A Potentiality Argument
November 24, 2009 — 7:25

Author: Michael Almeida  Category: Uncategorized  Comments: 25

Updated 11.25.09
I’ve been thinking in general about potentiality arguments. In some cases, they seem undeniably cogent. But there are what seem like analogous cases in which the cogency of such arguments is denied. It’s hard to see why. So, suppose the claim in (1) is true.
(1) The tumor in Smith’s abdomen is currently benign and presents no threat to him, but if it is left untreated it will develop into something that is seriously life-threatening.
Now imagine someone saying that the fact in (1) gives us no reason to treat the tumor in Smith’s abdomen. Imagine someone saying that since the tumor is currently benign we have no reason to remove it. That position is just bizarre. We think that the potential of the tumor to develop into something life-threatening is not just a reason, but an excellent reason for us to remove it now, before it has those life-threatening features. Intuitively, it seems a serious wrong to fail to remove it, providing we can, and other things are equal. So, we naturally and cogently reason this way:
(2) If Smith’s tumor will develop into something life-threatening, then we ought to remove it now, prior to it’s development into something life-threatening.
(3) Therefore, we ought to remove it now, prior to it’s development into something life-threatening.
That argument is pretty compelling, despite the fact that it tells us to remove something that is benign; something that does not yet have any life-threatening properties. Why doesn’t the same reasoning hold in this case?
(1′) The being in Smith’s womb is currently without much value, but if it is left untreated it will develop into something that is extremely valuable.
(2′) If Smith’s being will develop into something extremely valuable, then we ought not to remove it now, prior to it’s development into something extremely valuable.
(3′) Therefore, we ought not to remove it now, prior to it’s development into something extremely valuable.
In the initial argument we conclude that we ought to remove X because X IS NOT, but WILL BECOME threatening, IF LEFT UNTREATED. In the second argument we conclude that we ought not to remove X because X IS NOT, but WILL BECOME extremely valuable, IF LEFT UNTREATED. How can the first potentiality argument be cogent and the second potentiality argument not cogent? If the potential for developing extremely bad properties is an excellent reason to remove X before it develops those properties, then why isn’t the potential to develop extremely good properties an excellent reason not to remove X before it develops those properties? It’s hard to see.
Perhaps the relevant difference is (cf. Matt’s comments) that the initial argument involves potential instrumental disvalue (the threat to someone’s life) and the second argument involves potential intrinsic value. But then consider,
An Argument from Intrinsically Evil Beings
(1*) The being in Jones’ womb is currently without much value, but if it is left untreated it will develop into something that is extremely intrinsically evil.
(2*) If Smith’s being will develop into something extremely intrinsically evil, then we ought to remove it now, prior to it’s development into something extremely evil.
(3*) Therefore, we ought to remove it now, prior to it’s development into something extremely intrinsically evil.
That argument again seems cogent. But then why doesn’t (3′) follow convincingly from (1′) and (2′)?

Comments:
  • Eric

    Here’s a problem I can see with the potentiality argument. Why draw the line at the point where a “being” is created in the womb? It seems like people are always facing the potential for creating an extremely valuable being, but surely there’s nothing wrong with choosing not to procreate at any given moment. I can’t see a reason to distinguish the cases without presupposing that fertilization of an egg is morally significant.

    November 24, 2009 — 21:53
  • ZG

    I see no reason why the second reasoning doesn’t hold…it doesn’t seem unnatural or odd to me, at all!

    November 24, 2009 — 23:20
  • Matt

    One difference between the two arguments is that the reason we ought to remove the tumor is that it threatens Smith’s life, whereas, if we ought not remove the fetus, the reason has to do with the fetus itself. Putting this more in the language of your last paragraph, the tumor has the potential to develop properties that are bad for Smith, whereas with the fetus the relevant properties are (primarily) good for it. Whether this is a distinction that makes a difference is going to be controversial, I would think.
    One reason to worry about taking these arguments together is that they seem to imply contradictory obligations in the case of a pregnancy that threatens the life of the mother. The first tells us that we ought to remove the fetus, and the second tells us we ought not do so. This suggests that one of the arguments might be too general.

