Tim Sommers asks at 3Quarks if you can have an obligation to your past self? The answer is- sure. Why not? The problem is you have no remedy if you decide not to honour such obligations. Still, going forward you can pay a Mafia dude to enforce obligations to your present self against your future self.
A different approach would be to assert that ethical beings have 'genidentity'. They generate themselves through time in an ethical manner such that they have an existential authenticity based on keeping faith with themselves. There are two problems here
1) is purely mathematical. The fact is, only certain 'eigenstates' of a quantized dynamic system have a determinable parameter. This is why 'genidentity' could not resolve problems in quantum physics. The 'holistic' Gestalt of the object is only graspable in a global, non-local, manner. Of course, a non quantized dynamic system would not necessarily have this problem. But it would have another- St. Augustine's 'vanishing present'. That sounds more like spiritual theology than scientific theory.
2) something superior to obligation and entitlement arises with respect to what you were and what you will be. There is a type of 'self ownership', or conatus, which endows the self in the moment with a superior Hohfeldian immunity with respect to all past and future selves- including those of the other people you might have become or who might have become you.
Sadly, having 'genidentity' would be a serious handicap in any competitive field. Things which don't have 'survival value', subtract from 'conatus'. To remain yourself, you need to stop doing stupid shit and start imitating what smart people are doing. Tardean mimetics is 'regret minimizing'. You'll feel pretty darn stupid if you remained 'authentic' and starved while everybody else does sensible things and enjoys affluence and security.
One might say, for some immediate purposes it is better to see yourself as a a member of a statistical class and behave 'ergodically' (either mimetically or in response to Schelling focal solutions to coordination games). The alternative is to surrender to 'hysteresis' and do stupid sub-optimal shite so as to remain 'authentic' or some such retarded shite.
The same point can be made of those who go to Grad Skool but choose shite subjects. It may be a promise they are keeping to their younger, stupider, selves, but this does not change the fact that they are authentically cretinous- unless, of course, they are just phoning in shite of the following sort-
The question, as philosopher Ben Bradley puts it is, “If you desired something in the past but you don’t desire it anymore, do you have any prudential reason to bring it about?”
You may have a 'regret minimizing reason'. However, it may counsel against prudence. You may regret not throwing caution to the winds on certain occasions. This is why though 'prudence' is required of a Financial accountant, Regret-minimization is not. The story is different for Fund Managers, a particular Investment Strategy may be 'Hanan consistent' and yet be highly risky.
Some philosophers – Bradley focuses on Richard Pettigrew and Dale Dorsey – argue that you do. Dorsey’s view is that “[in a case where my future self ignores my current projects] my future self has failed me—failed to do something that my future self had reason to do given the effort I’ve put in.”
This is foolish. You future self can't be said to ignore something it can remember about its past. But it is entitled to refuse to do stupid shit you once thought cool.
According to Dorsey, Bradley says, “I now want my future self to carry out my current plans; so, it would be unfair of me to deny my past self the same courtesy. Expecting my future self to give some weight to what I now want, while denying my past self that courtesy, would be, in a way, inconsistent or ‘unsavory.’”
You should not want your future self to carry out your current plans. Plans must be changed in the light of unanticipated developments or fresh information. Perhaps what is meant is 'I want my future self to have the same commitment or objective that I currently do'.
Still, it is good to know that some people feel it would be unsavoury not to try to suckle on their aged mother's titty just because it is what they wanted to do when they were a baby. It must be said, courtesy can be taken too far.
In other words, we might be tempted to think about prudence (using reason to guide our actions) “as involving some sort of cooperation between one’s past, present, and future selves. When I make decisions, I should think of my past and future selves as other people. Just as I must take other people’s welfare into account in order to act morally, I must take my other temporal selves into account in order to act prudently.”
