Who’s Which? Which What?

My guess, really just a guess, is that the discussion of role ethics or relational ethics might benefit from some direct attention to a couple of fallacies available for commission—one minor, one major.  I don’t know whether they’re actually committed or directly discussed in the literature.  Possible examples of each can be found in Henry Rosemont’s essay “Rights-Bearing Individuals and Role-Bearing Persons” (in Mary Bockover, ed., Rules, Rituals, and Responsibility: Essays Dedicated to Herbert Fingarette, Open Court 1991, pp. 71-101).  I’ll make that my text.  I don’t understand it.

Similar questions might be raised in connection with Rosemont’s book Against Individualism (Lexington Books 2015), e.g. about the relation between pp. 33-39 and pp. 39-43.


A minor fallacy

In Plato’s Euthydemus, Dionysodorus interrogates Ctesippus and establishes that the latter has a dog who has fathered puppies.  “He is yours and a father; therefore he is your father” (298d).

In his seminal 1956 paper “Good and Evil” (Analysis 17, pp. 32-42), Peter Geach drew new attention to the point that (e.g. in English) for various reasons “is an A B” is often not logically equivalent to “is a B” and “is A.”  To be someone’s putative uncle is not to be putative and her uncle.  To be a big flea is not to be big and a flea.  A person can be a good liar without being a good person.

At a diner you ask for a sandwich and the waiter brings you one.  I ask for a sandwich and the waiter gives me yours.  You say, “No, give him a different sandwich.”  The waiter picks it up and takes a bite.  “Now it’s different.”

The waiter’s implicit argument is, “This is different, and it is a sandwich; therefore it is a different sandwich.”  The argument is fallacious.  When you said, “Give him a different sandwich,” you were not saying nor even suggesting, that the waiter bring me a dissimilar sandwich.  Rather, you were using “a different sandwich” in a perfectly standard way to mean simply another sandwich—with no suggestion of dissimilarity.  (Though of course we could use that phrase to mean a dissimilar sandwich.)  And as we see in the sandwich example, where “N” is almost any noun, an N can change, can become different, without becoming or being supplanted by another N.  Inferring that it is “a different N” in the sense of another N is fallacious. That is the “minor fallacy” of my section title.

For some special nouns, however, that inference is typically correct rather than fallacious.   It is typically correct with “point,” whether geometrical or propositional.  But the inference would be fallacious with “role.”  A role in a play can be altered somewhat without making it another role.

And yet if we were to replace “different” with “somewhat different” throughout the fallacious argument about the sandwich, the new argument would no longer be strictly fallacious.  If the waiter’s big bite made the sandwich somewhat different from how it was before, the bite would make the sandwich a somewhat different sandwich.  That is because being a somewhat different sandwich (from what yours was) does not imply being another sandwich—or at least, it doesn’t so definitely imply that.  The modifier “somewhat” signals that it is dissimilarity that is at issue, not distinctness.  The modifiers “slightly” and “very” would do the same.


A possible instance of the minor fallacy

One could imagine that Rosemont commits the minor fallacy when he writes:

For Confucius I am my roles. Taken collectively, they weave, for each of us, a unique pattern of personal identity, such that if some of my roles change, others will of necessity change also, literally making me a different person.  (p. 90)

A change in roles is a change; it is a difference.  So is growing up or sitting down, or the change in our material composition that occurs each day.  But being different and a person does not imply being literally a different person.

For, first, note that Rosemont is right in supposing that the word “literal” is meaningful in this context.  “A different person” has not only a familiar literal meaning, it has also a familiar figurative meaning.

Being literally “a different person” from X normally means being another person.  It means not being X.  If you are a different person from X, then you don’t own that house just because X bought it, you don’t have a degree just because X earned it, and you don’t have X’s spouse just because X married him.

Being metaphorically a different person, metaphorically another person, generally means being very different in personality or character.

If someone observes that I am “a different person” ever since the plane accident when all my blood kin and close colleagues died, we do not imagine that she might mean that I was not around when someone else married the person I am living with, and that today I am not someone who lost a daughter in the accident.  We do not imagine that she means to challenge one of the bedrock fundamentals of social order, that we take bodily continuity as the practically indefeasible standard of the difference between “same person” and “different person”—just as we do with apples.  (At least when persons are bodily.)  What she does presumably mean, stated literally, is that my personality or character has changed to a degree that (in the short term) is quite unusual for one person.  I am, metaphorically, someone else.

Writing to appeal the revocation of his admission to Harvard, student Kyle Kashuv apologized for his behavior of a year or two before, saying, “I am no longer the same person.”

We use the idea of being a distinct person metaphorically to mark a big change in personality or character.  “She is another person now,” or “She has become a different person,” or “She was someone else then.”   “She is not who she was.”   The changes marked by this metaphor may be very important.  Their effect may be that we no longer understand the person, or no longer want to be her friend or lover; or no longer want to hold her to account for what she did so long ago.  But the claim is not that there is literally a different person now, another person, someone else instead.  If Kashuv were literally a different person, then the writer of his note would have nothing to apologize for.  And if I say “Don’t judge her for that; she is literally another person now,” my statement is logically problematic.  The continuity marked by the personal pronoun signals that the phrase is being used metaphorically rather than literally.

One reason we want to keep track of persons over time is to know whether the person in front of us today is responsible for what someone did yesterday or five years ago.  If Y is literally a different person from X, then X’s iniquities are not presumptively Y’s fault.  Now, big changes in character over time, the kind that incline us to say metaphorically that someone “is a different person,” can actually reduce a person’s present moral responsibility for past iniquities, since moral responsibility for an action is in some part about the action’s reflecting the agent’s character.  It’s your fault if it’s your flaw.  That is part of why we are not inclined to think it was immoral of Oedipus to marry his mother, though it was a kind of failure.