    November 25, 2009 — 2:36
  • Mike Almeida

    Here’s a problem I can see with the potentiality argument. Why draw the line at the point where a “being” is created in the womb??
    Eric, that’s a fair question, but I don’t take the egg+sperm to be an object with a potential, since (i) I don’t take them to compose an object and (ii) if they did compose an object, it wouldn’t the sort of object that is potentially a very valuable being. The case you’re talking about is the potential to have the potential to develop into a very valuable being.
    ZG, I don’t see any important difference in the arguments either.
    Putting this more in the language of your last paragraph, the tumor has the potential to develop properties that are bad for Smith, whereas with the fetus the relevant properties are (primarily) good for it. Whether this is a distinction that makes a difference is going to be controversial, I would think.
    Matt, the reasoning appeals to the resulting value of not treating the object in the womb and the object in the abdomen. In one case the value is instrumental, and in the other it is intrinsic (though you could reasonable construe it as instrumental as well). I don’t off hand see why the potentiality for insturmental disvalue would give me a reason to act but the potentiality for intrinsic value would not give me a reason to act.

    November 25, 2009 — 7:23
  • It’s not obvious that the correct/complete explanation of (2) supports the view that there’s a symmetry here. If we should remove the tumor because Smith has a claim or right to (reasonable) assistance in avoiding serious future harms, then it’s not simply the fact of future harm that generates the obligation, but the presence of a right or claim as well. That is, there’s a fact about Smith’s moral status that helps generate the obligation – a fact that is arguably lacking in the case of the fetus (at least it’s a fact the potentiality argument can’t assume).
    Interestingly, the argument from intrinsically evil beings seems to require the lack of such a claim or right on the part of the fetus.

    November 25, 2009 — 15:48
  • Mike Almeida

    If we should remove the tumor because Smith has a claim or right to (reasonable) assistance in avoiding serious future harms, then it’s not simply the fact of future harm that generates the obligation, but the presence of a right or claim as well.
    I’m puzzled by the rights claim. Where would one get such a right against me? Do you have a claim against me that I warn about the slippery sidewalk I just noticed in front of your house or the gangsters loitering on your doorstep? Each of these might be as easy as making a phone call. But I don’t think you’ve got a right against me that I call you.
    Interestingly, the argument from intrinsically evil beings seems to require the lack of such a claim or right on the part of the fetus.
    I’m not sure I follow you. Note that I didn’t say it was a fetus that was intrinsically evil. I’m not sure a fetus or any human being can be so. But I should be more cautious here. I don’t mean a being that has a feature that is intrinsically evil. I have in mind a being–if such a being is possible–who is extremely intrinscially evil and not at all intrinsically good. I ought to add that the being is not contingently intrinsically evil; it’s intrinsic evil supevenes on its individual essence, let’s say.

    November 25, 2009 — 16:12
  • Jeremy Pierce

    Mike, I don’t think you need to assume welfare rights to make Nate’s point. If we have duties to ourselves, then we have a duty based on our own intrinsic value that explains why we ought to remove the benign but dispositionally-malignant tumor. This is because we uncontroversially assume moral status for ourselves. With the fetus, we aren’t assuming any moral status, and thus we aren’t assuming there’s anything that we have duties toward. The consequences in the tumor case have particular value that’s more than just consequentialist value, because there’s already a being who has rights or moral status. We’re not assuming any such being in the abortion case if we want to keep the argument neutral on moral status of the fetus. So at best we’d have consequentialist reasons, whereas in the tumor case we might have deontological reasons.
    If we can allow duties to future beings, then we can get the argument going more strongly, but then I think it’s a different argument. Then we’re back to Marquis, in fact. What’s distinctive about your argument is that it doesn’t assume duties to future beings. But that’s also where the analogy with the tumor might break down, and it might be a crucial place to break down.

    November 25, 2009 — 16:48
  • Mike Almeida

    If we have duties to ourselves, then we have a duty based on our own intrinsic value that explains why we ought to remove the benign but dispositionally-malignant tumor.
    But rights to oneself are not salient, are they? I mean, that can’t be what is behind the intuition that makes the initial inference cogent, since the appeal to rights to oneself is at least controversial. But then suppose this has some influence on our intuitions. The argument still seems cogent under the assumption that there is nothing I can do to remove my own tumor.
    Reconsider the argument from intrinsically evil beings. Suppose we have a human host for some non-human being. And let the being be potentially intrinsically extremely evil, though not life-threatening. The argument seems to go through in such a case, and it does not appeal to the violation of rights.

