University of St Andrews
In this paper I am going to be arguing that the account of identity through change of composition offered by E.J. Lowe is inadequate due to a misconception regarding what it means for one thing to be the same as the original. I wish to argue for a more common-sense view of the matter, as supported by Hobbes, where identity is dependent on physical composition and two things cannot be identical to one. The discussion will mainly be concerning the puzzle of the ship of Theseus as that is what Lowe’s main argument is based around.
This area of metaphysics is concerned with what it means for two things to be identical, but in what sense? According to Leibniz Law, whatever is identical with one thing is that very thing itself, and so whatever is true of one thing is also true of whatever is identical to that thing. More simply put: if a has all the same properties as b, a and b are identical. I agree with Lowe here, that this realisation is trivial and lacks any metaphysical significance; it is clearly just common sense. However, I feel it is important to mention this as a starting place for our conception of identity, even though it is not particularly profound. From here, Lowe turns to the problem of change, where philosophers usually distinguish between what is numerically identical and what qualitatively identical. Lowe’s focus in this section of text, though, is on change of composition. He addresses the question of how much something can change (if at all) before it is no longer the thing it once was.
The example, on which Lowe bases his views about this problem, is the puzzle of the ship of Theseus, a legend reported by the Greek historian, Plutarch. The story goes that the ship was left in the harbour at Athens after Theseus died and it was preserved for many years. After some time, the parts began to decay and they were replaced one-by-one by new parts. Eventually, nothing of the original ship was left and the question, of whether this renovated ship is the same one that Theseus sailed in, remains. Lowe’s immediate response to this is that ‘an artefact can undergo replacement of its parts […] if the replacement occurs in a gradual and piecemeal fashion’ (Lowe, 2012, p.25) and so the renovated ship is still identical with the original ship. It is clear that this change in the ship has occurred over a period of time, but how can a ship that contains none of the original parts actually be that ship as it was when it was made? Lowe returns to this point, but first explores what it would mean for us to place a limit on how much of something can be changed before there is a loss of identity. He claims that it would make no sense to argue that the ship could not undergo a change to any of its parts and still be identical with the original. Some philosophers are in conflict with this, though: notably Chisholm, Leibniz and Moore, who believe in what is known as mereological essentialism. This argues that ‘for any whole x, if x has y as one of its parts, then y is a part of x in every possible world in which x exists’ (Chisholm, cited: Plantinga, 1975, p.468). The implication of this, assuming it is true, is that nothing can survive any change and keep its identity. In some ways this is an easier position to adopt as it saves one having to decide how much change is allowed but, of course, since when have philosophers been concerned with what is easiest?
Lowe’s response to the first part of the puzzle seems to create a further conflict. He argues that the renovated ship is in fact identical with the original even though none of its parts are. I wish to partly favour the mereological essentialists on this matter, both because there seem to be no connection between the original ship and the renovated ship, and because of the next puzzle in the story and Lowe’s beliefs about it, which are as follows.
This final part of the legend was first considered by Hobbes, but it has been retold in several ways since (see Aune, 1985, pp.82-3). Hobbes asks us to imagine that the original parts of Theseus’ ship were kept in a warehouse and one day reassembled in the same way they were before. There would then be two ships which looked almost identical, but there is a worry concerning which is the original. Lowe argues that both have a claim to being the original ship, though accepts that it is not possible for both to be. He continues to maintain that the renovated ship is identical with the original ship and shows that it is not possible for both ships to be identical with original because they ‘are two quite distinct ships, each having a quite distinct location’ (Lowe, 2012, p.28). This leaves him with a problem and, to avoid being left with neither ship being identical (as the mereological essentialists would have to argue), Lowe suggests two solutions to the problem, which he labels radical and later intolerable.
The first solution is to bring back the conception of the two ships both being identical with the original and accept that the same ship can be in two different places at once – the harbour and the warehouse. Lowe quickly rejects this on the grounds that it goes completely against common sense, which is interesting because I would argue the same of his later concepts, and make it impossible for us to know how many ships with one identity are present in any one place. This is a bit of an impossible situation and allows us to make no progress. The second solution offered is little better and falls prey to the same criticisms. The idea is that both ships were originally in the harbour as two quite distinct ships and became separated by a process of renovation and removal. As well as the above issues, this view forces us to conceive no longer of the ship of Theseus as one entity, but the whole point of the discussion hangs on the fact that is.
We come now to the main argument Lowe’s offers as a solution to the puzzle. Interestingly, he claims that his opinion concerning the problem at hand is a common-sense conception but, as we shall see, the implications of such a proposition render it rather a misconception. His first claim is that, at the middle stage, the ship parts in the harbour belong to the ship in the harbour and those in the warehouse ‘belong to no ship at all at that time’ (Lowe, 2012, p.31). He continues by upholding his original view that the renovated ship is the ship of Theseus and it is a mistake to say that the reconstructed ship refers to one ship in the two different situations; one where renovation does occur and one where it does not.
The important part of this argument is where the greater issue comes in. Lowe made a claim and then later contradicted himself concerning whether identity can depend on the existence of another entity. He insists a few pages before his main line of argument that the identity of two things can only concern those things, independently of what else exists. However, his proposed solution seems to indicate that he does in fact believe this, arguing that the reconstructed ship would have been identical with the original if renovation had not occurred and attempting to make it sound more plausible by suggesting that the reconstructed ship could not have existed if renovation had not happened, in the renovation situation. I do not doubt that there would need to be parts in the warehouse for reconstruction to happen, but I find the notion of a renovated ship having a greater claim to being the original than a reconstructed one, highly contentious not to mention unsettling. Even considering all else Lowe says about his view of the problem, he still takes it for granted that the identity of the ship of Theseus has neither been lost nor moved.