This is not the case. Regret minimization is one thing- the notion of regret is psychological. That is what is at work here. Prudence is associated with foresight and the sagacity to understand that things might not turn out as you fondly imagine they will. The difference between prudential considerations and regret-minimizing considerations in this case is that the former has to be sceptical about your notions of what you owe to 'past' or 'future' selves. Regret-minimization is more loosey-goosey. I think the advantage is it permits more low risk 'discovery' which is why it is ubiquitous as an evolutionarily stable strategy.
In his fascinating new paper, “The Sacrificer’s Dilemma,” Bradley offers a unique argument against the view that we should “think of prudence as involving a negotiation or compromise between distinct temporal selves, each of which makes claims of justice on the other selves.” It’s this argument I want to focus on.
Hopefully, Bradley's next paper will argue against the view that people who start negotiating with their future and past selves aren't either crazy or stuck in a deeply stupid profession.
But first, there’s a well-known counterexample to the idea that the desires of our past selves might matter at all to us: Brandt’s Roller-Coaster. Suppose as a child you fervently wished to ride a roller-coaster on your fiftieth birthday. Now, it’s your fiftieth birthday, your joints ache and you have a bad back. You are not interested in riding a jerky coaster. Most people have the intuition that having wished to do so as a child, given that you don’t want to now, is no reason at all to ride the coaster.
At least not till you've gone and sucked the titty of your eighty year old mother because that was what you wanted to do on your actual day of birth.
Pettigrew and Dorsey respond that, while not every past desire – this one, for example – creates an obligation, some do.
A binding obligation is a vow or oath or a contractual or sacramental or other such grave undertaking. Speaking generally, there are reputable people you can consult on the circumstances in which such obligations are defeasible.
But Bradley uses time travel as a heuristic to get at how this view goes wrong.
But, if we had a time travel machine we wouldn't bother with stupid shit of this sort.
Suppose the most painful sacrifice you made for yo-yos was persisting in your work on the yo-yo museum despite your partner, first, threatening, and then actually, leaving you. You told yourself at the time that the sacrifice would be worth it – one day you would be a part of yo-yo history. Now that you don’t care about yo-yos you feel bad about the sacrifices made by your former self, but you still don’t want anything to do with yo-yos. Luckily, since you have access to a time machine, you can go back to the past and make sure that either your partner does not leave your past self or that you can compensate your past self in some other way. But wait! If the partner never left, because you prevented it via time travel, then your past self didn’t suffer the suffering that you were motivated to go back to alleviate. “We have a reason to travel back if and only we don’t do it,” Bradley says.
This is silly. We don't need a reason to do something novel- like revisit the Eighties or go ten thousand years into the future to see if dolphins have taken over the world.
But if you do it, you have no reason to have done it. Time travel always ends in paradox.
There is no paradox here. It's cool to visit the past or the future even if you gain nothing by it.
So, is this just another time travel paradox? Maybe not.
Bradley argues that we can get stuck with “some of the same kinds of paradoxes and dilemmas that genuine time travel or backwards causation would generate” in these cases even without the time machine.
Only if we are as stupid as shit or are condemned to teaching nonsense to retards.
One might try to capture the central claim of the ‘you owe your past self’ view with something like Pettigrew’s “Beneficiary Principle: A current self that has justly benefitted from certain sorts of sacrifice made by some of its past selves has an obligation to give a certain amount of weight to the preferences of these past selves.” My former self sacrificed an important relationship to be on the stage and yo-yoing at the grand opening (and more!), I now have some obligation to either follow through or compensate my past self in some way.
One can certainly speak of things you owe yourself. But this is merely a metaphorical way of speaking. It doesn't matter if you genuinely made sacrifices or not. The thing is not justiciable in any real sense. It is merely ipse dixit. You are judge and jury in your own case. The problem is that you may also be the sexy court stenographer. Next thing which happens is you are arrested for outraging public decency. Worse still, it turns out you yourself are the arresting officer. Shit. Maybe going off your meds wasn't such a good idea.
For this to make sense in the case where we don’t have a time machine, we must accept something like “The Redeeming the Past Principle (RPP): someone’s welfare at [a time] can be affected by things happening after [that time].” Dorsey explicitly endorses that principle. “If I value now climbing Mount Everest in [the future], and I do it, that I do so makes me better off now.”
