I’m happy to introduce our current Featured Philosophy, Gwen Bradford, who teaches at Rice University and has written a creative and insightful book on achievement.  Her post today is on the nature and value of uniqueness.  Please comment with your thoughts about the interesting new territory that Gwen is exploring!


I have been thinking about uniqueness and its relationship to value.

The issue first arises in one of the important moments in value theory. The orthodox conception of intrinsic value as value strictly in virtue of intrinsic properties was questioned by counterexamples pointing to extrinsic properties generating what’s plausibly intrinsic value. Monroe Beardsley in 1965 wrote this:

One inconvenience of this definition can be brought out as follows: A sheet of postage stamps has been misprinted – the central figure, say, is inverted. …[but] its value is not for the sake of anything else. (Beardsley 1965: 61-62).

And other philosophers since followed with examples of uniqueness shifting the conception of intrinsic value away from the traditional Moorean one (E.g., (Kagan 1998: 282-283; Rabinowicz and Rønnow-Rasmussen 2006: 120).

I think that there are (at least) three kinds of uniqueness, and none of them have the relationship to intrinsic value upon which the original examples rely. Here I am going to sketch the three kinds, and I would like to hear what you think.

Type 1 Uniqueness: “this one is different from all the rest”

Beardley’s stamps are unique insofar as they are different from the other stamps of the issue. This is the commonsense Type 1 Uniqueness: something is unique just in case it is different from other members of its kind. To be precise, when there is a norm of similarity across the members of a kind, a member of the kind is unique when it has some property or properties that distinguishes it as dissimilar from the other members of the kind. It is related to rarity – a member of the kind is rare when there are relatively few of them that have the properties that make them dissimilar to other members of the kind.

Clearly, however, not just any old thing accrues intrinsic value when it is dissimilar from other members of the kind. If there is a malfunction at the toothbrush factory resulting in a two-headed toothbrush, it would not deserve to be enshrined in a museum. This supports the conclusion that uniqueness enhances pre-existing intrinsic value, but does not imbues it to anything.

And yet, what have we just seen as the showcase example of the relationship between uniqueness and intrinsic value? Something that’s one of the very archetypes of not intrinsic but mere instrumental value, namely, a stamp. Rare stamps and coins are the most obvious examples of uniqueness enhancing intrinsic value – but stamps and money are clear instances of things whose value is instrumental and not intrinsic. So either uniqueness does imbue things with no prior intrinsic value with value, and a two-headed toothbrush is, in fact, intrinsically valuable, or we are mistaken in thinking that uniqueness is relevant for value in this way.

Perhaps you are not keen on philately and are willing to simply reject that uniqueness is relevant for value. But there are other very compelling cases that we should not be so willing to cast aside. Many of us consider a good use of our time to spend the day at the museum, admiring great works of art and historical artifacts, such as Ancient Egyptian sarcophagi or medieval suits of armor, and (at least part of) what makes these objects interesting and worthy of our attention is that they are unique – you can’t see anything like them anywhere else. At least part of their value, it seems, is related to their uniqueness.

Type 2 Uniqueness: “Nothing else like this one exists”

I want to suggest that in these compelling examples, it is a different kind of uniqueness that matters. And, moreover, the relationship to value that arises from Type 2 Uniqueness concerns a change not in amount of value, but kind.

I propose that Type 2 Uniqueness is a matter of having properties that can no longer be newly reinstantiated. Of course, properties by definition are such that they can be multiply instantiated, i.e., are borne by multiple objects. But some properties, simply as a matter of contingent fact, can’t be reinstantiated anymore, such as the property of being painted by Picasso or being an Ancient Egyptian sarcophagus. It is this sort of property that I want to suggest is responsible for our intuitions about the significance of uniqueness. Like Type 1 Uniqueness, Type 2 Uniqueness is also related to rarity – an unreinstantiable property may be instantiated in more than one object, and the significance increases as this number grows smaller.

Now, of course virtually everything has contingently unreinstantiable properties. The pen that you just used to write down a note is the only pen that will ever have the property of you having used it just now to write down that note. To be sure, uniqueness is ubiquitous. What it significant in some cases but not others?

