Does it make sense to say that someone would have been better off, had he never been born?
Jim was born with a disease that has throughout the years
caused him intense agony, and for which he has no hope of a cure. Jim has no friends because the pain makes him
extremely aggressive. In fact, he
behaves so aggressively that not even his family can stand him. He has no opportunity to achieve important
goals, such as acquiring interesting knowledge or developing an artistic
talent. Jim lives a life that is not
worth living. Jim’s life is a life that
is bad for him
However, despite the intuitiveness of the claim that Jim
would have been better off if he had never been born, there is a problem. Had Jim never been born, he wouldn’t have
existed at all. And it isn’t clear that
it even makes sense to say that Jim would be better off than he is now in a
situation in which he doesn’t exist. We have a puzzle. In this post, I attempt to solve it.
Bernard Williams writes that while
someone “can thinking egoistically of what it would be for him to live longer
or less long, he cannot think egoistically of what it would be for him never to
have existed at all” [Williams 1973]. In
general, it seems that you cannot meaningfully compare the degree to which an
entity might have had a particular property with the degree to which it
actually has that property – unless
the entity exists in both the counterfactual and the actual situations.
For another
illustration of this principle, consider Kenny. Kenny is extremely ugly. His
limbs are deformed, his face is a wreck, and he has large open sores on his
forearms. Although Kenny is extremely
ugly, it doesn’t make sense to say that Kenny would have been prettier had he
never been born. Assume for a minute
that there are facts about how attractive people are, that it makes sense to
talk of degrees of attractiveness, and that beauty and ugliness can be placed
on the same scale. In order for someone s to be prettier in situation A than in situation B, s must have some
amount of attractiveness in A that is
comparable to the amount of attractiveness that s has in B. But if s
doesn’t exist in A, then s has no amount of attractiveness; s does not have zero amounts of attractiveness in a situation in which s does not exist.
Likewise,
in order for someone s to be better off in situation A than in situation B, it seems that s must
enjoy more welfare in A than in B. But this seems to require that s exists
in both A and B, since it seems that one must exist in order to enjoy some amount
of welfare. So it doesn’t seem to make
sense to say that someone would be better off not existing.
In some circumstances, people are
better off in possible situations in which they do not exist. What we need is a reasonable theory that
tells us how this is possible. In what
follows, I sketch such a theory. Here is
a preview: the theory that I endorse holds that (1) the primary objects of
prudential evaluations are the lives that person’s lead and (2) person’s have
lives even in possible worlds in which they do not exist. This initially sounds outrageous, but,
fortunately, each element of the solution is defensible (and has been defended).
The solution makes use of the following claims:
(LIVES): A person’s life at a world w is the class of
propositions about that person that are true at w.
(BASICS): The basic bearers of intrinsic value are fine-grained propositions; classes of propositions have intrinsic value in virtue of containing such propositions.
(BETTER): A’s
obtaining would be better for s than B’s obtaining iff the life that s would have if A obtained is intrinsically better than the life that s would have if B obtained.
(NULL): If a life
contains nothing that has positive or negative intrinsic value, then the
intrinsic value of that life is zero.
I will argue that persons have
lives even at possible worlds in which they do not exist. Let us consider
Jim. Since Jim is a human being, he is a
contingent being; he could have failed to exist. Since Jim could have failed to exist, there
is a possible world w at which he
does not exist. Since Jim does not exist
at w, the proposition that Jim does
not exist is true at w. So propositions about Jim can be true at worlds at which Jim does not
exist. In general, propositions about
actual objects can be true at worlds at which they do not exist.
Given this
fact, we are in a position to exploit an interesting feature of LIVES. Since a person’s life at a possible world is
simply the set of propositions about him or her that are true at that world,
every actual person has a life at every possible world. Even if Jim does not exist at w, he has a “life” at w, i.e., there is a set of propositions
about Jim that are true at w.
This set is
not the empty set. It has at least one
member: the proposition that Jim does not exist. And presumably it has many other members as
well, e.g., the proposition that Jim does not exist and 2+2=4. Although it seems strange to call this set
“Jim’s life at w” since Jim is not
alive at this world, it is an appropriate object of axiological
evaluation. Jim’s life at w, whether we are comfortable calling it
that or not, has an intrinsic value. Presumably, the intrinsic value of Jim’s life at a world at which he
does not exist is zero. It is hard to
see how there could be anything intrinsically good or bad about Jim’s life at a
world in which he does not exist. Standard hedonism, for example, implies that Jim’s life at a world at
which he does not exist has an intrinsic value of zero, since it does not
contain any propositions of the form “Jim feels pleasure (or pain) to degree n from t1 until t2.” Any other reasonable axiological theory, such
as desire satisfactionism, perfectionism, or various versions of the objective
list theory, will have similar consequences. So unless sheer non-existence (or existence) is intrinsically valuable
(bad or good), then Jim’s life at a world in which he does not exist has an
intrinsic value of zero.