    November 25, 2009 — 17:50
  • I tentatively think both arguments are cogent, assuming that Smith has given her consent to removal. But that seems to sunder the analogy to the abortion question, where the host’s consent (and own personal flourishing) may be competing values.

    November 26, 2009 — 9:11
  • Mike Almeida

    Mike,
    The argument aims to show that potential properties of the relevant beings can provide excellent reason now either to terminate (evil case) or not. The thesis is that if they do in the evil case then they ought to in the initial case, too. That’s consistent with there being other relevant considerations. So, you can look at it as an argument from consistency.

    November 26, 2009 — 9:51
  • Mike, right, what I’m saying is that I basically accept your argument. I only suggested some reasons to be skeptical that the argument could illuminate the abortion question. But maybe illuminating the abortion question wasn’t your intent.

    November 26, 2009 — 14:46
  • Mike,
    In what sense are you using the word “ought”? Surely in the first case there is an “ought” attached to Smith getting the tumor removed, but it’s questionable whether or not this is a moral ought. Perhaps it is just a prudential ought: If Smith desires the consequence of removing the tumor (i.e. avoiding death), then she ought to do it.
    Suppose that’s the ought being applied in the first case. Then if the cases are supposed to be analogous, you must be careful not to slip in a moral ought regarding not getting an abortion. The ought might simply be: If Smith desires the consequences of not having an abortion (i.e. going through a pregnancy, giving birth, having a child to care for, etc.), then she ought not have one.
    There doesn’t seem to be anything controversial about using the potentiality argument in this manner (though note that moral oughts can conflict with prudential oughts, but this is commonly understood).
    Now your last argument seems to aim to show that this isn’t just about prudential oughts, it’s also about moral oughts. (Or at least that’s the way I’m understanding it.) In that argument you suggest that the potential intrinsic value of the fetus might be negative, in which case there is a sense in which Smith ought to remove it. I think a pertinent question at this point would be this: In what way is the potential value negative? And why is it negative? Suppose that the fetus will turn into a monster at birth and cause extensive suffering and death. In that case, just given the consequences (and let’s stipulate that they’re known consequences), it seems reasonable to say that there’s a moral duty to remove the fetus. But what is the analogy in the case of not aborting? That it is known that the fetus will end world hunger? If that were the case, then yes, maybe there’s a moral obligation to not remove the fetus. But notice that this doesn’t get us any closer to solving the abortion debate, because it’s just consequentialism based on imaginary foreknowledge. Given that we don’t know what the fetus will end up being like (a monster or a saint), we can’t use consequentialism to get us a moral ought in either case. (Unless a good probabilistic argument is made.)
    In short, there might be some ambiguity in describing the second and third cases regarding the future value of the fetuses. You’re suggesting some sort of foreknowledge of their intrinsic value (positive or negative), and it’s true that given such foreknowledge you can derive a moral “ought.” But absent the foreknowledge, I think, it’s questionable whether or not you can still derive the moral ought.
    (By the way, I think you slipped between “Jones” and “Smith.”)
    Also, I want to apologize if what I’ve written is incoherent. It’s late, I’m tired, and I was merely taking a break from reading epistemology.

    November 27, 2009 — 2:12
  • Mike Almeida

    There doesn’t seem to be anything controversial about using the potentiality argument in this manner (though note that moral oughts can conflict with prudential oughts, but this is commonly understood).
    I’m using moral oughts throughout. The reasons I’m considering are agent-neutral reasons that appeal to potential value.
    I don’t think what you say above is true. If it were then abortions would be rarely justified unless we disagree on the great intrinsic value of human beings. It must be at least as valuable to let a potential being realize that potential as it is extremely bad to kill that being. Let V be the intrinsic value B would have were it allowed to develop fully. In killing, we remove the total value V of a human being, and we are rarely justified in doing that. But in not allowing B to develop, we preclude V. So, we should be rarely justified in doing that. But the fact is that most people don’t think it is all that bad not to allow such a being B to realize it’s potential, and most people don’t deny the great value of human beings. So, they must find something wrong with arguing from potentiality in the way I have.