So does valuing valuing climbing that mountain as does valuing that valuing and so on. Equally, not valuing infinite regresses of valuing may be said to make you better off but only in the sense that every goldfish is the cat's whiskers.
So, here’s Bradley’s argument. Even if sacrifices in the past mean your past selves deserve compensation,
an investment is not a sacrifice. Going to Medical School may involve giving up many of the pleasures your peers enjoy to the hilt. But you believe you will be much better off than them. True, you may be stricken down by mental illness or suddenly suffer some sort of spiritual crisis. But the odds are in your favour.
and there are ways to compensate your past self in some cases (since someone’s welfare can be affected by things that happen later).
It is true that a guy going to Med School is treated with more respect by other people. Expectations do affect reality and you may be extended credit which is just as good as cash. But this has nothing to do with actual past or future selves.
To the extent that you compensate your past selves
which is the same extent to which every goldfish is the cat's whiskers
(time machine or not), you necessarily also take away any reason to compensate them.
But not the reasons why that extent may not suffice.
RPP allows you to change the past, in a way. But just like changing the past with a time machine, if you do it, you give your present and future self no reason to do it or have done it. If you do it, you have no reason to do it.
But there may be a reason you haven't done enough of it. The other problem is that once you admit liability in a particular case, you may find yourself on the hook for a large class of similar claims. This may involve your trying to suckle on your very elderly mum's titties just the way you planned to do when you were born.
“This conclusion is unacceptable; it cannot be the case that someone has reason to do something if and only if they do not do it.”
I have a reason to take a piss precisely because I am not currently pissing. What's wrong with that?
But here’s an objection to that argument. If I am thirsty and I drink a glass of water, I no longer have a reason to drink a glass of water. How is that different from the Bradley case? I had a reason to do something until I did it and now, having done it, I have no reason to do it. Isn’t that a case of someone having a reason to do something if and only if they haven’t done it?
Yes.
In the water case, first, at a specific time that we will call t, I am thirsty; then, at time t+1, I drink water; until finally, at time t+2, I am no longer thirsty. There’s no time at which I have a reason to drink the water only if I don’t drink it.
Yes there is. At time t you had a reason to drink water because you weren't drinking water.
I always have or had a reason to drink at time t. But in the RPP case the moment at which I act to change the past is itself one of the moments in which I have no reason to act if I do.
You don't act to change the past. Still, the RPP nutter can point out that you didn't stab yourself so as to frustrate your own thirst. Surely, this is at least partial redemption?
The kind of thing that I can do to affect the welfare of my past self is to now honor a preference of theirs that I no longer share. That doesn’t look paradoxical.
It is arbitrary or ipse dixit. Anyone is welcome to say they are honouring their past life as the Queen Victoria by shoving a radish up their bum and running naked around the Albert Memorial. Indeed, if you studied Chemical Engineering at Imperial College, I believe the thing was de rigueur.
I think Bradley’s argument, again, is that changing the past – even without retrocausality – creates a paradox because it still makes the past different in a way, say drinking water, does not. Once I fulfill my past self’s’ desire, there is a sense in which it was always fulfilled (because it was always in the past). If I can compensate myself in the past, even if just in this quasi-logical way, I am eliminating what would give me a reason to change it.
Sadly this is not the case. Elimination a reason to do a thing does not eliminate a reason for thinking that reason hasn't been fully extinguished. I may admit I have a reason to do the washing up. I may actually do the washing up. But I still have a reason to do the washing up because apparently washing up involves using soap and a scrubber and then stacking dishes in the dishwasher. Fuck that.
Here’s where Bradley’s titular “sacrifice” language helps. My former self sacrificed a relationship to make the yo-yo venue happen, that’s why I owe it to them to fulfill their desire to go inside and yo-yo. But if I fulfill that desire at any point, what my past self did is no longer a sacrifice,
because it is a radish up your bum?
and so I have no reason to compensate them for a sacrifice they ended up not making.