Not every kind of big difference invites the metaphor.  We do not normally apply the “different person” metaphor simply on the grounds of major amputations or the onset of paralysis or other major illness, big as those changes can be.   More extremely, even the people who think we continue past the corruption of the corpse do not, I gather, use the metaphor on those grounds.  Similarly, we do not apply the metaphor simply on the grounds of radically changed roles or positions.  We would apply it if these or other changes brought great changes in personality or character.  I think that’s what the metaphor standardly means—though we might consider making it standardly mean something else.  And any term can be used metaphorically in unconventional ways, if the speaker provides adequate signals of the new meaning.

Of course there is probably no fallacious inference in Rosemont’s statement.  He is probably not confusing undergoing a familiar sort of change with being supplanted by someone new.  Still, I think, attention to the distinction between (a) and (b)

(a) being different and a person

(b) being a different person

highlights an interpretive question or puzzle for Rosemont’s reader.  Which of the following is Rosemont’s intended point about being made “literally a different person”?

A. According to early Confucians, when some of a person’s roles change, that is or at least causes some change in the person. She is literally different from how she was before.

B. According to early Confucians, a person is the same thing as the roles she has at the moment. That is, when some of a person’s roles change, she becomes, or (more accurately) is replaced by, another person (a point that prima facie would make a radical difference to, say, her property rights and whether she is still her sister’s elder, and that should prompt us to ask what kinship relation might obtain between the new person and the old, and whether the new person has the same body as the old, and whether death brings enough changes in roles that a person does not typically survive her death).

C. According to early Confucians, our roles are so important to us that when some of a person’s roles change, her personality or character therefore changes greatly, in such a way as perhaps to reduce or remove her responsibility for what she did before, or to release her friends from some obligations of friendship. (That is, she is what Anglophones would call metaphorically “a different person.”)

D. Early Confucians not infrequently took phrases that literally meant “someone else” and applied them (metaphorically, if you like) simply on account of changes of roles, never mind changes in personality or character.

E. Something else?

Point A would not distinguish Confucians from anybody else.  Point B looks like a non-starter.  The minor fallacy would be to equivocate between A and B.  As for C and D, has a serious interpretive case for either of them been made anywhere?  I do not happen to recall any Confucian passage that would be relevant; but my reading is narrow and mostly not recent, and my memory is biased.

Of course we can imagine a society in which many of the practical implications that we associate with being the same person over time are suspended in case of certain major changes in positions and roles, marked perhaps by rites of passage: coming of age, say; or marrying, or dying.  That is, we can imagine a society in which, at certain standard transitions, some preponderant fraction of what today is mine would not carry over: my name, my past crimes or accomplishments, my personal property, my health insurance, my friends, my kinship ties.  How might that work?   What in particular might not carry over?  And, what is more to the point for Rosemont’s claim, how and to what extent do we see this in old China, or in the society envisioned by the Liji?  One thinks perhaps of a woman’s marriage.


A major fallacy

Here are two very different questions one might pose on different occasions, or on the same occasion.

1. What is an apple? (What are apples?)

2. Which apple?

Questions 1 and 2 might be replies to “Bring me an apple” or “Bring me that apple.”  Out of context, Question 2 is highly vague or indeterminate.  (Which apple what?)  Question 1 is indeterminate too: are we asking about the semantics of a word or about the structure of nature?  Nevertheless, Questions 1 and 2 are unmistakably very different.  We are unlikely to elide or confuse them.  Insofar as we do, we are gravely disadvantaged for addressing either of them.

And it’s not just apples. The distinction is just the same for sandwiches, points, and roles.

A sufficient answer to Question 2 can be the specification of a relation: “the one on top,” “the biggest,” or “the second one on the left.”  This is what we would often focus on, in addressing Question 2, especially if the apples look similar.  Now all of this is obvious, but note how different it is from focusing on that apple as a thin-skinned tree fruit with a characteristic flavor and texture.

In a way Questions 1 and 2 are opposites.  Roughly speaking, Question 1 asks only what all and only apples have in common.  Question 2 asks only for something to distinguish one apple from others (or to distinguish several from others if the question is “Which apples?”).  Hence, logically speaking, there should be no overlap between a proper answer to 1 and a proper answer to 2.  As a rule, none of the features that pick out one apple from others is part of what it is to be an apple, and vice versa.

First Objection:

Any account of what it is to be an apple that leaves out all the rich detail of actual apples must be false.  No apple lacks detail.  There are no Platonic ghost apples!  What it is to be an apple is to be a certain apple, some particular apple, with countless particular details (and relations to other apples).  Therefore the two questions are the same.  We simply misunderstand Question 1 unless we identify it with Question 2.


Of course not.   The objection misunderstands general terms in general.  Granted, every apple is a particular apple.  But no particular apple is every apple.  There is no apple that is always the one meant by “apple.”  Granted, having a specific current size is part of what it is to be an apple—a part that usually goes without separate mention.  But no specific current size is part of what it is to be an apple.  What it is to be this apple is a combination of two things:  (i) what apples are, and (ii) what it is to be this as opposed to any other apple.  (i) and (ii) do not overlap.  They are the distinct concerns of Question 1 and Question 2, respectively.

We might distinguish a third question, a tricky abstract question that people never ask:

12.  What is it in general to be a distinct apple, i.e. one apple rather than another? That is to say—what is the difference between same apple and different apple?

