In her very good article in the latest Ethics (117,
Oct. ’06), “Kantian Rigorism and Mitigating Circumstances,” Tamar Schapiro
brings to light a problem with the standard Kantian line for dealing with the
flexibility of moral rules, in particular the rule against deception.
On the one hand, the standard line holds that deception is
wrong (by Kantian lights) by virtue of facts intrinsic to deception itself—it
is manipulative, interferes with another’s autonomy etc.—which seems to entail
the rigorist conclusion that deception would be wrong no matter what the circumstances. On the other hand, Kantians often say that in some circumstances deception is permissible
(suggesting that it is not merely facts intrinsic to deception itself that make
deception wrong). How can this tension
be resolved?
Schapiro’s resolution has several important details that I’m
going to pass over, hopefully without distorting the picture too much. The basic idea is to understand “the spirit
of honesty” as the spirit “to live up to an (implied) demand to think and act
from a shared standpoint” and to understand honesty itself as “a form of address issued by one
colegislator to another” (50-51). Those
who either can’t be part of this community or won’t play by its rules can be permissibly
deceived. Those with incapacitated
autonomy can’t be part of the community and so can be paternalistically
deceived; those wrongdoers who repudiate the reciprocity of the moral
community, such as the murderer at the door, won’t be part of the community.
honest—in the robust sense—is simply impossible, in which case deception is
permissible.
I like this way of resolving the tension. But I don’t see how it can account for two cases
of intuitively permissible deception.
- Surprise Party: I deceive a loved
one about where we are going tonight, in order to distract her from the fact
that we’re going to her surprise birthday party. (She has never given me reason to think that
she either likes or dislikes surprise parties.) Here, the patient (the loved one) is autonomous and has not disavowed
the moral community. Indeed, she has done nothing that makes (robust) honesty impossible. Instead, it’s the surprise party that requires
deception.
- White Lie: My friend asks if I
like his new haircut, and I deceive him into thinking I do, when really I think
it is hideous. This isn’t one of those
cases where the friend desperately needs an honest answer from a trusted source
(as with, say, “Do you think my wife’s cheating on me?”). The friend sincerely asks in a spirit of
curiosity, and while his feelings would be slightly hurt if he knew the truth,
in the end he wouldn’t care that much. Again, there is nothing about him, voluntary or not, that is
incompatible with the reciprocity of the moral community.
In note 29, Schapiro considers two cases that are similar,
in that they involve patients who neither are non-autonomous nor have
repudiated the community. (1) We want to
deceive Tony in order to stop him from acting on his desire to get a huge rush
by bungee jumping from a “particularly high bridge using a particularly
dangerous form of cord.” (2) We want to
deceive Jack about where Jill is, because Jack will (innocently) annoy her by
talking too much. As with Surprise Party
and White Lie, one might explain these cases’ intuitive permissibility with the anti-Kantian story that
deception is permissible because of the value of the deception’s consequences. However, Surprise Party and White Lie are
different in that, first, the value produced by the deception is pretty small,
and, second, no third party (such as Jill) has anything at stake in the
deception.
Schapiro acknowledges that her account doesn’t
directly address the cases of Tony and Jack, and that they need to be
addressed. So I guess one question here is whether you see any
prospects for a unified Kantian account of the disparate cases where deception
can be rendered permissible, in a way that preserves the claim that deception is intrinsically wrong. Act-consequentialists
secure a unified account of when deception is permissible by tethering
permissibility to the production of good consequences. But this doesn’t preserve the claim that deception
is intrinsically wrong.
Now Kantians can also go down the non-preservation road by saying that deception is permissible just when it is part of a complex
maxim (one that includes reference not only to the act of deception but also to
the circumstances and end of the deception) that is warranted by the CI. While I admire the ambition of Schapiro’s
attempt to preserve the idea that something intrinsic to deception is wrong, I’ve
always been comfortable with the non-preservation approach. So another question is: why treat deception (or any act-type) as wrong in itself, rather than seeing it as wrong only when
coupled with certain circumstances and ends?
Here’s my proposal. Take these two claims:
1. Deception is wrong (by Kantian lights) by virtue of facts intrinsic to deception itself—it is manipulative, interferes with another’s autonomy etc.—which seems to entail the rigorist conclusion that deception would be wrong no matter what the circumstances.
2. In some circumstances deception is permissible (suggesting that it is not merely facts intrinsic to deception itself that make deception wrong).
Now, start from the Kantian idea that deception is an wrong-maker. Understand 1 in a way that deception always makes actions wrong to a degree. Maybe deception is a strong enough wrong-maker so that it almost always is enough to make actions also wrong over-all. But, a Kantian (maybe not Kant himself) can think that there are other wrong-makers and right-makers too. So, in some cases deceptive actions will have other features that are right-makers.
If this right-making outweighs wrong-making, then these deceptive actions will be permissible overall even if deception is making them wrong to a degree. This fits with 2 if we understand permissibility there as a over-all term. In the cases you give it seems that there are all kinds of right-makers to outweight the wrong-making of deception. I guess the point is that if some action has an intrinsic feature that makes the action wrong we don’t have to think that it makes the action wrong overall wrong in all circumstances.
Jussi,
I like that Rossian story, but just to play devil’s advocate: do you think there is anything at all wrong-making in the deception in Surprise Party?