You may do or you may not. The thing is purely arbitrary.
Bradley says that “the agent has reason at t to bring it about that P if and only if the agent does not at t bring it about that P.”
Nothing wrong with that. You can have antagonomic reasons and only do things you have no reason to do except maybe you don't for that same reason. The fact that a lot of reasons are useless, silly or not action guiding in any way doesn't change the fact that they are reasons though, obviously, there's no reason you can't call them the goldfish that is the cat's whiskers.
You have a reason, hypothetically, to fulfill your past selves preference in this case, but if you do, you no longer have a reason.
We don't know that.
The point is that “Paying back one’s past self is not a good way to think about the reasons we have, if any, to care about our past values,” Bradley says. That seems right.
It is arbitrary. It may be right in some situations. Gassing on about what you owe your past self alerts your interlocutors to your solipsistic narcissism. This lowers their expectations of what they can get out of you.
There’s a lot more there detail-wise. And philosophy is all in the details. But let’s hope we got this part straight. I will let you know when a full version of the paper is available. A few final comments.
(1) Recall the “Beneficiary Principle.” A current self that has justly benefitted from certain sorts of sacrifice made by some of its past selves has an obligation to give a certain amount of weight to the preferences of those past selves.
But those past selves have obligations to each other and obligations to you which they didn't meet and so 'netting out' is an infinite process because of impredicativity. Thus, this is a line of reasoning which can't be action guiding.
What kind of obligation?
An unquantifiable or wholly arbitrary one.
Do the preferences of past selves have moral or prudential weight? Any given past self is either you, or it isn’t you. If it isn’t you, you could only have a moral, and not a prudential, obligation. If it is you, then why shouldn’t you always do what you want now, and not what you wanted in the past – but no longer want. Derek Parfit, the most influential precursor to the kind of view Dorsey and Pettigrew take, intentionally blurred the line between morality and prudence – our obligations to our own future and past selves are very much like our obligations to others.
Not if those obligations are legal or otherwise enforceable against ourselves.
But I wonder what happens to the Dorsey/Pettigrew position if the obligation to past selves are not prudential. If you’ve changed enough to be a new person, it seems to me, you have no prudential obligations to a former selves.
Why not? You may have become a worse person. It might be safer to stick with your original plan.
If you have a moral obligation, why is stronger than your obligations to anyone?
For some ideographic reason.
Either way, this is not a good model of prudence.
It is not prudent to have a model for prudence though, if you don't have a good Structural Causal Model, you have to content yourself with doing what the guy with a reputation for prudence is doing.
(2) Finally, under all the abstractions, it seems to me, the view that you can owe it to your past self to do things that you no longer want to do because your past self put so much effort into it, hides some very bad advice about prudence. Saying “my future self has failed me—failed to do something that my future self had reason to do given the effort I’ve put in,” as Dorsey does, seems to embed the fallacy of sunk cost into our very conception of prudence.
The exception to sunk cost is if it is 'discovery'. You need a model with Knightian Uncertainty.
(3) Since I wrote (2), Bradley pointed out to me that Dorsey and Pettigrew would likely simply deny that “honoring sunk costs” is a fallacy, as some other recent philosophers also have (e.g., Doody and Kelly). This doesn’t seem promising to me. For example, Doody’s defense of following sunk costs as a rational way to act is “so that a plausible story can be told about you according to which you haven’t suffered.”
But you can always tell a plausible story about how doing stupid shit and ending up being sodomized in a South African prison gave you the dynamite idea for a new crypto-currency which is guaranteed to give early investors a ten thousand percent return.
A joke from my graduate school days seems apropos here. That’s not a counterexample to my argument, that is my argument. But I have gone on too long.
The joke was that he went to graduate school. The point about pointless arguments is that making them in some contexts has greater imperative force than shitting into the palms of your hand and flinging your faeces around.
No comments:
Post a Comment