Consider good-making properties in general. Many (if not most) good-making properties are reinstantiable. Being pleasant, being beautiful, or bearing the autograph of Beyonce are all good-making properties that are reinstantiable. Other good-making properties, such as being an Ancient Egyptian sarcophagus can’t be newly instantiated, simply as a matter of historical fact. (Of course, there’s the substantive question of which properties are good-making, but they seem to be treated this way and it’s not implausible to think we are correct in doing so.)

Here is the suggestion about unreinstantiable good-making properties: they result in irreplaceable value. Irreplaceable value is value that, if lost, can’t be replenished in the same way, even if it could be replenished in amount. There isn’t a great deal of discussion in the literature conceptualizing irreplaceable value, but one way in which things might be thought to have irreplaceable value is by having good-making properties are the sort of properties that can’t ever be reinstantiated to make anything else good again. So objects that have good-making properties that are unreinstiantiable are irreplaceably valuable. Their value, one might say, is unique.

So Type 2 Uniqueness is a way for value to be unique. This suggests that we were misled in thinking earlier that uniqueness is a formal property that generates intrinsic value. I’m suggesting that Type 2 Uniqueness and irreplaceable value, not Type 1 Uniqueness and augmented intrinsic value, explain intuitions about cases where uniqueness is relevant for value. Previous philosophers were mistaken in thinking that uniqueness enhances intrinsic value. Rather, it is a matter of irreplaceable value resulting from bearing unreinstantiable good-making properties.

What does this suggest about the Moorean conception of intrinsic value? Nothing conclusive yet, but it is down an important counterexample, suggesting that the debate merits revisiting.

Type 3 Uniqueness: “nothing else like this one could ever exist”

Now I haven’t yet said anything about people. Uniqueness matters for the value of persons, and, if anything does, persons have irreplaceable value. From this one might conclude that persons have Type 2 Uniqueness and this is what characterizes our irreplaceable value. But I don’t think this is right. Type 2 Uniqueness is a matter of contingently unreinstantiable good-making properties. To be sure, people have many good-making properties that are contingently unreinstantiable, such as personal histories or talents. But the value that people have in virtue of these properties surely does not exhaust the sense in which persons are irreplaceable. The irreplaceability of persons seems more aptly characterized as necessary, not merely contingent. Moreover, there is a relationship between Type 2 Uniqueness and rarity that is not true of persons. When unreinstantiable good-making properties are multiply realized, their bearers may be rare, approaching unique as the number shrinks. But people are not like that – even if sometimes we may describe someone’s talent as “rare” it’s not the case that the irreplaceable value of persons is a matter of exceptional traits. People with very common traits are just as irreplaceable as those with rare ones.

What distinguishes the irreplaceable value of persons seems more accurately characterized as something that could not possibly be multiply realized. It is a matter of something that is necessarily unreinstantiable.

But this sounds very mysterious. What is at the root of Uniqueness Type 3 and the irreplaceable value of persons? I look forward to hearing what you think about this question as well as my arguments above.


Works Cited

Beardsley, M. C. (1965). Intrinsic Value. Philosophy and Phenomenological Research, 1-17. Reprinted in Rønnow-Rasmussen, T., & Zimmerman, M. J. (Eds.). (2006). Recent Work on Intrinsic Value (Vol. 17). Springer.

Kagan, Shelly (1998). Rethinking Intrinsic Value. The Journal of Ethics. 2(4), 277-297.

Rabinowicz, W. and T. Rønnow-Rasmussen (2006). A Distinction in Value: Intrinsic and for its Own Sake Reprinted in Rønnow-Rasmussen, T., & Zimmerman, M. J. (Eds.) Recent Work on Intrinsic Value (Vol. 17). Springer.

13 Replies to “Uniqueness (by Gwen Bradford)

  1. Hello, Gwen. This is a thought-provoking set of examples and distinctions.

    I have never been impressed by the inverted-stamp example. Its (financial) value is a simple function of supply and demand. If we think of “value” as what price is or ought to be tracking (“Why is the price so high?” “It’s valuable”), then unique or rare things tend to be more valuable because supply is restricted. If however we think of “value” along the lines of what demand is or ought to be tracking (“Why do so many people want this?” “It’s valuable”), then appealing to uniqueness doesn’t help. It seems strange that there is a community of people who collect stamps, but no community of people who collect toothbrushes, but that’s the way it is.

