By Richard Joyce
University of Sheffield / RSSS, Australian National University
1. Cultural Treasures
This paper addresses some of the moral issues surrounding the question of returning cultural treasures to their places of origin. It does not pretend to be a complete discussion, and has two points of focus. First, the main body of the paper is concerned with disarming an argument commonly deployed against the proposal that such items ought to be returned: the slippery slope argument. The suspicions that are raised about such arguments are intended to be quite general, and applicable to other moral debates. Second, this paper has an interest in utilizing these general results for a particular casethat of the reliefs and statuary from the Parthenon (the so-called "Elgin Marbles"), housed in the British Museum since the early nineteenth century. This example has been chosen not merely because it is the cause célèbre of such cases, but because it lacks attributes that might otherwise cloud discussion (e.g., the marbles were not, in any obvious way, stolen), and because in nearly all formal debates on the matter over the years (e.g., in the House of Lords), "fear of setting a precedent" has been raised as the major consideration against return. Besides, it is pleasing to take philosophical discussion and apply it directly not merely to a type of moral case, but to a particular problem.
These issues are not discussed here from a legal point of view, but from a moral point of view. Nor (perhaps surprisingly) is this paper concerned with the issue of ownership, since ownership is primarily a legal category. It seems, however, that in the popular rhetoric over cultural treasures there is often employed a vague notion of "moral ownership": that, irrespective of the legalities, certain countries or groups enjoy "rights of possession" over certain objects, in virtue, presumably, of a lineage of transferal of rights. Perhaps a solid moral argument might be forged out of such ideas, but it is not attempted here. It is for this reason that this paper uses the term "cultural treasures", and eschews talk of "cultural property", the term favored by much of the debate on this issue. First, it will be established how one might create a moral presumption in favor of the return of a cultural treasure without recourse to an appeal to property rights. But the major part of the paper is devoted to the task of combating a common counter-argument that is raised even once such a presumption is grantednamely, that returning any such item will "open the flood-gates" to a disastrous result, and thus, irrespective of any prima facie case, no return should be countenanced.
In order to create the moral presumption, it is useful to think by analogy of items of sentimental value to individuals. Suppose that Jack buys a pocket watch from a used goods store. He buys it fair and square, under perfectly legitimate circumstances. Now suppose that he subsequently discovers that this very watch is of great sentimental value to some family. Nobody stole it from themlet's say that an irresponsible family member pawned the watch without the knowledge of the others. Finally, let's stipulate that the watch really is of immense significance to the familyperhaps it's the last tangible reminder of a deceased family memberand its absence is, quite reasonably, the object of significant emotional distress. What are the relevant moral features of such a situation?
The pocket watch is Jack's, not the family's, and they have no right to claim it. Legally, they don't have a leg to stand on, and it would seem that morally Jack is under no obligation to make them happy by returning it. Nevertheless, anyone who thought that there ends the moral issue would be misguided. Clearly, there are any number of terms of moral censure that could be legitimately applied to Jack if he keeps the watch, knowing of the suffering this causes: he is mean, hard-hearted, and unsympathetic. If he returns the watch to the family, on the other hand, he is kind, benevolent, and generous. One can be acting perfectly "within one's rights" while still being subject to reasonable moral criticism; it may be morally desirable for the watch to be returned without its being morally obligatory.
One obvious thing to note about objects of sentimental value (indeed, it's virtually what we mean by the term) is that they are irreplaceable: if someone loses her ordinary old watch, then she'll be just as happy with another; but if she loses her much-loved great grandfather's watch, then she'll be very upset, and no other watch will take its place. Another thing about such objects is that the value we gain from them comes largely from having them with us. It is quite nice for a person to know that her great grandfather's watch still existsbut it doesn't bring the personal value of having it in her possession. Exactly why we gain such value from items of sentimental worth is a question for psychologists. It is enough here to note that we do, and that although there is doubtlessly a vice of being overly-sentimental, or infusing trivial items with sentimental importance, there is nothing obviously unreasonable about this kind of valuing, and arguably much to be said in its favor.