For example, suppose you see an apple on my kitchen table.  Next day you come back and see an apple on my kitchen table.  You ask if it’s the same apple you saw yesterday, or another one.  I reply, “What does that even mean?”  My question is Question 12.  It’s a silly question.  But that kind of question might not be silly if it is asked about electrons, or points, or houses of cards, or documents (in the age of computers), or an apple on p. 12 and an apple on p. 21 of a novel.  That sort of question might be an important theoretical topic for real estate law in a rainy hilly river basin, or for copyright law about music.

On its face, question 12 (about apples) is at pains to avoid overlap with 1.  On the other hand, not so fast.  For as we might say, question 12 is really asking about a relation among appearances or encounters or time-slices of an apple, and so it is in principle relevant to 1.

Still—and this is important—a good answer to Question 12 will not help us much with Question 1.  The answer to Question 12 (about apples) should be pretty much the same as the answer to the similar question about oranges, or about carrots, sticks, bones, or rabbits.  But the answer to 1 (about apples) should be very different from the answer to the similar question about oranges, or about carrots, sticks, bones, or rabbits.

Second Objection:

For some kinds of things Questions 1 and 2 may be the same.  For example, a point in space, a geometrical point, is nothing but its spatial relations to other points.  And the set of those relations is also what picks it out as one point among others.


No, there is no set of spatial relations that would both pick out one point and answer the question “What are points?” If you ask me what points are, or what “point” means, I might start my answer by saying that they are not made of cheese.  That doesn’t get us very far, but it’s on topic.  If instead you ask me which point, it would be flatly irrelevant for me to say that the one I have in mind is not made of cheese.

Third Objection: 

When someone asks “Which apple?” and I say “the one on the left,” of course I mean the apple on the left.  Therefore a full conceptual account of my answer to 2 will include an answer to 1.  Therefore the topics of the two questions overlap.


Not really.  When you ask “Which apple?”, your Question 2 is assuming that it is an apple, not asking about that.  I might or might not mention that it is an apple in my answer, but that point is off-topic.  Indeed it is possible for me to give an adequate answer to Question 2 even if I don’t know what an apple is.  Suppose I don’t, though you have told me that each of the little boxes on the table has an “apple” in it.  I want to put something on the table, and I ask you to move that apple aside.  You ask, “Which apple?”  I answer, “the one on this end.”  I have given you a full answer but said nothing about what it is to be an apple.  I don’t even know.

Fourth Objection: 

What it is to be an apple constrains the kind of answer to Question 2 that might possibly be correct.  “The one with the ten-foot nose” can’t be correct—not normally.  In this sense any correct answer to Question 2 is informative about the correct answer to Question 1.


Granted.  In that sense there is overlap.

Fifth Objection: 

It is a vast oversimplification to say that Question 1 asks what all and only apples have in common.  (I don’t mean because the answer may be complex, e.g. that all and only apples have “quality A, and either quality B or relation C—unless they began south of the equator, in which case they have quality D and none of A, B, or C.”  We could have a word for all and only the things that fit a complex description of that sort, if we wanted it very much.)  Rather, it’s an oversimplification for at least two other reasons.  First, the natural or conventional standards for being an apple are almost certainly somewhat vague, so that the category has a core and a fringe, so that the most helpful answer to Question 1 might be, or at least include, a varied representative sample of apples and a wink.  Second, thinking of something as an apple normally involves thinking of the thing in connection with certain uses or values, and hence involves implicit standards for what it is to be a good apple.  One could argue that such uses or values are part of the meaning of the word, and therefore have something to do with what it is to be an apple.  Indeed these first and second reasons have some kind of kinship, at least for apples.  The main fringe cases of apples are bad apples: apples too young, or smashed, or half eaten.  We even sometimes use “a good X” to mean “a good example of an X.”  Hence if our answer to 2 is, say, “the best apple,” then there is some overlap between our answer and what it is to be an apple.


There is much validity in what this objection says, but also room for big caveats and further distinctions.   Never mind all that.  Suffice it to say that the objection does not really challenge the radical character of the distinction between Question 1 and Question 2.  Right?


So much for apples.

Here is a pair of questions analogous to Questions 1 and 2.

1a.  What is a person?

2a.  What person? Which person?

Just like 1 and 2, 1a and 2a are almost never conflated, at least outside of philosophy.

We commonly use “Who?” to express Question 2a.  We can’t so easily use “who” in an articulation of 1a.  This interrogative personal pronoun primarily means which person or what person, rather as “when” means what time, “where” means what place, and “how” means what way.

We use “who” in this way when we are asking, “Who is it?” or “Who is that by the window?”, and a name is often an adequate answer.  Or “That’s the new hire” or “It’s Pitxi’s sister.”  Or we might have the name and ask, “Who is Letu?”  The answer is that she is the person by the window.  When we ask “Who is __?”, usually we want some fairly definite identifier, to pick the person out from among others and enable us to seek further information and otherwise keep track of which past and future encounters are with the same person.  We want to distinguish the person from others in a way that we can carry with us.  Names are practically helpful not just because and insofar as they are unique, but also because they can help place a person in a lasting array of people, a reference grid or network: “Björk Guðmundsdóttir” or “Anselm of Canterbury” or “Philip Wheelwright” or “Eunice Kennedy Shriver.”  (For a relevant and fascinating historical discussion of the introduction of surnames in various places, see James C. Scott, Seeing Like a State (Yale 1999), pp. 64-70.)

But we also use “Who?” in some other ways.  A “who” question slightly different from 2a attempts to assign the person a place in some grid but is not concerned with uniqueness.  For example, we understand that a wedding usher asking “And who are you?” is seeking only minimal relational information; an answer may be perfectly good even if it is the same for three people arriving together.

We should not lose sight of just how open-ended or context-dependent is the question “Who?”  What would it mean out of context?