Suppose that things are going bad
for Jim, and that there is no hope of them getting better. In
the actual world, Jim has a bad life; his life is in itself bad for him. It seems that Jim would have been better off
if he had never existed. BETTER implies
that Jim would be better off not existing if and only if the life that Jim
would have had if he hadn’t existed is intrinsically better than the life he
has now. In order to evaluate whether
this is the case, we go to the nearest possible world in which Jim does not
exist and see what the intrinsic value of his life at that world is. At that world, Jim’s life has nothing in it
that is intrinsically good or intrinsically bad. NULL implies that the intrinsic value of
Jim’s life at the nearest possible world at which he does not exist is
zero. Since Jim’s life is quite bad, the
intrinsic value of his life at the actual world is less than zero. So the life that Jim would have had if he had
never existed is better than the life that Jim actually has. So Jim would have been better off if he had
never been born.
Cool.
The solution as stated doesn’t seem to me to work, and I’m not sure whether the problem is incidental or essential (and it seems a lot easier for me just to say what it is than to think through the ramifications carefully).
Since you give NULL as a premise, I expect you to argue that the class of statements true about a person at a world in which the person does not exist contains nothing that has positive or negative intrinsic value. But I don’t see any argument for that claim. Instead, you say:
And you give the example of hedonism.
Now, it is not at all obvious to me that there is nothing intrinsically good or bad about any of the propositions in the Jim class in a world at which Jim doesn’t exist. I can’t remember exactly how you decide whether a proposition is about Jim or not, but from the Jim does not exist and 2+2=4 example I infer that Jim does not exist and a million human beings are blissfully happy and get everything they have ever wanted would be about Jim. That seems like a reasonable candidate for a proposition with intrinsic positive value.
What’s gone wrong? Do we need a different “about Jim” criterion? As a general rule it is shockingly difficult to say what it takes for a proposition to be ‘about X’. (Cf. D. Lewis, “Statements partly about observation”)
On the other hand, I like the idea of thinking of something that exists in all the worlds, which an existing person then has, as a device for making sense out of my non-existence being better or worse for me.
I’ll try to think more about these prospects myself.
Jamie,
To get around that problem, I think Kris just needs to distinguish regular intrinsic value from “basic” intrinsic value. States with basic intrinsic value, not regular intrinsic value, are the ones whose values you add up to get the values of complex things like lives. Complex things have intrinsic value, but not basic intrinsic value. The states with basic intrinsic value according to hedonism are bare-bones attributions of pleasure or pain to people at times. *Jim does not exist and a million human beings are blissfully happy and get everything they have ever wanted* has intrinsic value, but not basic intrinsic value; for one thing it’s a conjunction, and no conjunctions have basic intrinsic value according to hedonism. (If they did, pleasures would get massively overcounted by being parts of millions of conjunctive states.) So what Kris should have said is something like: “it is hard to see how Jim’s life could contain any states with basic intrinsic value at a world where he does not exist.”
Consider a world where there are lives with no individuals that have them. In this world, is it true that the fact that these lives are not lived, that they are deprived of agents, is something bad for them? If so, it would “seem” to follow that this world is worse than we might intuitively think. All the lives that are not had, lives that exist in the world without their bearers, will contribute badness to the world we don’t intuitively recognize.
Perhaps only individuals with life can be deprived of life, but that isn’t obvious initially.
But the solution isn’t supposed to presuppose hedonism.
More generally, I have these questions, with my own tentative answers in parentheses:
I should come up with my own suggestion — I’ll try to do that. This question has actually baffled me for a long time. I’m kind of glad to see someone else worrying about it.
I think your solution is pretty cool, Dr. McX.
Another way to go might be to understand welfare — i.e., the “good for” relation — as a relation not between a life and the person who lives it but between a *world* and a person. If hedonism is true, the value of some W for S = the amount of pleasure minus pain S experiences in W. If some OLT is true, then that value = the amount of knolwedge, virtue, aesthetic appreication (or whatever) the person gets there minus whatever bads he suffers. If S doesn’t exist at W, then it turns out that the amount of pleasure minus pain S experiences at W is zero. The same would go for an OLT. So we can get it to be that it could be better never to have existed.
