’s ideas also allowed for the creation of moderate theories that link certain ideas from descriptivism and direct reference to create new, hybrid ideas. Furthermore, he developed original ideas that showed previously unimaginable cases of contingent ‘a priori’ and necessary ‘a posteriori’. He shows the former using Wittgenstein’s example of a meter ruler (Wittgenstein, 1953). We know ‘a priori’ that a meter ruler measures a meter in length, however changes in temperature may slightly vary the length of the ruler at a given time. So saying ‘a metre ruler measures a metre in length’ is therefore contingent. His cases of the latter depend on his ideas of rigid designation, but if this notion is accepted they are equally compelling, original cases of necessary ‘a posteriori’.
Kripke’s claim that proper names are rigid designators clearly falls short of being a fully-fledged theory, however it represents an important progress in the philosophy of language as it enforces an abandonment of the previously popular descriptivism. Kripke intended to provide a ‘picture’, or a platform for the philosophy of language to build on and it was extremely influential; it allowed for other philosophers to explore other avenues: Devitt refined his ideas of proper names being ‘rigid designators’ (Devitt, 1981) and Evans pursued a hybrid theory that joined both descriptivism and rigid designation (Evans, 1973). So Kripke’s ideas, while not convincing as a theory, represented an important progress to the debate on philosophy of language.