    Type 2 uniqueness, contingently irreplaceable value, seems analyzable like this: X has contingently irreplaceable value iff X is valuable in virtue of P, and Ps are contingently irreplaceable. (This artifact has value in virtue of being an ancient Egyptian sarcophagus, and ancient Egyptian sarcophagi are contingently irreplaceable.)

    Similarly, Type 3 uniqueness seems analyzable as: X has necessarily irreplaceable value iff X is value in virtue of P, and Ps are necessarily irreplaceable. (Adam has value in virtue of being his unique self, and that unique self is necessarily irreplaceable.) On this view, the irreplaceable value of persons is just a function of (a) the essential irreplaceability of persons, and (b) persons are valuable qua persons.

  2. Hello Heath,
    Glad you find this interesting!

    Indeed, the analysis that you propose for the value associated with Type 2 Uniqueness sounds like you’ve understood my proposal precisely.

    But the suggestion for the value associated with Type 3 Uniqueness is not yet illuminating. The very thing that I would like to understand is the irreplaceable value of persons. Properties, of course, are the sort of thing that at least in some sense could be multiply realized. This is simply what it means to be a property (i.e., a universal). It’s an oxymoron for a property to be impossible in any sense at any time to be reinstantiated. So one cant’ say that “Ps are necessarily irreplaceable” if we mean by that “Ps are properties that are necessarily unreinstantiable” because those Ps would not be properties. So whatever it is that is responsible for the irrplaceable value of persons isn’t a matter of properties. But then what is it? This is what puzzles me.

  3. Hello Gwen,

    It seems to me that quite a few properties are necessarily unreinstantiable. “Being (identical to) Socrates” is one. So is “being Socrates’ mother”, “being Socrates’ only dentist”, “being Socrates’ 912th toenail clipping.” Other examples would be “being the inventor of the calculus,” “being the first man on the moon,” “being the shortest philosopher in Oklahoma at t.”

    These examples indicate that Type 3 uniqueness is not enough by itself to confer value, since some bearers of these properties are valuable and some are not.

    Here is a thought: maybe it is significant that “being Socrates” is a property only one possible individual could have, while other properties could be had by multiple individuals. (Type 4 uniqueness?) The idea would be that if Socrates were not around to be Socrates, there would be an unfillable Socrates-shaped hole in the world; while if Leibniz had not invented the calculus, Newton would have. You could think of this as a version of economists’ “value at replacement cost.”

    I don’t think that this is the whole story about the value of persons, since it doesn’t say why being Socrates is valuable in the first place. But it maybe it helps.

  4. Gwen, thanks for the interesting post. I have one clarificatory question and one distinction to make.

    Regarding Type 2 Uniqueness, do you deny that unique value is also a kind of intrinsic value? Or is it another sort of non-instrumental value, distinct from intrinsic value? I’m inclined to think that some things are uniquely intrinscially valuable. I’m also inclined to think the stamps example downplays the significance of the counterexamples to hte Moorean view—-there are many better counterexamples one can come up with). But I’d like to hear your thoughts.

    Regarding the irreplaceability of persons, I’d make a distinction between (1) the unique value of particular people to other people in virtue of their relationship (see e.g. Christopher Grau on this) and (2) the unique value of persons in general.

    The former clearly involves some extrinsic properties in virtue of which people can be uniquely valuable to one another. This might even work without reciprocity. Two examples. First, the way we value some artists, athletes, and historical figures is also partly in virtue of some relation to us, our society, our values, etc. Second, everyone is unique insofar as they have personal projects, values and commitments. We think their lives are specially valuable in virtue of those distinctive relational features, not just because they bear properties that can be multiply instantiated (being a person’s life). In some sense this makes them irreplaceable (see below), but I wonder if this makes them uniquely valuable.

    On the other hand, persons in general are commonly taken to be irreplaceable in virtue of specific capacities that they have——e.g., some form of self-consciousness, sense of themselves over time, narrative identity, robust future-directness, rational autonomy, what have you. But we don’t think this is what makes each person unique, although this arguably sets persons in general apart. Here, irreplaceability means they cannot be killed (replaced) without loss (see e.g. Peter Singer); uniqueness seems to do little explanatory work here. After all, even if a person did not exercise their person-capacities in a way that made them unique, they would still be irreplaceable, right? Does this distinction seem right to you?