The Caterpillar and Alice looked at each other for some time in silence: at last the Caterpillar took the hookah out of its mouth, and addressed her in a languid, sleepy voice.

“Who are you?” said the Caterpillar.

This was not an encouraging opening for a conversation. Alice replied, rather shyly, “I — I hardly know, sir, just at present — at least I know who I WAS when I got up this morning, but I think I must have been changed several times since then.”

“What do you mean by that?” said the Caterpillar sternly. “Explain yourself!”

“I can’t explain myself, I’m afraid, sir” said Alice, “because I’m not myself, you see.”

“I don’t see,” said the Caterpillar.

“I’m afraid I can’t put it more clearly,” Alice replied very politely, “for I can’t understand it myself to begin with; and being so many different sizes in a day is very confusing.”

“It isn’t,” said the Caterpillar.

The Caterpillar is not, of course, asking Alice what constitutes her personhood.  He assumes her personhood, and that knowledge is already helping to orient him to the situation; presumably now he wants to know something else, something more about how she might fit into his practical world, how he might talk with her.  Yet his question is open-ended.  Even if Alice had not had a confusing day, she might guess very wrongly about what sort of answer the Caterpillar wanted.  She would have more to go on if he were sitting behind a desk that identified his role.  But if he were asking in some formal capacity, he would probably have asked a clearer question in the first place.

Alice felt a little irritated at the Caterpillar’s making such VERY short remarks, and she drew herself up and said, very gravely, ‘I think, you ought to tell me who YOU are, first.’

‘Why?’ said the Caterpillar.

Here was another puzzling question …

What if I ask myself who I am?  What do I mean?  That too depends on the case.  I might actually be confused for just a moment about whether I am Bill or Manyul or a butterfly, if I am just waking from a dream or a drug.  Or I may have forgotten just my name, if there is such a condition.  Or, like Oedipus, I may not know my kinship position—even if I am exceptionally good at recognizing a general account of human beings when so riddled.  Or I might be talking through my effort to think what sort of answer the person in front of me wants, or signaling my effort to remember my medical record number.  In most of those cases, unusual as they might be, I would be using the words “Who am I?” in more or less the normal sense of “Who is X” questions, to mean “which person,” seeking a stable unique identifier.  In other kinds of case, perhaps more common, I would be using “who” of myself to myself in a somewhat different sense, so that it isn’t only or isn’t mainly about distinguishing one person from others.

For example, I might be puzzled at someone else’s action toward me, and ask myself what that party finds distinctively significant about me, as in the lyrics of Casting Crowns’ “Who Am I”:

Who am I, that the lord of all the earth
Would care to know my name
Would care to feel my hurt?
Who am I, that the bright and morning star
Would choose to light the way
For my ever wandering heart?

Not because of who I am
But because of what you’ve done
Not because of what I’ve done
But because of who you are

In this case the question differs interestingly from Question 2a.  The song’s question does not require an answer that picks the singer out from all others (or from all others in a small group at hand), though the question does seem to suppose that the relevant answer is not equally true of everyone.  If the answer is the same for all human beings, the question is wrong: “Not because of who I am.”  Perhaps then just because I am a person, or just because I am a human being.

We might note that the world the lyricist envisions is the kind of world in which persons other than human beings tend to feature prominently (though none of those is actually mentioned unless we count Lucifer).  Hence the singer might have asked about human beings collectively, as being only some persons among others, “Who are we that …?”

Alternately, I might ask myself “Who am I to…,” asking for the grounds of my authority or standing.  An answer might be, “You are someone who has suffered harm from that law; that is what gives you standing to sue to overturn it.”  If we happen to know of several other such people in the room, that is in no way an objection to the adequacy of the answer.

Or I might ask myself “Who am I?” in order to take stock of my own basic concerns or values or character.  For example, if I find myself surprised at what I have done, or at what I am considering doing, I might ask myself, “Who am I?”, wondering about my character and/or personality.  The lyrics of Jessica Andrews’ “Who I Am” report her basic concerns, or what is important to her about herself.

If I live to be a hundred
And never see the seven wonders
That’ll be all right
If I don’t make it to the big leagues
If I never win a Grammy
I’m gonna be just fine
‘Cause I know exactly who I am

I am Rosemary’s granddaughter
The spitting image of my father
And when the day is done
My momma’s still my biggest fan
Sometimes I’m clueless and I’m clumsy
But I’ve got friends who love me
And they know just where I stand
It’s all a part of me
And that’s who I am …

Lyrics by Wade Bowen are less reassuring:

And I love it that you’re my girl
I love that I’m your man
Now that you’re in my life baby
I know exactly who I am

Andrews’ sort of “who” question is one we can ask not only about ourselves, but also about others.

The question Andrews is answering is not “What distinguishes me from others?”  Indeed it is not immediately obvious that this is any part of her question.  I mean, it is not obvious that when she canvasses various candidates for being part of her answer, one of the directly relevant questions she asks about a candidate is, “Does this help to distinguish me from others?”  (Distinguish me for whom?  For what purpose?)  Consider a case where you have an adequate answer to Question 2a about the person by the window—say, that it is Smith from Accounting.  That very answer would have been inadequate if there had been more than one Smith in Accounting.  Can we say the same of Andrews’ question?  Does it logically require an answer that picks her out uniquely?

Uniqueness might be one of her leading concerns about herself.  Must it be?  Should it be?  Granted, for many purposes she must be able to avoid confusing herself with anybody else.  She must usually be able to pick out which person is her.  Only rarely is that difficult.  (She can do it in a room, but can she do it in a picture or a database?  Rosemary’s phone bill?)  But that is not what she is after here.

