Regret as a sufficient condition of personhood; new version.

Regret as a Sufficient Condition for Personhood/1

Regret as a Sufficient Condition for Personhood

James DiGiovannaAssistant Professor of PhilosophyJohn Jay College, [email protected]

Keywords: personhood, personal identity, emotion, regret

Abstract: Some emotional states are complex enough to serve as interesting, sufficient conditions for personhood. Regret is presented as such a case, as it fulfills standard criteria for personhood, such as self-consciousness, continuity of memory, reflective judgment, narrative sense of self, judgment, moral consideration, care, and second-order volitions. Regret is especially interesting in that it secures some degree of identity across time and is not only a mark of identity, but anidentity-making emotion. That is, regret alters how we think ofourselves and establishes both a relation between past and present selves and an attitude towards future selves. By virtueof regretting, we put forward that we were a certain person, that we hold a (negative) judgment about that person, and that we would, if possible, choose to be a different person in the future. Thus, continuity across time, and the creation of character are united in regret, bringing together diverse strands of the personhood and identity debate, and establishingfurther that an emotional attitude may be a necessary and sufficient condition for personhood. Further, the capacity for


regret a necessary condition. An entity that was incapable of regret would be a person in a reduced sense. It may well be that by happenstance or accident there is a human who never does anything regrettable. But if such a human did do somethingregrettable, but could simple shrug it off with no sense of regret or remorse, we would think them deficient just as we find the sociopath deficient, or the person incapable of love. If we accept that personhood is a success concept, as Dennett seems to, then perhaps we have not succeeded without the capacity to feel bad for what we have done when we have done wrong. As Williams points out, there is something wrong with the truck driver who does not regret the accident that killed the child.

1. Introduction: Emotion and Personhood

There are a handful of emotional states that are individually

complex enough to count as sufficient conditions for personhood,

and central enough to what is sometimes called “the human

condition” to be necessary for personhood. It is hard to imagine

that any entity that had never experienced sorrow could truly be

a person. If an intelligent being that was utterly devoid of

sorrow existed, we would have a certain difficulty in ascribing

full personhood to it (it might be something like an angel, or an

intelligent but unfeeling robot) because it would lack a certain


capacity to be injured that we count as essential in assigning

rights and protections1. Further, we could not converse with it

about something that we take as central and essential to our

definitions of ourselves.

However, a being could experience sorrow and not be a person. I

imagine that dogs might feel sorrow at times, say, upon being

bereft of their companions for extended periods.

But certain emotions contain so much cognitive content, and of

such a specific sort, that they alone would show satisfaction of

most, and perhaps all, widely accepted and proposed conditions of

personhood. In this paper I will make a case that regret is such

an emotion. I am quite convinced that it is a sufficient

condition of personhood, and at least tentatively will hold that,

1 One problem in saying what counts as a condition of personhood is that the process is in part a process of defining personhood, so there’s the risk of circularity, or at least of having no clear standard to appeal to. I’d note, though, that the standard claim that persons are those who are protected by laws and accorded rights provides a decent heuristic: if an entity could not feel pain, grief, sorrow, or any negative emotion, it’s hard to see how it could be harmed, at least from its own perspective, and since much of law is concerned with protection from harm, there is at least a question of how to apply law in regard to such a being. On the other hand, a being of this sort might well have consciousness and the ability to reason, so we could conceivably hold it responsible for actions that would harm others. As such, it would have an interesting status, a kind of partial personhood. We ascribe partial personhood in this legal sense to children or the extremely mentally impaired, though in the inverse manner: we often don’t hold them responsible for the actions, but do protect them from harm. So an emotionless being would be a sort of inverse child.


just as it is for sorrow, the capacity for regret is necessary

condition of full personhood. I would accept that some beings

that currently feel no regrets, and by happenstance will not feel

any regrets in the future, are indeed persons, if only they have

the capacity to regret. I think it may be the case that very

young children might not count as persons on this account, as I

assume there may well be a period before which true regret is

impossible, but I do not think that is a terrible flaw with the

standard proposed.

Some other emotions that might also serve this purpose, though

I won’t argue for them here, and I could well be wrong about any

or all of them, are grief, anxiety (there’s already a long

literature on anxiety, of course, starting with Kierkegaard’s

Concept of Anxiety (1980a) and moving through Heidegger and the 20th

century existentialists), hopefulness, the happiness we feel when

we someone besides our self does well,2 etc. I’ve chosen regret

both because it seems to fulfill conditions of personhood, and

because it’s a particularly undervalued emotion. People actually

2 There’s no separate name for this emotion in English, but it strikes me as distinct in its cognitive and affective qualities. In Buddhist circles the term “mudita” is sometimes used for this feeling.


take pride in saying they have no regrets, though I think they’re

probably lying (or are, at the moment they make this claim, at

least, not aware that it’s false) and if they really did have no

regrets at any point in their lives, then they missed out on an

essential and singular experience, an experience that adds

importantly to the richness of life.