    November 27, 2009 — 8:16
  • Mike,
    Alright, I didn’t get the sense from the first example that you were talking about moral oughts. It makes sense to think of it as a prudential ought, but I understand that doing so does not allow you to make the point you seem to be trying to make–that potentiality arguments give us a moral ought regarding abortion.
    When you say that you don’t think what I wrote is true, I’m not sure what part you’re referring to. But let me reconstruct an argument that you might be trying to make:
    (1) In killing, we remove the total value V of a being B, which is a bad thing.
    (2) In aborting a fetus, we preclude the total value V of a being B.
    (3) The ultimate end of both actions are equivalent (i.e. precluding future values).
    (4) Thus, aborting a fetus must be a bad thing too.
    This may not correspond to the argument you’re trying to make, but it seems like a plausible interpretation of your comment to me.
    Let’s simplify the discussion. In what way do you think it is bad to kill another sentient human being? Is the bad-making property merely the removal of the being’s value? Or the preclusion of the being’s future value?
    It seems to me that given foreknowledge of the consequences (e.g. this fetus will develop into a monster–or saint), we can derive the moral oughts you’re trying to derive. Perhaps the argument you’re making is that the potential value of human organisms is generally very high, thus giving us a moral ought in most cases. I might be able to agree with that, given that you can support the premise that the potential value is generally very high.
    I’m sorry, the more I think about this argument the more confused I’m getting. It would be helpful to have a nice structured step-by-step argument.

    November 27, 2009 — 12:43
  • Mike Almeida

    Maybe this. Let B be a being whose potential moral value is equivalent to the value of H, a normal, adult human being. So, assumption one, where PV is potential value and V is actual value.
    1. PV(B) = V(H) = extremely high
    2. In the absence of very good moral reasons that override the moral reasons not to kill a being of value V(H), it is wrong to kill H.
    3. In the absence of very good moral reasons that override the moral reasons to kill a being of value PV(B), it is wrong to kill B.
    4. It is almost never the case that there is sufficient moral reason to kill a being H such that V(H) = extremely high.
    5. Therefore, it is almost never the case that there is sufficient moral reason to kill a being B such that PV(B) = extremely high.
    Perhaps the obvious response is that killing H is unlike killing B in that killing H violates an extant right and killing B does not. It would take too long to pursue this here, but moral rights are supposed to supervene on natural properties that, in fact, come and go: properties that some normal adult human beings possess only in potential (e.g., adults that are temporarily incapacitated in ways relevant to the possession of the relevant natural properties). But de facto these individuals are not denied their rights, nor are their rights considered diminished. So it is difficult to make the case that infants and fetus’s are denied such rights, either, since they too are potentially in possession of the relevant natural properties.

    November 27, 2009 — 14:16
  • Mike,
    Thanks for the reformulation. Now I can see that the argument is resting on some notion of rights. A normal adult human being has the right not to be killed, so a normal developing fetus should have a similar right not to be killed. Here are a couple of worries:
    (1) Beings with potential value do not necessarily have the same rights as beings with actual value. When Obama was elected, he did not have all of the rights that came with being President until a couple of months later. So it’s an open question whether or not the right not to be killed is one of the rights that the fetus shares with the adult human being.
    (2) Remember Judith Jarvis Thomson’s argument in “A Defense of Abortion.” When you’re connected to the violinist, does the violinist have a right against you not to be unplugged?
    This latter worry brings up an interesting question. Suppose abortion procedures are changed so that, rather than intentionally destroying the fetus with medical equipment, we instead surgically remove the fetus, set it on the counter, and let it survive or expire on its own. This would be more akin to unplugging the violinist, or unplugging a brain dead relative from life support. Surely there’s no overriding moral rights being violated when we unplug in the latter two cases, and since they’re analogous, doesn’t it follow that the new abortion procedure just described also does not violate such rights? If so, then the argument at best serves as a moral reason to change the way we perform abortions.
    Regarding the numbered argument you provided, I wonder what the connection is supposed to be between premises (2) and (3). In premise (1) you say that the fetus has the potential to achieve the same amount of value as the normal adult human being. In premise (2) you say that such value makes it wrong to kill the adult human without overriding moral justification. In other words, in (2) you’re describing a right that the adult human has. Now, in (3) you somehow infer that the fetus has the same right, but that doesn’t seem to follow from (1) and (2). What does seem to follow is that the fetus has the potential to have the same right as the adult human. But such a conclusion is unsatisfying ammunition to use against the permissibility of abortion.