No doubt it is important to her that others not actually confuse her with anybody else.  But that is not the same as saying it is important to her that her leading concerns about herself are dissimilar from everyone else’s, or indeed from anyone else’s.  Her unique labels (name, face, ID number), and her unique location at any given time, are likely to be enough to avoid such confusion.

No doubt each person’s set of leading concerns is in fact unique.  Even if Rosemary had twin granddaughters with all the same neighbor friends, only one of the twins could be concerned about her relation to her sister Tweedledum.  And even if we are two nameless cult members, I cannot share your concern with your breathing, try as I might.

(In science fiction or the religious imagination there can be conditions or devices that greatly disrupt our ability to distinguish who is who, and even disrupt the conceptual distinction between same person and distinct person.  Such cases may force us to distinguish different standards of “same person” for different purposes.)

More interestingly, part of what makes Andrews’ answer true might be that it connects with the way others distinguish and classify her among people.  She is expressing comfort with being pigeonholed in certain ways, positioned among people.  Of course, her personhood does nothing to position or classify her among people.

We might compare her kind of “who” with the psychologists’ term “identity”.  (I don’t mean the logicians’ term, which applies to sticks and stones.)   Is it a logical requirement of the psychologists’ term “identity” that it seem to the bearer to pick her out uniquely?  Is that point crucial to the very idea?  I am not sure.  I gather that social psychologists in speaking of “identity” tend to stress group membership.  It is said that when Abraham Lincoln asked Robert E. Lee to command the Union forces, Lee answered that he was a Virginian first, an American second.  He wasn’t talking about what made him unique or different.  If he had been talking about that, he might have made it a logical point: “Even if I am American, being Virginian contributes more to who I am, because there are fewer Virginians than Americans.”

To catch in a big net the uses of “who” suggested by the various song lyrics, we might distinguish a third question, just as vague and indeterminate as 2a’s “Which person?”

3a.  What is important or relevant or qualifying about this person?

One version of that question is:

3aa.  What is important to this person about this person?

I guess that’s an approximation of one of the things that gets called “identity.”

I suppose a main function of “self” as a word (not a prefix or suffix) is to equivocate as between 1a and 3aa.

Note that question 3a and its version Question 3aa each differ from Question 2a in an important way.  The proper answer to 2a will not overlap with the proper answer to 1a, but the proper answers to 3a and to 3aa may well overlap with the proper answer to 1a.

Another version of 3a is the interview question, “Tell me about yourself.”

While we’re at it, we might distinguish yet another question or request:

4a.  Describe this person completely.

And of course we can distinguish 12a, analogous to Question 12 about apples.

12a.  What is it in general to be a distinct person, i.e. one person rather than another? That is to say—what is the difference between same person and different person?

We might go on to distinguish ontological, prescriptive, and ethnographic versions of 12a:

12ao.  (See 12a.)

12ap.  Given that X is a person and Y is a person, under what conditions should we regard and treat X as a distinct person from Y?  What should be our tests?

12ae.  What (if anything) does Society S or Culture C regard in general as constituting or at least as marking whether or not two appearances or encounters are with the same person or someone else?

Two other questions are distinct from Question 12ae but might help answer it: What kinds of people does the society recognize?  What are its institutions (roles and relations)?

Perhaps, regarding human persons at least, my reader tends to understand the (literal) difference between “same person” and “another person” in pretty much the same way she understands the difference between “same human being” and “another human being” and the difference between “same horse” and “another horse.”  Or bug or tree.

If my reader thinks that approach to 12a is the best approach in real life, at least for all but purely imaginary or extremely weird cases, then perhaps she thinks the best answer to 12a would not give us much help toward answering 1a, just as (I proposed above) getting the right answer to Question 12, about apples, would not give us much help toward answering Question 1, about apples.  The answer to 12a might help us understand what persons are, but might not give us so much help as to help us to say how people generally differ from crickets or trees.

But my reader may think that thought experiments about imaginary cases can help us answer 12a differently and better.

There is a gulf, if not an impassable one, between attending to our everyday practices of answering question 2a (e.g. “It’s Letu” or “the person by the window”) and working out our implicit general view about 12a.

There is also a question 13a, related to 3a as 12a is related to 2a:

13a.  What is important or relevant or qualifying about a person?

13a is not asking what is important about people as opposed to, say, cows or guns—the kind of question whose answer would not help us sort among people. Rather it is asking what sorts of features of people (as opposed to other people) are important or relevant or qualifying about them.  It is no objection to an answer that the features it mentions do not characterize all people.  We might argue about whether it is an objection to an answer that the features it mentions do characterize all people.  In any case the indeterminacy of Question 13a is quite plain. Each version of Question 3a brings its own version of Question 13a, with its own set of answers.

13aa.  What sorts of things tend to be important to people about themselves?

Note that questions 13a and 13aa are not approximations of Question 1a.

Someone commits the “major fallacy” named in my section title if her argument or exposition crucially conflates some of the several alphanumerically-named questions about persons, and especially if she conflates 1a with any of the others.


One passage suggesting the major fallacy

Rosemont seems to be thinking of Question 1a when he offers to compare two traditions’ views of “what it is to be a person” (p. 71).

He first discusses what he takes to be a standard Western “concept of human beings” (p. 73); i.e. a view of “what makes them uniquely human” (p. 76) and “what it is to be a human being” (pp. 85, 86, 92) and what “human beings are” (pp. 86, 88), a “general picture of human beings” (p. 86) or “model of human beings” (pp. 88, 89).