Though the analogy is not perfect, it seems that there is a fruitful comparison to be made between things of personal sentimental value and the items we are calling "cultural treasures". Some objects are a focus of pride for a group or a nationfrom them comes a sense of heritage and historical rootedness, which contributes to cultural identity and self-esteem. The Statue of Liberty, the Eiffel Tower, the Pyramids, the Great Wall of China, and so on, are the most obvious examples that come to mind (to say nothing of the biggest). They are objects "charged" with cultural significance. When such things are irretrievably lost, it is rightly perceived as a tragedy. Replacement with a convincing replica just won't suffice; like a great grandfather's watch, their value comes from being that very object. A cultural treasure is not merely an artifact that has historical, archaeological, ethnographic or artistic significance, however great that significance may be; it is an item that plays a functional role for a certain group of persons. It acts rather like a figurehead (though "figurehead" sounds singular, and a culture may have many treasures).
Some will object that the notion of "cultural treasure" is so hopelessly vague as to be useless. It certainly is imprecise, but it doesn't follow that it is useless. Regardless of vagueness, there are going to be paradigms that are clear instances of cultural treasures. Some obvious and grand ones were just mentioned, but treasures can be considerably less impressive, for it is the role that they play that counts, and within a certain group that function might be filled by items of clothing, pieces of pottery, masks, etc. Additionally, there will be clear foils for the concept: a banana preserved since Tudor times in the mud of the Thames, a pair of trousers owned by Woodrow Wilson. These, though in their own way fascinating, are not cultural treasures. It is no challenge to find something for which one just isn't surethe wreck of the Mary Rose, perhaps?but the fact that one could not decide on the Mary Rose, and a thousand other instances, in no way undermines the truth that The Statue of Liberty is (and the banana is not) a cultural treasure.
The argument of this paper, then, is confined to the things that count as cultural treasures. On any reasonable specification, it is worth noting, the Elgin Marbles will count: the Parthenon is Greece's primary national monument, and the marbles are an integral part of that structure. It would not be too strong to say that if the marbles do not count as a Greek cultural treasure, then they have no cultural treasures at all. "Granted," one might object, "that there will be items that are clear cases of cultural treasuresbut how are we going to decide on the myriad of uncertain cases?" The best answer is to point out that we do make decisions concerning the cultural importance of things all the time. We grant things the official status of "historic landmark", of "national heritage site", of "area of outstanding beauty", and so on. Sometimes these categories involve rankings, or subcategories (e.g., "area of beauty" vs. "area of outstanding beauty"). Moreover, they often involve a special treatment, such as preservation by law, and therefore a decision is made (if only by default) on every instance within the relevant domain. All such conferrals involve making decisions in a gray area. Sometimes it's not easy, sometimes it involves negotiation and argumentbut that's a far cry from saying that such categories are so vague as to be impracticable and useless.
Although the preceding is brief, it satisfies the desire to create a presumption in favor of returning things that clearly count as cultural treasuressuch as the Elgin Marblesto the people who treasure them. The argument has nothing to do with ownership (except in so far as it might bear on the question of who should own the items), and doesn't pretend to deduce any moral obligations to return. Rather, we start with the assumption that the return has prima facie desirability, and from here examine whether there are obstructions to prevent this desirability becoming all-things-considered. It will be argued that one perceived obstacle (the slippery slope)so often appealed tocarries no weight.
2. Slippery Slopes
Slippery slope arguments are deployed only when there is some admission that the action at the top of the slope, as it were, has some prima facie desirability. They are employed against actions that we are tempted to perform, that seem acceptable on the face of things. It has often been noted that there is not obviously a single slippery slope argument, but various arguments hiding, sometimes entwined, under that vague heading; therefore it is difficult to give an initial generic statement of the objection. Nevertheless, its expression is something we are familiar with: Sir Alexander Cadogan (Permanent Undersecretary of State of Britain), contemplating the return of the Elgin Marbles in 1941, remarked "I don't know where this is all going to end", while another official in the foreign office had spoken previously of its setting "a dangerous precedent". The vague idea is that returning the marbles will lead to something obviously unacceptablesay, museums throughout the land stripped of everything but local artifacts.