Before coming to what Hobbes argues about this, I want to show the implications for such an argument to indicate what we are allowing by accepting what Lowe says as true. From there, I will present a more plausible explanation of the issue at hand and show that common sense is not as far out as we might at first think. Although arguably a fairly silly illustration, the following situation shows what Lowe has forced himself to believe. In the TV show Only Fools and Horses, there is a character called Trigger who earns an award for saving money by having the same broom for twenty years. Upon investigation, he reveals that said broom has in fact had seventeen new heads and fourteen new handles in that time and, unsurprisingly, one of the men listening demands to know how it can still be the same broom. Trigger responds by showing them a photograph of himself and his broom as proof, but this fails, and I think quite rightly, to convince his listeners. It is clear that, for Trigger, replacing one part of the broom at a time does not do anything to its identity as the ‘original broom’, but surely with both a new handle and a new head, the resulting broom is a completely new one; just because Trigger thinks it is the same broom does not mean that it is.
We can make this dilemma equal to the one of the ship of Theseus by supposing that the original handle and head of the broom are kept and later glued back together, leaving us with two brooms. Which then has the identity of the original broom? I would question whether there might be people who follow Lowe on the ship, but disagree with Trigger on the broom. My worry is that, in the former case, the time period and number of pieces tricks people’s perception. With the ship, there are numerous pieces that are changed over a very long time, whereas the broom has just two and these changes have taken place over twenty years. It seems that this difference allows one argument to be sensible and the other quite the contrary. The reality is that they are essentially the same situation and what applies to one, must apply to the other.
This leaves us having to decide how else to solve the problem. Thomas Hobbes puts forward a different and seemly more plausible answer to the puzzle:
Lastly, if the name be given for some accident, then the identity of the thing will depend on the matter; the accidents that were, are destroyed, and the other new ones are generated, which cannot be the same numerically; so that a ship, which signifies matter so figured, will be the same as long as the matter remains the same; but if no part of the matter be the same, then it is numerically another ship; and if part of the matter remain and part be changed, then the ship will be partly the same, and partly not the same. (p.138)
Hobbes’s argument is quite simply that identity depends on original matter and something with none of the same matter is numerically a different thing. He even goes on to add that at the in-between stage, the ship is partly the same and partly not the same. This might leave us with the question of where the original ship is once renovation has happened as it cannot be in the harbour. Presumably, the ship is now where its parts are – in the warehouse – and so in the non-reconstruction situation, the original ship of Theseus simply ceases to exist as a ship. If the parts are just thrown into the warehouse and never used for anything, the ship just exists in parts, if they are actually burnt, then there is clearly nothing left of the ship. I do not want to argue for intermittent existence however, as it seems to make no sense to say that something can be between existing and not existing. All I, and I believe Hobbes, are trying to say is that the ship is just in another form; the entity itself still exists, but it is no longer ship-like in appearance. Provided the parts are restorable, its appearance can be made to resemble a ship once again. Therefore, according to Hobbes, the ship now in the harbour is very similar to Theseus’ ship, but it is not identical with it.
To show support for this argument, I feel that it is necessary to explain what the proposed conception of identity is based upon. No-one would argue that identity is not about things being the same, but there might be disagreement about what is the same. Hobbes, as well as the mereological essentialists, hold the view that identity is based on constituent parts and a change of parts is a change or loss of identity. To them, an entity’s parts are its essence and therefore a necessary part of its identity. This seems to be common sense, for if identity is not based on constituent parts, what else can it based on? Surely arguing that something with completely new parts is identical to the original leaves the criteria for identity completely arbitrary. I feel that this is nonsense and that identity is not open to allowing complete change.
As mentioned previously, I do not wish to argue for mereological essentialism, but for Hobbes’ slightly weaker conception of identity. The main issue with mereological essentialism is that it would deny that a reconstructed ship or a ship left in the harbour without renovation is identical with the original. This is because this view allows no change to occur; in the first case, the ship would have to be put back together with glue or nails and other new material which would change its composition, and in the second case, the decaying would also change it, albeit naturally. For the mereological essentialists, it is very difficult for anything to keep its identity; it would keep having a new one.
As a result of this, it seems that I have only given a definition of identity which works for objects and this could be a potential problem for the discussion that would follow this; that of personal identity. This particular discussion really concerns artefacts, as in the puzzle of the ship of Theseus, and so I will not go into too much detail about this further issue. However, I feel that it should be noted that the current argument would lead us either to accepting that we do change identity when all our cells are renewed (as is said to happen every seven years) or, if we refuse to do that, we would need to conceive of the mind as not entirely material and argue for a dualist position. Unfortunately, I do not have time to go into this, but it would be another interesting discussion.
In conclusion, then, Lowe’s conception of identity and the ideas he puts forward to support it, fail to account for the way we understand identity and leave us with implications that are implausible. Lowe argues for the renovated ship being identical with the original, but this leaves us with rather arbitrary criteria on which we place our understanding of identity. Because of this, I have proposed a view first suggested by Thomas Hobbes that makes constituent parts an essential aspect of the identity of something. It seems that the mereological essentialists go too far with an allowance of zero percent change and leave us with a never-ending cycle of renewed identity. On the other hand, I would argue that Lowe goes too far the other way and so his view is equally problematic. By appealing to the common sense view I have suggested, we avoid any issues of arbitrary criteria, two things having the same identity and something with more identical parts having less identity.
Aune, B. (1985), Metaphysics: The Elements, United States: University of Minnesota Press.
Lowe, E. (2002), A Survey of Metaphysics, Oxford: Oxford University Press.
Hobbes, T. (1839; 1655), The English Work of Thomas Hobbes, Volume I, London: John Bohn.
Plantinga, A. (1975), On Mereological Essentialism, The Review of Metaphysics, 28 (3).