This avoids having to say that you have a life even at worlds at which you don’t exist.
Is this solution cheating in some way?
I’d like to try to push Jamie’s point a bit further, if you excuse me, because I too wonder why it would follow from your account of lives that Jim’s life at w is neutral in terms of value.
Here’s a first list of propositions about Jim that are true in w (list 1):
‘Jim does not feel pain’
‘Jim does not suffer’
‘Jim is not discriminated against’
‘Jim is not tortured’
‘Jim is not forbidden to be speak against the establishment or to vote’
‘Jim is not lied to consistently’
‘Jim is not dispised’
and so on – you get the picture hopefully.
Here’s another list of propositions about Jim that are true in w (list 2):
‘Jim does not have friends or family’
‘Jim does not experience joy’
‘Jim does not have an access to education’
‘Jim does not have ground projects’
‘Jim does not smell the flowers’
and so on.
All these propositions about Jim are true in w so it seems like on your view Jim’s life in w would consist of them. They are also propositions that are bearers of intrinsic value on many if not most plausible accounts of intrinsic value. Items on the list 1 are the kind of propositions that the actual Jim would very much desire to be true and items on the list 2 things that Jim would desire not to be true. So on the desire satisfaction view these propositions would bear positive and negative value, and thus if Jim’s life at w consists of them, then his life there is going to end up having positive or negative value depending on the sums. And, many would put items like the ones on list 1 to objective list theories as well.
You might of course accept this consequence (if it follows), but for me it just sounds like a reductio that Jim has a life at w which is more or less valuable even if he doesn’t exist there. It also sounds like a reductio to me that I have infinitely many lives as there probably are propositions that are true about me in infinitely many possible worlds. I find it hard to find any grounds that would make me want to think that *I* have more than one life.
Hi everyone,
Thanks for the comments. Some responses, in no particular order:
I don’t want to change the account of lives, or committ myself to hedonism. (Which, I guess, I reject. I think I’m some sort of wishy-washy, touchy-feely, lovey-dovey guy when it comes to what makes a life worth living.)
I agree with Ben that an appeal to the notion of a basic bearer of intrinsic value is key.
A life, world, or total consequence of an action has ivalue in virtue of containing basic bearers of intrinsic value. This claim belongs to “formal” axiology. A substantive, material axiological theory will determine what these basic bearers of value are, and the functions that take us from the values of the basics to the values of the things that contain them.
Here’s the revised list of principles I want to endorse:
(LIVES): A person’s life at a world w is the class of propositions about that person that are true at w.
(BASICS): The basic bearers of intrinsic value are fine-grained propositions; classes of propositions have intrinsic value in virtue of containing such propositions.
(BETTER): A’s obtaining would be better for s than B’s obtaining iff the life that s would have if A obtained is intrinsically better than the life that s would have if B obtained.
(NULL*): If a life contains nothing that has positive or negative basic intrinsic value, then the intrinsic value of that life is zero.
(EMPTY): If S does not exist at w, then S’s life at w contains nothing that has basic intrinsic value.
I think that by replacing NULL with NULL*, and making explicit my substantive assumption about what is valuable, i.e., making EMPTY explicity, I can make the proposed solution clearer.
Although EMPTY is a substantive material axiological claim, it’s consistent with a number of plausible axiological views — in fact, I would think it’s consistent with every reasonable axiology of which I am aware. Hedonist’s, who think the form of a basic bearer is:
S feels pleasure (pain) to degree n at time t for duration d
will accept EMPTY. Desire satisfactionists,, who think the form of a basic bearer is:
S desires P at t to strength n for duration d and P is true (P is not true)
will accept EMPTY. Perfectionists, who think the form of a basic bearer is:
S exhibits/instantiates perfection (imperfection) P to degree n at time t
will accept EMPTY. And pluralists (like me),who think that there are a plurality of kinds of basic bearers of intrinsic value can accept EMPTY as well.
In order to reject EMPTY, you would have to think that something like the items on this list are *basic* bearers:
‘Jim does not feel pain’
‘Jim does not suffer’
‘Jim is not discriminated against’
‘Jim is not tortured’
‘Jim is not forbidden to be speak against the establishment or to vote’
‘Jim is not lied to consistently’
‘Jim is not dispised’
And I doubt anyone holds that these are actually basic bearers of value.