Josh,
I’ve been thinking about that today. Had couple of thoughts about what to say to this. Much seems to depend on the circumstances. Say that she had started to have some expectations about where you are going on the basis of what you told her. In that case the deception seems to be making the action wrong.
Maybe she also hates surprise parties. In that scenario too, the deception seems to be dubious. That there is even such a chance that she will feel humiliated by the party may be enough to create some wrong-making. Maybe one should at least inquire in some sneaky way in advance how the person feels about surprise parties.
But, these are the kinds of wrongs that seem to be outweighed by the sort of tender relationship that includes surprise parties that make your loved one happy. Whatever the right-makers are don’t necessarily have to be any consequentialist sort of values. Many of them can probably be generated within the Kantian framework from the GI (I think Audi does something like this – generate Rossian Prima Facie duties from the GI).
Also, in the Kantian story it seems like deception is wrong because it undermines autonomy or fails to respect it. It’s uncertain whether this happens in the surprise party case. The ‘victim’ is free to leave the party at any point. So, it’s not like in deceiving her you use her as a mere tool to some end of yours.
Finally, maybe the Kantian can go from Rossianism to Dancyanism. It seems like a possible Kantian stance that deceiving is intrinsically wrong (i.e., the wrong-making features are in the act of deceiving itself) when deceiving is wrong even when there are contexts where some considerations disable that wrong-making. Something like this might fit what Schapiro is arguing.
Jussi,
Note that if we say that Surprise Party’s deception is not autonomy-undermining (and in no other way violates Kantian principles), we’d fail to preserve the idea that deception is wrong in itself.
On another note, I’d be curious to see how your last paragraph might be developed: as you understand it, when the wrong-making is disabled, is the disabling due to the context, or is it due to something internal to deception itself? Only the latter would fit with Schapiro’s approach (as far as I can tell, anyway), and I’d be curious to know more about how that might work, because my (undeveloped) intuition is to go with the former: it’s the context that renders the deception in Surprise Party permissible.
And, lastly, to return to an earlier question and pick up on a couple of your earlier points, let’s stipulate that the loved one does not hate surprise parties, and that the only expectation she has formed as a result of my deception is that we will have fun tonight (and that, as a separate matter, we will in fact have fun at the surprise party). Given these stipulations, and again just to play devil’s advocate, is it your — or anyone else’s — intuition that there is still something wrong (at least a little, and perhaps a wrong that is easily overridable) with the deception in Surprise Party?
Josh,
I’m not sure I can reply to all these questions but here is an attempt.
You might be right about the first point. But, I’m not sure that it is a worry necessarily for the Kantian. She could be talking about deception*. This would be a technical notion of deception. These are the acts in which we hide our ends deliberatily from others so that they cannot decide whether to contribute and thus cannot excercise autonomy. It is an internal feature of these actions that they undermine the autonomy of others. Therefore, there is a Kantian story about their intrinsic wrongness. But, maybe the class of actions of deception* is only a subsection of the actions that we ordinarily call deception. This would imply that some actions like the surprise party would not count as deception* even when they are deception. That’s just one way to go though.
About the second point. What seems to be doing a lot of work in Schapiro’s account is the value of honesty which is a quality that enables certain moral relations in the Kingdom of ends. It is that value that seems to explain that the wrong-making is disabled in some contexts. This reminds of Pekka Vayrynen’s account of hedged principles and how values play explanatory roles in determining when the principles are disabled. So, someone might claim that it is the context that does the disabling but it is the value that is essentially related to deception that explains why this happens.
The last one is also a tough one. I’m not sure what my intuition is. Part of the problem is that I’ve never been in a surprise party let alone been given one. If you think that there is nothing wrong with it maybe the Kantian could argue that here the agent is autonomously accepted to be deceived in which case the deception cannot undermine autonomy.
You might think that it would be wrong to a small degree to throw a surprise party to a stranger. But, if we know the person well and that she likes that sort of parties, and so on, then she is in a constant state of agreeing to being mislead about the party she would want to take part in.
Jussi,
I agree that a Kantian could resort to deception*. But Schapiro, given her particular account, can’t do that. On the one hand, her notions of deception and honesty are understood in the robust sense I discussed in the original post, and are in that sense somewhat specific. On the other hand, for her, the idea that acts are intrinsically wrong is a uniquely deontological idea (this is one of the details I skipped over in the OP). So it won’t help her account to just talk about deception*, for every moral theory could make an analogous move. For example, consequentialists can say that that an act is an instance of deception** just when it is a case of deliberate deception that fails to maximize good consequences, and then hold that deception** is wrong due to features internal to deception**.
Another detail: in the original Surprise Party case, it was stipulated that the agent has no reason to think that the loved one likes surprise parties, so the loved one has offered no consent to be misled at some unspecified later date. (Yet that doesn’t seem to make it more likely that there’s something wrong with the deception in Surprise Party.)
Also, thanks for the note about Pekka’s work–I’ll have to go back and take another look at his stuff with this in mind.
I agree with the first point. But, I also wonder what deception without stars are. Do I always deceive with white lies? Maybe but I’m not sure. The Kantian version of deception seems more closer to what is going on in deceptions in general than the consequentialist term.
About the second point. But, now we get back to the point that as we don’t know how the victim is going to react to the party, the risk that she will be appalled and embarrassed might be considered a wrong-maker.