In what follows it will be convenient if we formalize this. Treating A as Φ may seem acceptable, or even desirable, but it will lead inevitably to treating B as Φ, which is clearly unacceptable; therefore treating A as Φ is, all-things-considered, not acceptable or desirable after all. The virtue of treating the argument in such terms is that it will be easier to see how it transfers to other debates. In the euthanasia debates, for example, it is often argued that although it might be seem permissible to refrain from preventing a terminally ill patient from committing suicide in certain extreme circumstances ("It seems acceptable to treat A as Φ"), this would inevitably lead to less desirable actions, such as actively killing a terminally ill patient after receiving permission from his family ("It is unacceptable to treat B as Φ"), and this is used as an argument against the first proposed action ("So it is not all-things-considered acceptable to treat A as Φ"). And the slippery slope is well-known in the abortion debate as well: although it may be permissible, on the face of it, to terminate a pregnancy after one week, this just leads inexorably down through the months until one is accepting third trimester abortions (or even condoning infanticide); and since these latter actions are (the argument goes) clearly unacceptable, so too must be any abortion. However, despite the critique having general scope, it is beyond this paper (and would, besides, prove repetitive) explicitly to show the failings of the argument for these other debates.
This paper will offer four versions of the slippery slope argument presented in general terms, criticize each (often arguing by reductio), and also draw the moral for our test-case treasure, the Elgin Marbles. The argument as it appears in the abortion debate is a good example of the first kind of slippery slope to considerrevolving, it seems, around the question of "Where would one draw the line?" Hence we will call it (just for ease of reference) "the Continuum Argument".
2.1. The Continuum Argument:
You can't draw a principled line between things of type A and things of type B; there's just a continuum of cases between them, and any attempt to draw a border would be arbitrary. Therefore, if you're going to treat A as Φ, you'll have to treat B as Φ, too. Anything else would involve an arbitrary and subjective decision. But since treating B as Φ is clearly undesirable, treating A as Φdespite initial appearances to the contraryis also undesirable.
This argument is easily and justly parodied. Suppose a man wants to have his house painted yellow. He leaves the painters to the task, and returns later to find the house painted bright orange. Could the painters reasonably defend their choice by pointing out that yellow and orange merge imperceptibly into each other, challenging the man to draw a non-arbitrary line between the two? Of course not. It is simply absurd to claim that whenever the boundaries between two categories are imprecise, we are committed to treating instances of the two in the same way. Yellow and orange, despite fuzziness, are perfectly serviceable distinct concepts.
If forced, we often can draw boundaries between vague groups. As to the question of whether any such boundary is necessarily "arbitrary and subjective" the answer is "Not in any pernicious way". If one were to put the boundary between yellow and orange right up towards one end of the spectrum, then that is just wrong. It's as wrong as someone drawing a line between cats and dogs that includes small fluffy dogs in the cat category. The correct procedure is to get some sort of consensus on what counts as "the gray area" (or in this case "the yellowy-orange area") and draw a line roughly in the middle. There's a degree of arbitrariness involved. It's much the same kind of arbitrariness that faces authorities when they have to decide on speed limits on roadsdrawing a precise line between "too fast" and "not too fast". Perhaps putting the speed limit anywhere between 45 m.p.h. and 55 m.p.h. would be acceptable, in which case a precise speed of 50 m.p.h. is decided upon. Arbitrary? Somewhatin the sense that there was a range of acceptable outcomes to the decision (i.e., a speed limit of 52 m.p.h. would not have been "wrong")but, in so far as the law requires such precision, reasonably so.
This is not to say that we need to draw lines between vague conceptsmerely that sometimes circumstances might require us to make a decision one way or the other on every case. Consider a real-life case, already mentioned above: the conferral of the status of "National Heritage Site" on certain buildings. We start out with an indistinct notion that certain buildings are of cultural and historical value and others are clearly not. Nobody thinks that there is a natural crisp boundary between the two: the variables that determine this kind of value are many and nebulous. Then somebody comes along with the admirable idea that a certain statusone that involves legal protection and fundingshould be conferred on the valuable buildings. There are going to be a lot of buildings that immediately get that status, being obviously on the right side of the line. But then there's the gray area. A decision has to be made on every building in the gray areaeither it gets the funding or it does notand so certain boundaries will need to be stipulated. Here there won't be a single continuum, but the complex decision might involve deliberations over a variety of continua: being sixteenth-century will count for more than being seventeenth, being visited by a number of tourists greater than n per year will count. And other, equally vague matters will be considered: whether someone famous ever owned it, whether it is particularly beautiful, etc. The point is not to discuss how such decisions may be made, but simply to observe that we frequently do make yes-or-no decisions concerning utterly imprecise matters, and the fact that these decisions are complex and difficult in no way commits us to sliding from treating things up one end of the continuum in a certain way to treating similarly the things at the other end. We don't end up conferring "National Heritage" status on apartment blocks of the 1970s.