(In fact the Western philosophical tradition strongly distinguishes “human being” from “person,” mainly so as to avoid ruling out by mere definition the personhood of, for example, gods (most notably two of the “three persons in one” who are God), angels, certain extraterrestrials (such as the Na’vi, victims of human exploitation in the movie Avatar), Neanderthals, porpoises, and pets.  For all these, common usage avoids impersonal pronouns such as “it,” using e.g. “she” instead—and not figuratively as with ships.  Some thinkers also value the distinction as a line of defense against the view that fetuses or embryos are persons.  But I take it Rosemont means to be discussing only human persons.)

Rosemont contrasts this Western view with what he takes to be the Confucian view of “what it is to be a human being” (pp. 81, 89, 92), “what it is to be a person” (p. 92), “qualities of human beings, as a natural species” (p. 89), and “personhood” (p. 91).

That is, he claims to be comparing two traditions’ answers to Question 1a.

Having sketched the Western answer, Rosemont turns to the Confucian answer.

Against this background, let me attempt to sketch briefly the early Confucian view of what it is to be a human being. If I could ask the shade of Confucius “who am I?” his reply, I believe, would run roughly as follows … (p. 89)

Note the surface shift from Question 1a to Question 2a or perhaps 3a.  Let us read on.

… roughly as follows: given that you are Henry Rosemont, Jr., you are obviously the son of Henry, Sr. and Sally Rosemont.  You are thus (sic) first, foremost, and most basically a son; you stand in a relationship to your parents that began at birth, has had a profound influence on your later development, …

Of course, now I am many other things besides a son.  I am husband to my wife, father of our children, grandfather to their children; … my neighbor’s neighbor; …

Now all of this is obvious, but note how different it is from focusing on me as an autonomous, freely choosing individual self…

The shade is consulted in order to bring out an answer to 1a.  But the question he is hypothetically asked is 2a—toward answering 12a?—; or perhaps the question he is hypothetically asked is 3a—toward what?  The beginning of the shade’s answer suggests that he is trying to answer 2a.  But later the picture seems to shift toward 3a or 4a, perhaps toward answering 13aa.  And yet while the focus of 3a and 3aa is importance, some of the importance Rosemont seems to have in mind here is importance in causal explanation of character—not a kind of importance that tends to qualify something as part of someone’s “identity.”

Rosemont is right that it is obvious that a good general-purpose answer to “Who is he?” in sense 2a is very likely to include the distinguishing, locating, situating points that the shade mentions about Rosemont.  These are points that pick him out among people, points that can answer Question 2a.  Some of these points are important enough to be part of an answer to Question 3a too.

Indeed Rosemont’s words do not prove that he is confusing any pair of our various Questions.  One could read him as hinting indirectly and obscurely at a point like the following:

Persons are different from apples in the following way.  For most practical purposes, one apple is as good as any other similar apple.  If there were an angel who went around switching apples for similar apples when nobody was looking, it would make no difference.  But people aren’t like that.  It is in the nature of persons that the distinctions even among similar people matter to us.  For example, people’s particular relationships matter.  It matters that I am the son of Henry and you are the son of George, not vice versa.  Switching us would not do, even if the four of us are peas in a pod.  Therefore non-internal features of persons, relational features, are crucial.  So a good definition of personhood in general will harmonize with that point.

(And he might further suppose for some reason that a Kantian view of personhood does not.)

Certainly we have any number of compelling reasons to keep track of who is who over time—which of our encounters are with the same person and which not—reasons that do not apply in the case of apples.  That is, 2a is an important sort of question about a person.  But note that the line of thought I have imagined and italicized above carries not the slightest suggestion that the qualitative dissimilarities among dissimilar people are unimportant, any more than its apple example suggests that the qualitative dissimilarities among dissimilar apples are unimportant to us.  In other words, the italicized line of thought carries not the slightest suggestion that one’s personal (or other) relations or relationships exhaust what is important about one (3a), much less that they exhaust one’s personhood (1a).  What could suggest that?


A second passage suggesting the major fallacy

This second passage is on p. 72f.

Nothing is more plain, however, than the facts that every one of us does have a definite sex, a color, an age, an ethnic background, … There are not, in short, any culturally independent human beings.  Each of us has specific hopes, fears, joy, sorrows, values, and views which are inextricably linked to our definitions of who and what we are, and these definitions have been overwhelmingly influenced by the cultural community of which we are a part.  F. H. Bradley has made this point—in another context—well:

Let us take a man, an Englishman as he is now, and try to point out that apart from what he has in common with others, apart from his sameness with others, he is not an Englishman—not a man at all; that if you take him as something by himself, he is not what he is … he is what he is because he is a born and educated social being, and a member of an individual social organism; … if you make abstraction of all this, which is the same in him and in others, what you have left is not an Englishman, nor a man, but some I know not what residuum, which never has existed by itself, and does not so exist.

What point is Rosemont making here, and what point is Bradley making well?   One might imagine that each is presenting a thought analogous to the mistaken “First Objection” above about apples.  There is no such thing as a person without any age, sex, color, or culture, so such specifications have to be part of our account of what it is to be a human being (or, for Bradley, “man”).  But charity forbids that reading.

When Bradley says of the Englishman, “he is what he is because he is a born and educated social being,” what is the because explaining?  What does Bradley mean by “he is what he is”?  Does he mean the full description of our Englishman?  That would be too much; it would make the explanation false.  Does he mean “he is a person”?

The explanation is that “he is a born and educated social being, and a member of an individual social organism.”  This is indeed common to all or virtually all human persons, which is a point in favor of its being part of the answer to “What are human persons?” rather than an objection to the question.  It is also true of virtually all wolves—even perhaps those raised by humans—and for all I know it may be true of bees.  No doubt it would be true of humans raised by wolves.