In section 2, I’ll spell out in brief how regret fits into the

existing literature on emotions, and how the notion of personhood

I wish to address fits into the tradition of philosophizing on

persons.

Regret, I will claim, is a cognitive emotion, in that it

includes propositional content, and that it is not simply a

transient, emergent state. Rather, it is something that lingers,

and may not always be present to consciousness. In this regard it

is not like a mood, which colors all experiences while the user

is under its spell, but is rather a “sentiment,” “condition,” or

“established emotion,” like lasting love or long-standing grief.

While not always present to consciousness, it returns and when it

returns it is identifiable by its object.


I’ll use Locke’s idea of “person” as a forensic concept as

central to the notion of personhood, insofar as any notion of

person must be of an entity that can be the object of blame and

praise. The ancient notion of person as rational agent who

struggles with the irrational parts of the soul, Kant’s idea of

the person as the end-in-itself, and the narrativist’s notion of

the entity that creates for itself a story of it itself, all fall

into this category. Most strongly, Frankfurt’s claim that persons

are entities with second-order volitional states is a

clarification of Locke’s view, and Dennett’s list of person-

characteristics should imply Locke’s idea. I’ll differ only from

those who reduce person to human being, as I’m uninterested in a

species term, looking rather to the sort of entity that should be

subject to the laws and rules of a society.

In section 3 I’ll discuss the failure to include emotion as an

essential trait of persons in soe of the philosophical

literature. I’ll look to exceptions to this, and to the

psychological literature on personhood and emotion, to see how

emotion may be necessary to personhood, and then prepare the


ground for regret as the sort of emotions that are not only

necessary to personhood, but unique to it, and sufficient for it.

In section 4 I’ll go through Dennett’s conditions for

personhood , assuming this list covers most of the standard

conceptions of the concept, and show that regret meets the

conditions. I’ll also consider Frankfurt’s condition, and show

that for there to be regret, there must be second order

volitions.

In section 5, I’ll address narrativist accounts of personhood,

which regret is well-situated to fulfill. I’ll focus on Marya

Schechtman’s narrativism, because of its general influence and

clarity, its understanding of the role of emotion in narration,

and because her conditions apply well to other narrativist

accounts. Regret, as a narrative emotion, is interestingly

person-building, in that it creates a story that evolves over

time and ties a person to a past event. It thus is a central

narrative ‘hook,’ making a person one with a past self while at

the same time creating narrative structure by making that past

self a source of conflict.


In section 6 I’ll look at this conflict and how it complexifies

identity. Any concept of identity across time must grapple with

the problem that, while we have a degree of continuity of

identity, we also have many discontinuities. Regret highlights

this, and shows us in alienation from a past self, and thus

internally split. But cannot simply not be that self, or there

would be no conflict, so it also shows us, perhaps inextricably,

bound to the past self, and establishes that, whether we prefer

it or not, we see ourselves as continuous across time. That we

dislike this continuity and cannot escape it makes it more

powerful, though that we can see a lack of identity between how

we are now, or how we would choose now, and thus who we are as

agents, and how the past self was and would choose, highlights

the inherent tensions and multiplicity within any identity across

change.

In the conclusion, I’ll sum up why regret is so uniquely

situated as a criterion of personhood: not only does it create

identity across time, it does so by relating a person to him or

herself through blame, which establishes a moral consciousness,

one of the hallmarks of personhood. And it demands a narrative,


and reflective thought, and higher order intentional states:

regret involves wishing we had had some other intentional state

in the past. It is propositional, it includes judgments, and of

course it is linguistic. Further, it is not merely rational,

though it is that, it is also emotional, and so ties into a

central realm of being a person. There may be no other emotion so

well situated, and it may be that truly having never had any

regrets is the mark of a non-person: either an angel who has

always done and thought perfectly, or an animal who cannot feel

self-dissatisfaction and the desire to change for the better.

2. What persons, which emotions

a. Where regret falls in the taxonomy of emotions

In making a claim that a certain emotional state is a

sufficient condition of personhood, we should first clarify what

we mean by “emotion” and “person,” two terms with long,

contentious histories in philosophy, psychology and law.

I do not wish to take a strong stance here on the dispute

between cognitivist and affectivist theories of emotion. However,


regret, as I understand it here, must have cognitive content.