The application of these considerations to the case of the Elgin Marbles should be obvious. The argument claimed that it is desirable for the marbles to be returned on the grounds of their status as "a cultural treasure"a status that is to be distinguished from an item's being culturally significant, or an important historical document, or a work of art. It is has been admitted that this status is thoroughly imprecise, and if we decide to treat cultural treasures in a special way there will be cases lying within the gray area that will naturally be disputed. Perhaps there will be an item like a Japanese emperor's kimono regarding which we're simply unsure whether it has the status of "cultural treasure". But the fact that we may argue about the kimono in no way shows that we should refrain from treating the cases that are clearly cultural treasures in the desirable way, nor is it to say that the difficult cases could not be decided in due course. In any case, we would not be on a slope to returning to their original societies things that are clearly not cultural treasures.
2.2. The Principle Argument:
Although treating A as Φ is prima facie morally desirable, you need to consider the principle justifying this treatment (call this principle "P"). P also justifies treating B as Φ. But since treating B as Φ is clearly undesirable, treating A as Φdespite initial appearances to the contraryis all-things-considered undesirable.
It isn't obvious that this should really count as a "slippery slope" argument, but it's sufficiently in the ballpark to warrant examination. Up to a point the argument is perfectly acceptable, and resembles a form of reasoning that philosophers rely on frequently. Suppose that one were to think that killing animals for food is permissible, and that the principle underlying this appears to be something like "Animals have no moral rights because they are not sufficiently rational". But then we subject this principle to scrutiny by imagining what results it would yield if followed generally. We might be struck by the thought that young children and mentally retarded adults are also "insufficiently rational", and yet the thought of killing them for food is obviously unacceptable, and so we are forced to revise our initial position.
The problem is that "revising our initial position" is ambiguousthe initial position consisted of two things: a judgment that eating animals is permissible, plus a principle justifying that judgment. Showing that that principle would also license actions that are clearly unacceptable casts the principle into doubt; it doesn't undermine the judgment that eating animals is permissible. If the principle in question were the only possible way of justifying meat-eating, then the permissibility of the action would also be discredited, but this is unlikely to be the case.
If we are initially inclined to consider one action acceptable and another unacceptable, then (unless we are simply being irrational) there must be a difference between the two actions that explains this difference in our attitudes. What we must therefore do is try to articulate a principle that accommodates the first action's acceptability and the second action's unacceptability. Therefore if there's some principle P that tells us that we should treat them both as Φ, or neither of them as Φ, we have a reason for rejecting that principle and looking for another, P*, that satisfies the intuitive differentiation. And we know (again, unless our difference in attitude is just irrational bias) that there is a P*, though it may be hard to find and articulate. This is the most important thing to remember about slippery slopes. The very fact that one thing (at the top of the slope) appears acceptable, and another thing (at the bottom of the slope) appears unacceptable suffices to show that there is probably a difference between the two cases to which we are sensitive. Bringing this difference out into the open is what should prevent us sliding down the slope.
Concerning the Elgin Marbles, we might start out thinking that they ought to be returned because artworks should be located in the place of their origin. But then we think this through, and realize that the implementation of this principle would lead to a MOMA with no Picassos, a National Gallery with no Raphaels, etc. The unacceptability of this does not force a revision of our attitude towards the marbleswe merely come to see that we were trying to justify it with a foolish principle. Given that we do think that returning all Picassos to Europe would be absurd, whereas returning the marbles to Athens would be desirable, what we should be doing is searching for the factor that underlies this difference in attitude, and a decision-guiding principle that captures this factor. Progress is made when we realize that it is not their being wonderful artworks, or historical documents, etc., that underlies our attitude towards returning the marblesbut their being of a more distinctive category: "a cultural treasure". We then may exchange our initial reason for a better one: "It is desirable to return them because cultural treasures generally ought to be housed with the culture that treasures them." This principle no longer justifies the return of all those Picassos, Raphaels, etc., because although such artworks are obviously in one sense of the word "treasures", they do not play the symbolic role of being a focal point of national identity that the term is intended to indicate. (Perhaps there are some exceptions, such as Guernica.)