But maybe I’m wrong about that. Maybe some of these are basic bearers. Then there are some things about your life at worlds at which you don’t exist that are good. This wouldn’t show that my proposed solution is unworkable. The lesson I would draw (if I were to be convinced that some of these were basics) is that it is easier to have a life not worth living than I might have first thought.
I have to admit that I find this example from Jamie somewhat troubling:
“What about that Jim’s sister is blissfully happy? Isn’t that (a) about Jim and (b) intrinsically valuable and (c) basic?”
Here’s what I want to say about this example. There are two propositions one might have in mind, one of which is a basic intrinsic value state, and one of which is about Jim. Neither of them is both.
The first proposition is this:
That person (Jim’s sister) is blissfully happy at t to degree n for duration d.
This propositon is a basic, is about Jim’s sister, but is not about Jim. It’s logical form is:
Hxtnd
There’s another proposition you might have had in mind:
There is a unique individual who is Jim’s sister, and that unique individual is happy at t to degree n for duration d.
I would think that this proposition is about Jim, but arguably it isn’t a basic bearer of value.
Heathwood: I have no deep objection to this response to this puzzle. The nice thing about the solution I offer is that it fits in with a systematic, formal theory of how value works. You don’t have something analogous, but I don’t doubt that you could work something out. Maybe they would be merely notational equivalents? I really don’t know.
A quick response to Jussi:
“All these propositions about Jim are true in w so it seems like on your view Jim’s life in w would consist of them. They are also propositions that are bearers of intrinsic value on many if not most plausible accounts of intrinsic value. Items on the list 1 are the kind of propositions that the actual Jim would very much desire to be true and items on the list 2 things that Jim would desire not to be true. So on the desire satisfaction view these propositions would bear positive and negative value, and thus if Jim’s life at w consists of them, then his life there is going to end up having positive or negative value depending on the sums. And, many would put items like the ones on list 1 to objective list theories as well.”
I think that this is exactly what the desire satisfactionist should not say.
The desire satisfactionist, in my opinion, should identify the basic bearers of value with propositions like this:
S desires P to degree n at time t for duration d and P is true
S desires P to degree n at time t for duration d and P is false.
These propositions are the ones that have basic value, not the propositions that are the *objects* of the propositional attitudes.
I think Heathwood has a paper on this point. Can you share the link (love), Wood?
Well – those are already second-order propositions that have other propositions as their content and they are conjunctive. Hard to see that they satisfy BASICS thesis. That would threaten to make desire satisfaction view formulatable in your framework.
Hm.
Could you tell us what ‘basic intrinsic value’ means? According to Ben it can only be had by simple things; but you attribute it to conjunctions. I have no view about this — I figure ‘basic intrinsic value’ is a technical term.
I was thinking of a proposition with this form, by the way:
Hf(Jim)tnd
where the ‘f‘ is a functor. I think that proposition is intuitively about Jim, but of course you don’t especially need to stick to the intuitive notion.
Jussi:
I want desire satisfacitonism to be formulatable, so I don’t feel threatened by this. I’m ok with the basic bearers of ivalue being second-order propositions that are conjunctions. To be a basic bearer is not necessarily to be an “atomic” proposition.
More on this soon, by way of answering Jamie’s question…
I just can’t stop here. This is getting way too interesting. And, sorry if I am getting a bit off-topic – that’s the way philosophical discussions usually go. But, I just find the question what kind of propositions desire satisfactionists should think to be the bearers of value very interesting.
I haven’t read Chris’ paper yet so probably there is a good answer to this question. I always thought that the bearers of value on this view would be the basic states of affairs (or propositions) which are desired, and then the desires are satisfied and this confers value to the original states of affairs. I thought that the virtue of this view was that it had explanatory potential. The view can answer the question why those and those states of affairs (or true propositions) are valuable by saying that the agent desired them and the desire was satisfied. If the agent hadn’t desired the states of affairs or the desire had not been satisfied, then the states of affairs/true proposition wouldn’t have had value. That looks like a counterfactual claim used in many sorts of explanations, and it seems to work here.
But now, if the basic bearers of value are instead propositions like ‘S desires P and P turns out to be true’ then it is not clear what explains that this proposition bears the value it is the bearer of. At least it is not clear how desire satisfaction theory itself could anymore explain how this proposition comes to bear the value it bears. Either this leads to more robust realism about value than desire satisfactionists usually accept or maybe there is a way around this with higher order desires (regress problem may threaten though)…
Sorry about this. I’m getting side-tracked here. Thanks for patience.