In Bradley’s line “If you make abstraction of all this, … what you have left is … ,”  it is obscure what is the intellectual operation Bradley means to indicate.  If you abstract from an apple’s being a fruit, what do you have left?  Does what we have left include the fact that it is a plant part, or a material object?  Or are those points not left because they are part of being a fruit, which we have abstracted away?  Perhaps Bradley’s or Rosemont’s proposal is that if you describe X as born and educated a social being, and always a member of a community, then you add nothing by adding that X has one or more of the other features often associated with personhood as such, such as various capacities for feeling, thinking, and considered action.  That proposal would be a reminder that those capacities are part of what it is to be born and educated a social being, and always a member of a community.  You add nothing because what you are trying to add is something you have already put in the pot.

If Bradley’s description of what he is looking for—what we have left when we abstract from this and that—if his description is opaque, then what fits the description “never has existed” because really there is no description.

The quote from Bradley is more intelligible to me if I read what directly follows in the original:

If we suppose the world of relations, in which he was born and bred, never to have been, then we  suppose the very essence of him not to be; if we take that away,  we have taken him away; and hence he now is not an individual,  in the sense of owing nothing to the sphere of relations in which  he finds himself, but does contain those relations within himself as belonging to his very being; he is what he is, in brief, so far as he is what others also are.

That is, being in social relations with others is a necessary condition of personhood or manhood.  Charity’s only worry about this interpretation is that Bradley does not here make that point well.  But the answer to this worry, I think, is that Bradley is also communicating a particular version of the point, or reason for it.  His thought is not just that sociality happens to be one of the conventional conditions for application of the word “man”; his point is instead that sociality is necessary in fact for the other features of personhood or manhood.

I am not sure what parallel point might be made about age, sex, or color.

(If Bradley had said “human person” rather than “man,” then we might have asked how much education a human being has to receive before she is a person by his lights.  More than one gets in one’s first day outside the womb?  Can one be a social being before one learns to recognize and identify other people as such?  Must one have been born to be a social being?    And how social does someone have to be to be a social being?  Is a certain level of autism disqualifying?  Such questions have been familiar parts of the mainstream Western discussion of the social aspect of personhood for a long time.)


A third passage suggesting the major fallacy

Here is a passage from the essay that might seem to conflate Question 1a with 2a, perhaps in turn conflated with 3a and/or 4a.

The early Confucians would insist that I do not play or perform, but am and become the roles I live in consonance with others, so that when all the roles have been specified, and their interconnections made manifest, then I have been specified fully as a unique person, with few discernible loose threads with which to piece together a free, autonomous, choosing self.  (p. 91)

Which of the following propositions, if any, is the line about loose threads trying to express?

(A) A certain set of points about a person (relational points) that is sufficiently detailed to distinguish her from all other persons does not include or imply a certain account (the autonomy account) of what all and only persons have in common;


(B) The complete description of any one person focuses on relational roles, and these do not include or imply freewill or autonomous choice; either because (i) our roles are such that freewill is not necessary to our performing them (and there is no other reason to hypothesize it), or else because (ii) our roles are such that freewill is incompatible with their adequate performance, perhaps because freewill would mean we are wholly unreliable.

Thought (A) does not on its face challenge any account of personhood.

Thought (B-i) may misunderstand human roles.  Our roles and relationships require certain capacities that set them apart from the roles of bees.  Ours require whatever freewill or autonomy is necessary for responsibility.  If I support my parents by accidentally killing their mortal enemy as I help her pack the C4, or if I support my parents in ignorance by making negligent investments that turn out to provide an excellent return, or if I have always been respectful only because you have been watching with a gun to my head, then I am not a good son.  Conversely, Oedipus’ marrying his mother did not represent a moral or ethical defect.  He was not a bad son, nor a good one.  He was just a bad son to have.

Thought (B-ii) would seem to misunderstand the Western terms like “freewill, or else to accuse the West of confusion in thinking that the freewill essential for responsibility is consistent with responsibility (consistent with the reliability that gives us what Nietzsche called our “right to make promises”).  If Rosemont’s point is that the Western view is incoherent on this matter, he should situate his point in the context of the extensive Western philosophical literature directly addressing this sort of worry, rather than in the context of views about personhood or identity or Rosemont.


A fourth passage suggesting the major fallacy

In the essay’s next paragraph we find a passage that might seem to conflate personhood (1a) with identity (3aa), suggesting what seems to me a dangerous conclusion about personhood.

Much of who and what I am is determined by the others with whom I interact, just as my efforts determine in part who and what they are at the same time. Personhood, identity, in this sense, is basically conferred on us, just as we basically contribute to conferring it on others.  (p. 91)

(Is there a confusion here between relations and relationships?  That is not my concern today.)

Granted, many of the things about me that should be mentioned if one is saying who I am, in sense 2a or 3a—many of those things came to be true of me only over long periods of time.  (But not all.  For example, I have always been and will always be son of Betty and Oscar.)   Indeed, to use the psychologists’ term, my identity today is different from what my identity was forty years ago.  But I did have an identity back then.  The particularity of our particular relations and relationships, and the slow changes to our identities, do not imply and should not suggest that we become people or “become human” only gradually, over many adult years.  The idea that fetuses are not people strikes some people as inimical to civilization and decency; the idea that becoming a person is something that happens to adults (or only some of them) strikes me as far more so.  Indeed there is philosophical literature on whether and how recognition by others is part of what makes something a person, at least for moral purposes.  Cats (domestic v. feral) and the next century’s machines are interesting cases.  On the one hand, “If I don’t bond with my fetus it doesn’t count so much” has something to be said for it.  On the other hand, justice may have a special place in its heart for the least advantaged.  Xunzi was right to advise governors, “Gather in the orphans and the widows, give assistance to the poor—if you do this, the common people will find security in those who govern.”  And as the Xiaojing points out, a good governor “不敢侮於鰥寡” (Rosemont and Ames, The Chinese Classic of Family Reverence (Hawaii 2009); p. 30 for the Xunzi and p. 109 for the Xiajing, translating the latter line as “would not presume to ignore the most dispossessed”).  A cultural tradition can place an especially high value on hospitality to strangers and the lost.