But imagine that we do come up with an explanation of our differentiating attitude, but it turns out to be an unacceptably unprincipled one. Perhaps we want to return the marbles but not, say, a tribal mask from New Guinea, simply because we "like" Greece more than we do New Guineawe read all their ancient playwrights, we study their philosophy, we relate to them more, we take holidays there. In other words, we uncover not a principle but a bias (at worst, simple racism). The first thing to note is that to uncover a bias for, say, Greece over New Guinea would no more show that we were mistaken in finding desirable the return Greece's cultural treasures than it would show that we were mistaken in finding undesirable the return of New Guinea's cultural treasures. Discovering that a difference in attitude towards A and B is unprincipled bias, and attempting to formulate a policy of action that does not reflect that bias, could go either way: it might show that we ought to treat neither A nor B as Φ, or it might show that we should treat them both as Φ. What would adjudicate which way to go? Answer: the principles that underlie our unbiased attitudes. We have already seen an unbiased moral principle that justifies the return of cultural treasures (not showing their return to be obligatory, but at the very least desirable). If we find that there are items that count as cultural treasures that we don't want to returnsay, the New Guinean maskbut we cannot find any grounds for distinguishing between the two cases, then the only conclusion to draw is that our resistance to returning the masks stems from an unprincipled (and therefore unjustifiable) emotive response.
2.3. The Collective Wrong Argument:
Although treating A as Φ is prima facie morally desirable, you need to consider the principle justifying this treatment (call this principle "P"). P will also justify treating B as Φ, C as Φ, etc. Although any one of these actions, considered alone, seems unobjectionable, all these actions, considered collectively, would clearly be a terrible thing. Therefore treating A as Φdespite initial appearances to the contraryis all-things-considered undesirable.
On the face of it, this is an odd argument. How could a number of actionseach morally correct by itselfadd up to a morally undesirable state of affairs? If we interpret "correct" as a strong moral prescription, like obligation, then the answer is probably that they cannot. If action Φ is morally obligatory, and so is y, and w, etc., then the conjunction of those actions is also morally obligatory. And if the conjunction of those obligatory actions seems intuitively unacceptable, but we are confident of each conjunct taken alone, then so much the worse for our former intuition.
However, if we interpret "correct" as something weaker, like moral desirability, then the Collective Wrong Argument gets a little further. It may be morally desirable for a particular person to give some money to the charity Save the Children. It is also, let's say, equally morally desirable for her to give some money to Greenpeace, and to Amnesty International, and to Save the Manatee, and to all those other charities. But it is not morally desirable that she gives money to all of them, for then she would be bankrupt, her family would suffer, etc. It is precisely because of this consequence that we generally resist claiming that it is morally obligatory for people to give money to any particular charity (though we may hold that it is obligatory for people to give some money to some charity or other). And so it is possible that some action Φ might be morally desirable, and some action y equally so (and w, etc.), but the conjunction of these actions would be morally undesirableit may even be morally prohibited.
The obvious thing to recognize is that from these observations the conclusion of the Collective Wrong Argument doesn't follow at all. From the fact that giving to every charity would be morally undesirable (perhaps even prohibited), the silliest thing to conclude would be that one shouldn't give to any charity. That would be like thinking that one shouldn't give birthday presents to anyone on the grounds that one cannot give birthday presents to everyone. So if we accept (for the sake of argument) that it would be quite unacceptable for the museums to be drastically depleted, and that there are a sufficient number of items in the museumseach of which it is morally desirable to returnsuch that if all such items were returned the museum would be drastically depletedit by no means follows that none of these items should be returned. Rather, the appropriate way to proceed would be to assess carefully which of the potential actions would be most morally desirable. In much the same way one distributes a limited amount of money among charities, all of which seem to a greater or lesser degree deserving: one gives it to the most deserving first.