I didn’t mean to suggest that, in general, basics can’t be conjunctions. (I actually do think that might be true, but for complicated reasons.) Only that *hedonists* don’t think the basics are conjunctions. (Unless they are Feldmanians.) I only brought up hedonism because the original example was based on the intuition that it’s good for people to be blissfully happy.
Fascinating discussion here. I would like for Chris’ suggestion to work, but doesn’t it do too much? He suggests we “understand welfare… as a relation not between a life and the person who lives it but between a *world* and a person.” Are there principled grounds for stopping there, or should we also hold that other modes of evaluation (e.g. good looks) hold between worlds and people? But we don’t want to say that an ugly person looks better in a world where they don’t exist. As Kris pointed out, the value there isn’t zero, but incommensurably blank. Shouldn’t we say the same of welfare? And if not, why not?
Hi Kris,
I find this issue so interesting that I really cannot stop thinking about it… Shows how good the original posting is. Another worry is beginning to dawn on me.
Here’s again how you phrase lives:
(LIVES): A person’s life at a world w is the class of propositions about that person that are true at w.
I wonder if a further condition is needed to this definition. What I have in mind is that the relevant life-constituting propositions need not only be (i) *about* that person (say, Jim) and true at the relevant possible world w, but futhermore also propositions about *the person within that world*.
I think that there is good motivation for that requirement. The proposition that ‘Jim actually is suffering’ is about Jim and true in all possible worlds but it’s not clear how this proposition could be substantially part of what constitutes Jim’s life in all possible worlds. The third requirement would stop this objection. The proposition even though true in many worlds is not about Jim in those worlds it’s about Jim in the actual world.
Now, if the third requirement is accepted then it’s not clear if the proposition ‘Jim does not exist’ is about *Jim in w*, even though it is about Jim and true in world w. The proposition just seem to be true in virtue of there being nothing in w that the name Jim picks up, not because of any feature of Jim in that world. If this is the case then, Jim doesn’t get a life in the world he exists in, and for me that sounds right. Lives just sound like dents to me – someone’s got to have them.
Sorry if this wasn’t well enough expressed and there is probably a good reply as always. Good work with the posting and replies.
Hi Jussi,
That’s an interesting worry. Given this framework, there are many propositions that will be elements of lives that initially we might not have thought were elements of lives. For example, “Jim is actually suffering” is an element of Jim’s life, as you point out– and an element of whatever life Jim has at any possible world. Pretty weird, I agree.
I like the framework for largely methodological reasons: it provides a neutral, er, framework for formulating various substantive axiological claims, and as far as I can tell, rules out no substantive axiological claim. I have pretty loosey goosey pluralistically views about what is valuable, but I can imagine someone far more radical than myself, who wants to insist that not only is pain bad, but so is actual pain! That is, the basic bearers of negative intrinsic value have the following forms:
Person p feels pain to degree n at t for duration d.
Person p *actually* feels pain to degree n at t for duration d.
I think this view is implausible; the second set of alleged basics seem funny to me. But someone, I suppose, could hold this view while (consistently) holding the framework. So I don’t want to adopt the move you suggest, independently of whether doing so would muck up my proposed solution to this problem about being better off never existing. (This axiology really is crazy — if things go bad for you in the actual world, that screws you over in other possible worlds too!)
My strategy in general has been this: let as many things into the lives as is possible (so to speak), and let substantive theories of welfare identify which things in those lives are basics.
I know I haven’t clarified the notion of basic bearer. This concept was developed by Fred Feldman. The paper can be found here:
http://www-unix.oit.umass.edu/%7Effeldman/BIV.pdf
The rough idea is that basic bearers of ivalue are the “atoms” of value; they are those bearers of ivalue such that all other bearers of ivalue derive their ivalue from them. That’s the rough idea; Feldman expands this notion in several ways. I’ve been thinking about putting this as a seperate post, so i’d like to come back to this more later…
Kris, thanks for the URL for the Feldman paper. (Why does PEA Soup not allow it to show up as a link in the comments?)
An idea suddenly occurred to me. It’s pretty obvious so there’s probably something wrong with it.
Whatever lives are, let’s just invent a special ‘waste case’ of a life, the null life, or just λ, and say for each person who does not exist at a given world, w, that she has λ at w.