  5. Nicolas,
    That’s right, I don’t deny that irreplaceable value is intrinsic (aka final) value. quite the contrary, it’s plausibly understood to e a species of it, or a “way” to have intrinsic value. I think that there are other very compelling counterexamples to the Moorean conception of intrinsic value (the dress worn by Princess Diana, the pen used by Lincoln to sign the Emancipation Proclamation being two of the most well known). It might be worth investigating whether all these other seemingly compelling cases are also mistaken – perhaps they too are shifts to irreplaceable value – but it seems to me that they are good examples nonetheless, illustrating augumented intrinsic value in virtue of extrinsic properties. The only case of which I’m fully skeptical is Type 1 Uniqueness, as I argued.

    About persons: the philosophical literature about love and relationships often takes up something similar to Type 2 Uniqueness (such as the paper by Grau). What makes people valuable to us in relationships may involve contingently irreplaceable history.

    But this sense of “uniqueness of persons” doesn’t exhaust the other sense — this is the illusive Type 3, which has to do with the irreplaceable value of persons more generally. And, indeed, just as you point out, many philosophers hold that the irreplaceability of persons has to do with something like rational autonomy, practical identity, or whatever it is that fundamentally characterizes persons as such. But this is interesting: the property that all persons have as such – and so is multiply instantiable – grounds irreplaceable value, the very hallmark of which seems to be irreplaceability. The answer, presumably, has to do with the nature of whatever it is that characterizes all persons as such. Only once we understand the nature of this property will we see how it can be multiply realizable yet irreplaceable. Perhaps it is kantian autonomy, or something yet to be explored.

    And to clarify – indeed Type 1 and Type 2 uniqueness do little explanatory work as far as irreplaceability of persons in general goes. The point here is that there seems to be a third sense of uniqueness.

  6. Thank you, again, Heath. That’s helpful. But some of the issue depends on the theory of properties, and the puzzle reiterates.

    On one entry-level theory of properties, properties are universals – this precludes the sort of thing that necessarily could not be multiply instantiated.

    But of course there are other theories of properties. Suppose “the property of being Socrates” is indeed a property. But now, as you point out, one might ask, why is this a good-making property? The answer cannot be what’s in common with the answer to “why is “the property of being Alcibiades” a good-making property?” because, then, of course, it would be some further property instantiated by both. So ever individual needs to have a unique property of being themselves and that property would have to be a good-making property. There would then be zillions of very different good-making properties. We would have a perplexingly messy axiology. This is the quesiton, then: why, for any such “being the one and only X” is this good-making? So there is another version of the puzzle that arises.

  7. Hi Gwen,

    I wonder whether we could also distinguish between irreplaceable and unrepeatable value. Many people value their first kiss and other first experiences of some type. With Socratic encouragement they often say things like, “I will never forget it. No later kiss was the same. Not even the kisses that had more meaning”.

    I am not sure that first kisses and other valued firsts are valued or valuable because they are unique. But if so, I am thinking they may point to unigueness sometimes having unrepeatable value rather than irreplaceable value.

    Here is what makes me think they can be replaceable: Imagine that I think I am going to have my first kiss tonight but the date gets rained out. Happily, I will have a kiss tomorrow that has replacement value of the same kind but, let’s say, slightly lower quantity (the rain delay casts a shadow on the proceedings). I would describe this as a case in which the missed value was replaced but not fully compensated. Of course we can keep the quantity of value the same to have replacement and compensation in another case. The basic thought is that my actual first kiss on the second date will provide me with replacement unrepeatable uniqueness value.

  8. Hi Gwen,

    For what it is worth I’ve written on these topics in a number of places. I’m not sure which paper is being referred to in the comments above by Nicolas (probably “Love and History”) but “Irreplaceability and Unique Value” is the most relevant given your particular concerns. I don’t have a satisfying answer to your queries about “the illusive Type 3, which has to do with the irreplaceable value of persons more generally” but Nora Kreft has some unpublished work on this issue which I think makes headway.