Separately: the proposal that our relational roles exhaust our personhood should give us pause because we should worry that we are perhaps each being told, “You are only a functionary in our institutions.  You have no other existence or meaning.”  We should be careful about exactly what is being proposed and denied, and what reasons are offered to support the claim.

Here are three opinions of mine.  (I) A human is a person within a year of conception, and remains a person until death—with arguable exceptions for extremely grave maladies.  Additionally, (II) she remains literally the same person.  She doesn’t become that person; certainly not after she is old enough to read.  And she doesn’t become somebody else; not literally.  But (III) metaphorically she can become “another person,” as discussed above, and in that sense become a certain person, a person she was not before.  We can have that sort of failure or success.  People can change.

Objection to (II):

Even in middle age I can literally become the person who is Jackie’s husband (and has my other features), and thus literally become that person.  Therefore (II) is false.


Sure you can become that person, and so much more.  In middle age you can even become the person who has recited the alphabet at a certain place and time, and thus literally become that person.  This alphabet example is plainly not a counterexample to (II), though the case for its being so looks the same as for the objector’s example.

(Claims (I) thru (III) are consistent with the view that there are degrees of personhood, and that e.g. a cat is a person to some degree but not fully.  The idea that there are degrees of personhood does not imply that there is no such thing as full personhood, nor that the typical child of 4 does not fully qualify.  Similarly with apples: even a poor apple may be fully an apple, while some things may be apples only to an extent.)

Objection to the imputation of the fallacy to the fourth passage:

Here is partial prima facie defense for Rosemont against the charge that he is conflating questions 1a and 3aa.  He may instead be assuming that the Western philosophers are committed to that conflation, because they are committed to an idea beyond their autonomy answer to 1a.  Their further idea is that personhood, autonomy, is the whole ground of morality.  From this further idea, Rosemont may suppose, it follows that what is fundamentally important about each person is just her personhood, so that her true fundamental identity—probably not her self-concept, but nevertheless the only thing that is qualified to be her correct fundamental self-concept —is just that she is a person (autonomous etc.).  That is, who she is in particular, at bottom, is the same as everybody else; and it is something she has always been, not something she has come to be.  And this, Rosemont may want to say, is a false view of who she is in particular. Rosemont’s intent may be to suggest that we need the Confucian answer to 1a in order to avoid this absurd conclusion.


Even a Western philosopher who thought that the personhood of all persons or all humans is the fundamental moral or practical principle would not infer that my personhood is the whole of what is fundamentally morally or practically important about me.  For—to be very schematic about it—general moral principle has no concrete practical implications about our immediate actions (swinging a sword here or there or not, studying this or that) except in conjunction with facts about the particular cases at hand: including facts about me, my life, my unalterable parentage, my present marital status, my financial situation, what promises I have made, who is or is not coming at me, and how swords work.  What has concrete implications is the conjunction of at least two sorts of thing: (a) general moral principle and (b) factual premises about the case at hand.  (a) gets no purchase on action without (b).  The facts needed for (b) are quite a lot of material, and they are not derivative from general moral principle, and some of them are about the persons involved.  So even if the autonomy of each person were the fundamental principle of morality, it would not be fundamental to all of what is important about anyone.  It would not be all of what is fundamentally practically significant about a person, nor even what is permanently fundamentally practically significant about her.

The point that general principles can’t have particular practical implications without particular facts as co-premises is a commonplace of the Western philosophical classroom, not widely rejected in the profession as Rosemont may seem to suggest on p. 38 of Against Individualism.  Also this point need not presuppose a distinction between facts and values.  The point is valid even if we think of moral principles as general facts, e.g. that the gods enforce certain rules, perhaps the rule that we should perform the roles conventionally associated with whatever family positions we may happen to hold.  The point is rather about the difference between the general and the particular—though of course some of the necessary non-moral facts may be rather general.  Some are about my basic situation, or other morally relevant general truths about me in particular.

Indeed, while the above defense of Rosemont relies on introducing a strange concept of true fundamental identity in order to generate a problem in the Western view, what seems to avoid the problem for the Confucian view is that the defender I have imagined simply refrains from introducing this concept in that connection.  The defender above has made no defense of the idea that the concept “true fundamental identity” has any presence in Western philosophy.  Indeed that concept would seem to be simply a device for conflating 1a and 3aa.  The problem the defender points to seems to be supplied to the West by the defender, and not supplied to Confucianism.



None of my worries about the language in Rosemont’s essay need amount to objections to Rosemont’s actual views or intended arguments.  They point to some obstacles to understanding him, at least for me; obstacles that do not necessarily indicate errors.  I have expressed no opinion on the truth of his account of any Western view.  And of course I have not touched on the many other rich ideas in the essay.  Here I have only glanced at a few features of its wording, as I am gripped by the analytic philosopher’s obsession with the rectification of words and phrases—a battle that can never be won.

But the distinctions I have been concerned with, especially between 1a and 2a, are not subtle.

One possibility I have not touched on is that the fallacies I have considered ascribing to Rosemont are actually deep confusions Rosemont ascribes to Western philosophy and his assumed reader; mistakes that his essay is trying, dialectically, to cure.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.