Thus I won’t be abiding by the “James-Lange” theory that reduces

emotions to bodily feelings3. It may well be that this is an

accurate account of emotions, in which case regret is not an

emotion, or not just an emotion. Further, regret will not be

understood simply as a passing feeling/cognitive state, but

rather as something that is importantly recurrent, and perhaps

active even when not experienced (I’m agnostic on the later

point.) So it will be, again perhaps, more like what Aaron Ben-

Ze’ev calls a “sentiment4” than an emotion: while it has a

specific intentional object, it endures, like grief or love, and

needed be emergent, like rage, for it to be said to exist.

Perhaps better would be to describe regret in Roddy Cowie’s terms

as an “established emotion5,” that is, one like grief or shame

that may form a background to experience and makes repeated

appearances in conscious life, and which includes an intentional

object that is, to some extent, narrative. Cowie later calls this

sort of emotion a “condition” as opposed to a “state,” and refers3 James, William, “What is an emotion?” Mind, 9: 188-2054 p.55 , “The thing called emotion,” Oxford Handbook of Philosophy of Emotion, 41-615 P. 3520, “Perceiving Emotion: Towards a Realistic Understanding of the Task”Philosophical Transactions: Biological Sciences, Vol. 364, No. 1535, pp 3515-3525


to the period during which the emotion persists as an “emotional

era.6”

This terminology is probably most helpful in discussing the

sort of thing regret is; it is clearly not necessary for someone

to be presently having an emergent emotional state in order to

say that he or she regrets something. Nonetheless, it is

necessary that there be associated cognitive content regardless

of the somatic content, so, without endorsing their entire

program, regret will fall in line with cognitivist views on

emotion insofar as it contains propositional attitudes.

While I accept that Schwitzgebel is correct in that we may have

very poor cognizance of our emotions, I don’t think this impedes

the value of regret as a marker of personhood. As Schwitzgebel

notes, citing a study by Russ Hurlburt, “A person might very

frequently have angry thoughts about his children, as he reports

when sampled at a random moments, and yet he might sincerely deny

that it is so in the general case. 7” Similarly, a person could

have recurrent regretful thoughts but not be able to access these

6 P.74: “Describing the forms of coloring that pervade everyday life,” Oxford Handbook of Philosophy of Emotion, pp 63-947 P. 5-5, “Self-ignorance,” forthcoming in JeeLoo Liu and John Perry, eds., Consciousness and the Self (Cambridge: Cambridge)


well when asked; the fact that the thoughts are recurrent and that

they then color his or her outlook, and that they have an

intentional object, is all that’s needed for them to do the work

required here.

Recently, questions have been raised about the universality of

emotions, and it has been suggested that emotions are all

culturally specific, or at least the interpretation of emotions

is culturally specific8. Regardless, regret point to general

human capacities: the ability to form a personal narrative, and

the ability to have a negative reaction to what one has done, and

the ability to feel responsibility for a prior act or omission.

Whether this collection of capacities, when experienced

collectively, has a name in some language or not, it would at

least be odd if it never occurred in some large sample of people.

I await some evidence that this is the case, but meanwhile will

accept, to at least a limited degree, Ekman’s claims about the

universality of at least some emotional content.8Tanaka A, Koizumi A, Imai H, Hiramatsu S, Hiramoto E, de Gelder B. “I feel your voice. Cultural differences in the multisensory perception of emotion” Psychol Sci. 2010 Sep;21(9):1259-62.; Jeanne Tsai, Brian Knutson, Helene Fung,“Cultural variation in affect valuation,” Journal of Personality and Social Psychology, 2006, Vol. 90, No. 2, 288 –307; Jeanne Tsai, “Ideal Affect: cultural causes and behavioral consequences,” Perspectives on Psychological Science (Wiley-Blackwell). Sep2007, Vol. 2 Issue 3, p242-259


So, in short, I will understand regret to be a (usually)

recurrent, affective state, accompanied by a negative judgment

towards a past action or failure of action on the part of the

agent having the state. It will thus have a reasonably consistent

object. While there may be a sense of “regret” that includes a

non-recurrent state, insofar as one can briefly regret something,

I’ll be more interested in the form of regret that is longer

lasting, like grief or love, and which, while not always present

to consciousness, forms an element of the lasting psychological

content of the agent, and is identifiable in part by having a

consistent object. In this, it is not only like a memory, but can

be an intentional state towards a certain memory, or towards a

large memory set by virtue of that set being indicative of the

failure of the agent to complete some action. So, one could

regret not going to college, which is not a regret towards a

memory as such, though it may be a regret about a decision to not

go to college. But even if no such decision were made, there

would still be an accompanying set of memories: having a job one

doesn’t like, being left out of discussions of one’s college


experiences, etc. It can also be that the attitude frequently

doesn’t address memories, but rather emerges when one encounters

those who did go to college, and, in response to envy, feels

regret, but I think this would still involve an attitude towards

one’s past, which, if not strictly a memory, shares with memory

that it is personal and backwards-directed.

b. The sorts of person-concepts addressed here.