    There was also a useful thread on irreplaceability and Erich Matthes’ paper on PEA Soup a few years back which seems relevant (though I realize you want to focus on persons rather than objects):



  9. Hello Chris,

    Nice to virtually meet you! I’ve read your work and got quite a lot out of it. It was in fact “Irreplaceability and Unique Value” that I had in mind above. I enjoyed that paper, and the view you describe is very similar to the value associated with Type 2 Uniqueness. If I understand it correctly, the conclusion of the paper is that the “unique value” of persons and, eg, irreplaceable historical objects is value in virtue of extrinsic unshared properties.

    What I’m suggesting for Type 2 Uniqueness is that it’s less important that the properties aren’t shared. What matters more is that they could not be instantiated anew, as a matter of contingent fact. Consider a Stradivarius violin. More than one exists, but they all have a kind of irreplaceable value in virtue of their contingently unreinstantiable good-making properties. If any Stradivarius is destroyed, we could never replenish its value in kind.

    While persons have properties that are like this, and no doubt they shape the way that we value another person, especially in relationships and whatnot, our extrinsic unshared properties do not exhaust our irreplaceable value as persons as such, impartially. It doesn’t fully explain the irreplaceable value of a stranger to point to their blend of idiosyncratic talents and personal traits. Importantly, the irreplaceable value of persons seems to be a necessary feature of our value, whereas it is merely contingent in the other cases. This is what leads me to articulate the third sense in which people are unique that is associated with irreplaceable value of this sort.

  10. And thank you, I didn’t realize there was a discussion of Erich’s paper! I will enjoy reading over that. Our views also differ in a few key ways, but are also similar. I look forward to looking into the other work you suggest.

  11. Hello Brad,

    That’s an interesting suggestion. It’s not clear to me that there is a distinction between “unrepeatable” and “irreplaceable” value. Supposing that “being the first” can be a good-making property, for any even that has this property, by its nature the property couldn’t be reinstantiated in the same event type. So it ti is a good-making property, it would be associated with Type 2 Uniqueness. So what you describe seems tome to be an example of the sort of value I have in mind.

    What the first kiss example seems to show is that something can have more or less value in quantity even when it has irreplaceable (or “unrepeatable”) value, which is qualitative. That seems right to me. There are other examples also illustrating this point. Consider the various works of Picasso – all of them are irreplaceably valualbe, but the most ingenious and important works, one might think, have a greater quantity of value. If one of his lesser works were destroyed, it would be a loss of irreplaceable value. But if one of his greatest works were destroyed, it would be an even worse loss – a greater quantity of irreplaceable value would be lost, one might say.

    I think the second attempt at the first kiss has irreplaceable value (can only be one first!) but perhaps a lower quantity of value. It’s not that it “replaces” the value of the first — that one simply didn’t happen, and this is the sort of property that can only be instantiated in the same type once.

  12. Hi Gwen,

    Yes, nice to (virtually) meet you! I can definitely see how you take “Irreplaceability and Unique Value” to have that conclusion, but I think my considered view is more complicated (and spelled out better toward the end of “Love and History”). So, I wouldn’t quite want to say that something that has unique value (as I use that phrase) has the value “in virtue of extrinsic unshared properties”. Or at least I don’t want to say that if by “in virtue” you are implying those properties *justify* the attribution of value. Extrinsic historical properties help us track the identity of the entity and so they are crucial for understanding irreplaceability, but I don’t think they can provide justification of the valuation. ( I also think such a justification is not usually needed — I’m probably more Wittgensteinian than you would be willing to accept here.)

    Regarding the idea that “the irreplaceable value of persons seems to be a necessary feature of our value, whereas it is merely contingent in the other cases”, I try to explore this in a chapter of my dissertation that discusses Vlastos’ writings on “individual worth” (in “Justice and Equality”) in relation to the irreplaceability of persons. I’ll spare you the details here (and I’m not completely happy with that now rather old treatment) but I conclude along these lines:

    “Both a wedding ring and a spouse may be valued as irreplaceable, but only the latter possesses the dignity (and deserves the respect) that comes with individual worth. Accordingly, the concept of individual worth is a richer and more substantial notion than that of unique value. To acknowledge the individual worth of a person is to, in part, recognize the unique value (i.e. the irreplaceability) of that person. It involves more, though: individual worth brings with it the idea that a person ought to rightly be regarded (by herself, and others) as “beyond price” and irreplaceable. In other words, valuing a possessor of individual worth as irreplaceable is not simply permissible, it is morally required. […] Following Vlastos, I want to argue that the respect due to all persons involves our recognition that they possess individual worth, and this recognition is in turn linked to our awareness that persons possess both the capacity for choice and the capacity for consciousness (i.e. what he refers to as well-being). Though Vlastos does not explicitly say this, these capacities function as conditions in the following respect: something that possessed neither capacity could not plausibly be thought of as possessing individual worth. It does not follow from this, however, that something lacking both these capacities (e.g. an artifact or natural object) could not possess unique value (that is, be legitimately valued as irreplaceable). It should also be clear at this point that it does not follow that individual worth ought to be seen as proportional to the degree to which these capacities are manifested – the existence of the capacities for choice and consciousness does not function as a justification for attributions of individual worth in that manner (e.g. the diminishment of one’s agency, as in the case of a person with dementia, need not bring with it the diminishment of one’s individual worth.) Finally, I want to reaffirm Vlastos’ point that recognizing the importance of these capacities is fully compatible with leaving the question open regarding whether other capacities or features of persons are relevant for a complete and proper understanding of individual worth and the dignity that comes with it.”

    If you want to see the rest, I’m happy to send it your way, but like I said I’m not fully happy with it, and I think my views may have shifted somewhat such that I am now less sure that there is either a rational or moral requirement to grant all persons irreplaceable value. To put the issue bluntly, I don’t currently think someone like Parfit (or perhaps a certain kind of Buddhist) is *irrational* or necessarily *immoral* in denying irreplaceable value to persons. I do think enough good arguments exist to show that Parfit’s position is not rationally *required*, and a different framework in which irreplaceability is accepted (and perhaps even viewed as a morally necessary feature of persons) is also defensible as reasonable. Or, at any rate, these days I find myself shifting between this more pluralistic position and my earlier stronger position that attributing irreplaceable value to persons is required by morality. Perhaps your work will be able to convince me to go back in that direction!

  13. Hi Gwen, thanks for your reply.

    “But this is interesting: the property that all persons have as such – and so is multiply instantiable – grounds irreplaceable value, the very hallmark of which seems to be irreplaceability. The answer, presumably, has to do with the nature of whatever it is that characterizes all persons as such. Only once we understand the nature of this property will we see how it can be multiply realizable yet irreplaceable.”

    My train of thought was that we’re equivocating on the term “irreplaceable”. Persons qua persons perhaps don’t have irreplaceable value, rather a distinctive type of worth or final value in virtue of a characteristic that is multiply instantiated. But since their value is not irreplaceable value, the mystery disappears. Treating them as “irreplaceable” is just the appropriate response to their final value and/or their person-making features (autonomy, practical identity, self-consciousness, etc.) rather than a response to something that is uniquely instantiated. On the other hand, persons qua individuals have not just final value in the above sense but also irreplaceable value in virtue of irreplaceable histories and other non-reinstantiable features.

    I think the distinction is important. We don’t want the value of persons to depend on interpersonal relationships; on the other hand, it makes sense to think that some such relationships are and should be grounded in part (though certainly not exclusively) in mutual responses to the value persons have qua persons. Or a weaker possibility, since this starts to sound like a Kantian view I do not share, is that person-making features enable the construction of interpersonal relationships, and vice versa——in turn, through those relationships we come to see our friends, partners, lovers, fellows, etc. as persons in a way we may not have otherwise.

    So, to sum up, there the irreplaceability of persons qua persons doesn’t have to do with uniqueness. You can have a consistent story about properties, uniqueness and irreplaceable value, while holding that persons qua persons are irreplaceable in virtue of the characteristics that ground their final value. Of course it’s puzzling, because the multiply instantiated properties in virtue of which we are persons still do not allow for replacing persons with one another. You can’t kill people and make up for the loss by creating new individuals with person-making features.

    Now, here’s a hint at a potential resolution. If persons cannot be killed and replaced this is not because those features aren’t multiply, repeatedly instantiable. Instead, this is because the exercise of these features produces irreplaceable experiences. I’m not fully confident this is a palatable resolution because it would seem to make the irreplaceability of persons contingent on their exercising their person-making features in a certain way (some utilitarians gesture toward such a view): their lives have irreplaceable value simply because they have Type 2 uniqueness. The distinction between type 2 and type 3 seems to collapse.

Comments are closed.