    November 27, 2009 — 15:17
  • Mike Almeida

    (1) Beings with potential value do not necessarily have the same rights as beings with actual value. When Obama was elected, he did not have all of the rights that came with being President until a couple of months later. So it’s an open question whether or not the right not to be killed is one of the rights that the fetus shares with the adult human being.
    First, my argument does not depend on any assumptions about rights at all. It is an argument that appeals to value. I refer to rights in closing only to dismiss a natural line of objection to the argument. What you say above is just the old Feinberg line that Singer reproduces elsewhere. The analogy is no good, as Boonin and others make pretty clear.
    But this is largely irrelevant to my position on the rights a potential adult human being has. I don’t deny that normal adult human beings have rights that fetuses do not. I have no need to deny that. Rather I claim that we have no reason to deny that fetuses have the right to life, if normal adult humans have it. The distinction between adult humans and fetuses that allegedly justifies the distinction with regard to the right to life is that adults have the necessary set of natural properties subvening the right to life (the properties that make the adult a person and so a rights bearer) and fetuses do not. I offer a reductio ad absurdum for this line of argument: the argument entails that newborns, and toddlers, for that matter, are not persons and so do not have a right to life. That they do not have a right to life is absurd, and so I deny that the alleged natural properties are necessary & sufficient to the possession of a right to life. I conclude that there is no basis for the distinction between fetuses and adults with regard to the right to life.

    November 27, 2009 — 15:58
  • Mike,
    If I’m following your argument, it goes something like this: There is no basis for distinguishing between fetuses and adults regarding their having the right to life. So just as it’s wrong to kill an adult human being, it would also be wrong to kill a fetus.
    Suppose a fetus at, say, ten weeks of development, does not have cognitive capacities. I don’t know the actual fact of the matter, so if you know that this is false then subtract weeks until our scientific knowledge gives us a clear answer on the fetus not having a developed complex cerebral cortex. At this point, a fetus is a developing body (which, you say, has a right to life). Perhaps the right that it has is a right not to be killed unjustly. Now suppose we perform a “schmamortion” instead of an abortion. A schmamortion is where you surgically remove the fetus from the woman’s body instead of intentionally destroying it. Does the fetus now have a right against us to keep it alive, rather than letting it expire on the operating table?

    December 4, 2009 — 13:01
  • r.chappell

    Mike – doesn’t the disanalogy stem from the arguments invoking different kinds of ‘value’? Some values ought to be promoted, such that we have reason to realize the potential for additional such value. Others ought to be respected, such that only actual instances ground reasons.
    Plausibly, the value of human life is of the latter kind. We do not necessarily have reason to bring additional people into existence. Their “value” is instead such that once they exist, they will be morally significant: we ought not to harm them, etc.
    N.B. I take it that none of your cases really depend specifically on embodied or de re ‘potentiality’ (not sure what’s the best terminology here). Even if there were no particular tumor with the potential to become malignant, so long as a course of action would (equally) decrease the probability of Smith later developing some tumor, then that surely provides (just as strong) a reason in favour of that preventative course of action.
    So I think Eric is exactly right that the argument you’ve offered here lacks any basis for distinguishing between pre- and post-conception “potentiality” for human life. It’s true that only the latter is de re potentiality. But de re potentiality plays no essential role in the original case.

    December 10, 2009 — 10:54
  • Mike Almeida

    Plausibly, the value of human life is of the latter kind. We do not necessarily have reason to bring additional people into existence.
    I don’t contend that we ought to bring human life into existence; why would anyone claim that? I don’t even claim that we ought to bring normal adult humans into existence. I deny that there is such a thing as pre-conception potentiality of a human being. There cannot be a pre-conception potentiality of a human being since there exists no “peconception human being” to have the potentiality. But even if we could make sense of a potentiality to be a human being, transitivity principles for potentialities are false. If X has the potential to be Y and Y has the potential to be Z it is not in general true that X has the potential to be Z. So if something is potentially a fetus, and a fetus is potentially extremely valuable, we cannot conclude that X is potentially extremely valuable. What is potentially extremely valuable is something that X potentially is, not something X actually is. X is at best potentially potentially valuable.