The following question naturally arises. Suppose one returns the item for which there is the most compelling moral case, and then the item for which there is the second most compelling case, etc.at what point does one stop? Aren't we on a slippery slope to an unacceptable state of affairs: barren museums? As usual, there is more rhetoric than genuine gradient to this slope. It is useful again to compare the case of giving to charity (not that returning cultural treasures falls into the category of charitable actions). The fact that it is highly desirable for someone to give some of her money to charitylet's say, 10%in no way raises the possibility of her giving away all her money. But where to draw the line? Why 10% as opposed to 15%? And if 15%, why not 20%? It is true that it is a difficult decision to make, and perhaps it is too much to expect strict percentages. Still, in so far as one is able to judge that giving some money to charity would be good, but giving all one's money to charity would be a disaster, then one should be able to formulate some sort of policy for approximately how much to give. And even if one is not entirely confident that one has gotten it right, one might at least be certain that 10%, as a rule of thumb, is closer to "the right amount" than zero. Exactly the same thing goes regarding the return of items in museums: if one has a conception of how much depletion would count as "unacceptable", then one can afford to return treasures until one approaches that mark. The fact that a certain vagueness and complexity surrounds the assessment should no more hinder this decision than it should paralyze a person when deciding how much to spend on his mother-in-law's birthday present. Taking various factors into account, he might decide that $50 is about right. Of course, it's absurd to think that $50 is "just right" and a penny more would be unacceptable: if he sees something nice for $55 or $60, then maybe he'll stretch that far. But the fact that there's no exact line to be drawn hardly proves that he's on a well-oiled slope to spending thousands of dollars on her birthday present.
It should be stressed that to some degree the above discussion is incidental regarding the return of cultural treasures. This category of items is defined in such a way that in all probability in any given museum there are not a great number of objects satisfying the criteria. If we consider the British Museum, which is presumably an extreme case, the idea that the removal of its cultural treasuresthe items that are of paramount significance to a living groupwould in any serious way deplete the museum, is untrue. (It might be noted that the museum has so much material in its collection that much is in storage, never seen by the public.) And even if it were true that their removal would result in unacceptable depletion, that is no justification at all, this paper has argued, for refraining from returning those items for which there is the most pressing case, and merely ensuring that in this process one stops well short of approaching "unacceptable" limits.
2.4. The Hardened Heart Argument:
Although treating A as Φ is prima facie morally desirable, doing so will alter our attitudes towards treating things as Φpretty soon we'll come to think of treating B as Φ as acceptable too; and if we decide to treat B as Φ we will soon "harden our hearts" further, and we'll begin to think that treating C as Φ is acceptable too (etc.). But treating C as Φ is clearly unacceptable. Therefore treating A as Φdespite initial appearances to the contraryis all-things-considered undesirable.
This and the Continuum Argument are, it seems, the most frequently employed slippery slopes, but this argument is no more compelling than the earlier one. The argument involves a projected chronology:
Time 1(us now) Returning A seems acceptable, but returning B or C seems unacceptable...
Time 2... so we return A (but keep B and C).
This brings about a change of attitude:
Time 3 Now returning B seems acceptable, but returning C still seems unacceptable...
Time 4... so we return B (but keep C).
This brings about a change of attitude:
Time 5 Now returning C seems acceptable...
Time 6... so we return C.
Result: Empty museums!
Here "A" might denote the Elgin Marbles, "B" a rather nice Mexican statuette, and "C" some pieces of unexceptional Egyptian pottery. (Three steps have been chosen for clarity, but obviously that's just for illustrative purposes.)
The argument seems to trade on a rather pessimistic view of human weakness: that we can somehow talk ourselves into judging as acceptable things that we now think of as morally unacceptable. Why, exactly, we would grow so feeble as to lose all capacity to distinguish the acceptable cases from the unacceptable, is far from obvious. After all, at time 1 we have a difference in attitude between treating A as Φ and treating C as Φ, and something must account for that differentiation. Unless we are being simply indefensibly biased at time 1, there must be some difference between A and C to which we are sensitive. The important question is whyas we proceed from time 1 through to time 2 and onwardswould that sensitivity be eroded? After we move past time 2 we will be treating A and C in a different mannerand surely that, if anything, is likely to reinforce a sensitivity to the difference between them. Treating them both the same, on the other handeither treating them both as Φ (as at time 1) or treating neither of them as Φ (as at time 6)is likely to dim our awareness of the difference between them.