Then we just have to give a place in the ordering of lives to the λ. (To solve the ‘better never to have been born’ puzzle, we don’t even have to give λ a cardinal value, just a place in the ordering.)
Hi Jamie,
Here’s a solution that I believe is relevantly similar to yours. We offer this analysis of “x would have been better off if x had never been born”:
x would have been better off if x had never been born iff the value of x’s life is less than zero.
(If you think it takes more than that, make the right hand side say “sufficiently less than zero.”)
I think that this biconditional is obviously true, and since the notion of a negative welfare is pretty easy to grasp, what’s the puzzle? Why are so many people puzzled?
I could make myself be satisfied with this “cheap solution”, but it seems that many people won’t like resting with this analysis. They think a genuine comparitive claim is being made, and this “analysis” seems to deflate this intuition. The solution that I offer does make use of a genuine comparitive claim to ground the claim that one would have been better off had one not been born, and the solution is rooted in an independently motivated framework.
Another issue is this: we could introduce a similar analysis of “would have been less ugly had he never been born”:
x would have been less ugly had x never been born iff x is more ugly than the norm (or sufficiently ugly or somesuch)
But this analysis won’t satisfy those people who have the intuition that remarks like “would have been less ugly had he never been born” are simply confused or vacously false.
Hang on. My suggestion does make the analysis relational. There is an ordering of lives, and you would have been better off had you never been born iff your life is worse than the λ.
Your cheap suggestion does not make the apparently relational judgment have a truly relational structure. That seems to be a significant difference.
Your complicated suggestion does make the analysis relational. It has this advantage: that it embeds the answer in an independently motivated framework. It has this disadvantage: the framework is metaphysically loaded, and (I think) normatively loaded as well.
What about our simple solutions – what is to be said in favor of one or the other? Well, maybe it depends on whether it is more plausible that relational valuations are explained by monadic ones, or vice versa. More explicitly: is the fact that one life is better (to live) than another explained by its having a certain value that is greater than the value of the other, or does the explanation go the other way around? My view is that it goes the other way around. The facts about which is better come first, and values are derived from and explained by them. If (as I suspect) you hold the contrary view, then that might be a reason for you to prefer the account that has monadic cardinal values at the ground floor.
By the way, I liked the Feldman article, but a lot of its implications seem pretty dubious to me. For example, it seems to me to be a very strong assumption that there are basics whose values can be summed to yield the value of the wholes that contain them. Hedonism does work that way (maybe) but not all coherent axiologies do
Hi Jamie,
“For example, it seems to me to be a very strong assumption that there are basics whose values can be summed to yield the value of the wholes that contain them.”
I don’t think Feldman accepts this; where in the paper did you see this? He definitely accepts a supervenience thesis, but I don’t see a summation thesis explicitly endorsed. Regardless, I agree that building summation into what it is to be a basic is not a good idea.
Hm, maybe I misunderstood and he was just using an additive theory as an example. Supervenience is much weaker.
In “On It Being Better Never to Have Been Born,” there is a quotation from Bernard Williams. From which work is this taken?
I think it’s from “The Makropoulos Case”.
Kris?
Yes, the quote is from “Makropulos.” If you have Fischer’s _Metaphysics of Death_ it’s on p. 79.
Hi guys,
yes, that’s right. Sorry for the slow response– been on the road the last week.
Kris,
Maybe someone said this above. I’m sorry that I missed this post, it is very interesting. You write,
“Since Jim does not exist at w, the proposition that Jim does not exist is true at w. So propositions about Jim can be true at worlds at which Jim does not exist”.
I’m not sure this is right. The proposition that Jim does not exist is not a proposition *about* a nonexisting object (Jim) in that world. (1) is true in w,
1. Jim does not exist.
But in w the name ‘Jim’ does not refer to anything. So, (1) just states (2),
2. ~(Ex)(x = Jim)
But every object that any proposition *could* be about has the property of being self-identical (indeed, has that property necessarily). But what about Jim?
3. Jim = Jim
If (3) were true then, we could existentially generalize to the negation of (2). But obviously we can’t do that. Since everything in w has the property of self-identity, and Jim lacks that property, no propositions in w could be *about* Jim at all. There is nothing for them to be about. Otherwise, you’d have to contend Jim lacks a necessary property of all objects. Difficult to see how that’s possible, unless you’re going Meinongian and contending that there are (exist?) objects that don’t exist.