    December 10, 2009 — 11:36
  • r.chappell

    I understand your view — that’s why I referenced pre-conception “potentiality” in scare-quotes. So long as we can contrast particularized potentiality with generic opportunities to create value, the word we use for the latter obviously doesn’t matter. (Did you read the rest of my comment?) My point is that potentiality is morally irrelevant in the cases you discuss: it does no work over and above establishing that one has the opportunity to promote a value (preventing cancer).
    It is just as morally important to prevent one from developing cancer in general as it is to prevent a particular tumor from developing its cancerous potential. So if you really endorse the analogy, you should be concluding that it’s just as morally important to bring more people into existence in general, as it is to enable a particular embryo to develop its potential.

    December 10, 2009 — 11:57
  • r.chappell

    At risk of labouring the point, let’s explicitly consider an analog of (1) that involves no de re potentiality. Suppose Smith has no tumor, but if we don’t phi then he will acquire a cancerous tumor in his abdomen, as in your original case.
    Do you think that I have introduced a morally significant difference into the case? It seems not. Our reason to act, in either case, is just to prevent Smith from getting cancer. The fact that in one case the possible harm or benefit stems from a particular entity’s “potential” is neither here nor there. So the case presents no reason to think that potentiality as such is morally significant. At most, it merely suggests the general consequentialist recommendation to promote benefits and prevent harms. That is, the case supports acting to bring additional people into existence just as much as it supports acting to realize the potential of a particular embryo. If we don’t think that the analogy really supports the former, then we shouldn’t think that it really supports the latter either. Right?

    December 10, 2009 — 12:08
  • Mike Almeida

    At risk of labouring the point, let’s explicitly consider an analog of (1) that involves no de re potentiality. Suppose Smith has no tumor, but if we don’t i then he will acquire a cancerous tumor in his abdomen, as in your original case.Do you think that I have introduced a morally significant difference into the case?
    Yes it does. Now more labor. It’s true that in this case we have reason to act in certain ways. But your case is disanalogous to the case of the fetus which has a potential to be a valuable being (I know you know that). I am denying that bringing about good states of affairs is in general a reason to act. Otherwise we’d have reason to bring about that state of affairs of there existing normal, adult human beings. But we don’t. We do have a reason to help Smith (in your case and in mine). So that reason cannot appeal to some principle requiring that we promote good states of affairs and prevent bad ones. I claim that the reason in the case I describe appeals to the potentiality of the tumor (and the potentiality of the fetus). I don’t deny we sometimes have reasons to bring about good states of affairs and to prevent bad ones.

    December 10, 2009 — 12:35
  • r.chappell

    I claim that the reason in the case I describe appeals to the potentiality of the tumor
    But why think that? In the absence of any other explanation, you might fall back on this feature as a possible explanation. But I showed that it’s redundant: You agree that in my case we clearly have reason to prevent Smith from getting cancer. But whatever the source of this reason in my case, it will presumably also suffice to explain our reason to prevent Smith from getting cancer in your case. So there’s no need to appeal to potentiality — it does no essential work in explaining why we ought to prevent Smith’s tumor from turning cancerous. We ought to prevent this because we ought to prevent Smith from suffering the harms of cancer (whether the present risk stems from a particular tumor or not).

    December 10, 2009 — 13:25
  • Mike Almeida

    But why think that? In the absence of any other explanation, you might fall back on this feature as a possible explanation. But I showed that it’s redundant
    Richard, what you said did not show redundancy. What you showed was that there is a dissimilar cases in which I still have reason to do something to help Smith. This is like saying the fact that Smith’s foot is gangrenous really gives me no reason to sever it, since Smith might instead have had the potential for a gangrenous foot which would alone give me a reason to sever it. So it’s being gangrenous really gives me no reason at all to sever it. That argument obviously does not show redundancy.

    December 10, 2009 — 14:22