Furthermore, with a view of human attitudes being so malleable in the face of precedent, the argument undermines itself. It accepts that we could get ourselves into a "hardened heart" situation (at time 5) where we judge that treating A, B and C as Φ are all acceptable, even though we are quite wrong, but we are seemingly unable to recognize this fact (at time 5), having unwittingly corrupted ourselves. But if that could be true of time 5, how do we know it's not also true of our starting attitudes, at time 1? Perhaps the attitude we have at time 1, wherein treating C as Φ is seen as unacceptable, is itself a product of having been influenced by earlier precedents, but we are unable now to see this. Somehow, we are supposed to be able to "tell" that the attitudes we have at time 1 are the correct ones, though the argument depends on the possibility of that we might get ourselves into a situation wherein we have the wrong set of attitudes but are oblivious of the fact. This is a troubling contradiction.
Two criticisms have been made of this argument. First and more importantly, why would our attitude change? If our initial differentiating attitude was based on some principlesome defensible difference between the marbles and (say) the unexceptional Egyptian pottery, such as the former but not the latter being a cultural treasurethen there's no reason at all to think that at any later date we'd "forget" about that principle. If someone were later to ask "Well, you returned those marbles, why not now return that pottery?" we'd have an answer to give: "Because the marbles play a crucial role in some culture's sense of national identity while the pottery does not". And if there were no such difference in principle that could be offered, then the attitudes that initially distinguished between the cases would be thrown into doubt (see discussion of the Principle Argument).
Second, suppose we were so malleable as to change our attitude towards the pottery at some time after returning the marbleswhy would that be such a bad thing? The fact that now we see it as a bad thing (we're supposing) is no more significant than the fact that in that projected future (at time 5) we'd look back on how we had once been inclined to keep both the marbles and the pottery and think that that was a bad thing. It is not a defensible practice to privilege the attitudes we have now for no other reason than they are the ones we have now. It is no more acceptable than it is for me to privilege the moral attitudes of my culture for no other reason than their being the attitudes of my culture. Justifying a moral decision with "These are my attitudes that I have now" is no justification at all; to make a valuable moral decision involves looking beyond such contingencies, and examining the general principles that underlie one's attitudes. If we can't find such a justification, then our attitude is unjustified, and we certainly shouldn't be basing moral decisions upon it. But if we can find such a principle, then this makes a nonsense of the claim upon which the Hardened Heart Argument depends: that we might set off down a slope upon which we become increasingly unable to distinguish right from wrong, until finally at the bottom, with hearts having been hardened by our journey, we are willing to countenance all manner of callous and foolhardy actions.
Not only is the slippery slope argument a fallacious way of arguing, it is self-undermining. To base one's reasoning upon it is to admit that a certain action has prima facie desirability, and that another action, or set of actions, is undesirable. But to admit this is to acknowledge that there must be some difference between the two cases, accounting for these opposed attitudes. This, in turn, implies that there is a difference to be brought into the light and written into the decision-procedure, thus preventing one sliding. If, on the other hand, one denies that the first action is even prima facie desirable, then an appeal to the fear of slippery slopes is entirely superfluous, and one should instead be able to establish the undesirability of the proposed action on its own merits and without reference to the undesirability of other possible actions.
Disqualifying the slippery slope argument against some proposed action is not, of course, to show that action to be perfectly permissible; it is merely to turn aside one obstacle that might stand in the way of the prima facie desirability translating into an all-things-considered desirability. But it is noteworthy that considerations of "precedent" loom so large in all debates over the return of cultural treasures, the case of the Elgin Marbles being no exception. Such fears are based on fallacious reasoning and should be put aside, allowing us to judge the moral arguments in favor of and against returning each cultural treasure on a case-by-case basis. Issues of who legally owns the item in question will, naturally, be of great importance in any such debate. But, as was stressed at the outset, the moral argument does not end there. If you find in your possession an item of immense sentimental value to another person, and of comparably lesser value to you, then your refusal even to consider returning the item, or to openly discuss the possibility, certainly would reveal a serious moral flaw. How morally desirable (or otherwise) it is for you to return the item depends on the particulars of the matter. But it is certainly possible that even if you are judged "within your rights" to keep it, you might still be accused legitimately of a range of moral vices: of being mean, miserly, and heartless. The debate over who owns the Elgin Marbles is a worthwhile matter to pursue, but anyone who thinks that the question of ownership exhausts the ethical question is guilty of impoverished moral thinking: a peculiar modern impoverishment that recognizes only one type of ethical crimethe violation of a rightand forgets that one can morally err in many ways.