[Three important aspects of personhood: Each of this is meant to be necessary, and the

first two are meant to be necessary and sufficient, at least according to the

philosophers listed. There are other interesting sufficient conditions that might cover

these, for example, the suggestion that persons are players (Johan Huizinga, Bernard

Suits), that persons are futural (Heidegger, Sartre),

a. Rational (note Locke’s discussion of this, also Descartes, Aristotle, Plato, etc.)

b. Accountable (Locke’s “Forensic” sense, Frankfurt, legal definitions of person,

Kant): Blame and praise (third and first person), moral agency, sense of self as

moral agent

c. Relational. The sort of entity that enters into meaningful relations with others


a. Persons are sought out by other persons not merely for reproduction and

safety, but because the company of persons is considered important

b. Note Schechtman, Friedman’ “Autonomy and the Split-Level Self,”

Aristotle’s social persons

c. Akan conception of a person, Yiddish term “real mensch”=real person: the

idea in both being that the extent to one is socially valuable is the extent

to which one is a person…(more to be said here)

d. Probably the non-regretter would not be able to form social relations

because he/she would be unable to adjust his her behavior to social cues.

We probably have all met the non-regretter. “I hurt your feelings? So

what? I’m just being me!”

Others: Aristotle said that humans are “the only animals that laugh”( Aristotle,

Sense and Sensibilia, IV, 441a1–2, in The Complete Works of Aristotle, vol. 1, ed.

Jonathan Barnes, trans. J. I. Beare (Princeton: Princeton University, 1984), p. 10.,

cited in “A philosophy of tickling” aaron Schuster,

http://cabinetmagazine.org/issues/50/schuster.php)]

“Person,” and its variants, is even more fraught a term than

“emotion,” but the leading positions probably have more in common

than the most divergent of the leading positions on emotion.

Since “person” has a legal sense, definitions of “person” should


meet the criteria of being applicable to a being to whom one

attributes responsibility and blame for actions. Some versions,

notably hard-core behaviorist and determinist conceptions, might

not meet this condition, but only for the hardcore behaviorist or

determinist; if the entity described is one to whom most would

ascribe responsibility and blame, even if described in

behaviorist/determinist terms, then it’s probably meets this

constraint for definitions of “person.” Recent research indicates

that even in determinist situations, people will still blame or

praise an individual. Eddy Nahmias, for example, has found that

certain wordings of determinist cases still elicited blame

towards the characters described in experiments with naïve

subjects9.

The discussion of personhood dates to ancient times.

Aristotle’s analysis of “man” (anthropos) in Nichomachean Ethics,

though perhaps only a species-term, does seem aimed at the goal

of explaining personhood. He explains that humans, unlike other

animals, have a “rational principle” in addition to their

9 Eddy Nahmias & Dylan Murray , “Experimental Philosophy On Free Will: An Error Theory For Incompatibilist Intuitions” forthcoming in New Waves in Philosophy of Action


irrational principles, and that it is precisely this struggle

between the rational principle that explains the sort of moral

praise and blame we reserve for persons:

“There seems to be also another irrational element in the

soul-one which in a sense, however, shares in a rational

principle. For we praise the rational principle of the

continent man and of the incontinent, and the part of their

soul that has such a principle, since it urges them aright

and towards the best objects; but there is found in them also

another element naturally opposed to the rational principle,

which fights against and resists that principle. For exactly

as paralysed limbs when we intend to move them to the right

turn on the contrary to the left, so is it with the soul; the

impulses of incontinent people move in contrary directions.

But while in the body we see that which moves astray, in the

soul we do not. No doubt, however, we must none the less

suppose that in the soul too there is something contrary to

the rational principle, resisting and opposing it. In what

sense it is distinct from the other elements does not concern

us. Now even this seems to have a share in a rational


principle, as we said; at any rate in the continent man it

obeys the rational principle and presumably in the temperate

and brave man it is still more obedient; for in him it

speaks, on all matters, with the same voice as the rational

principle.10”

This struggle between inner principles is coherent with the

following discussion of regret, and I think having regret would

suggest this very struggle, so I believe it will serve as a

criteria for personhood in the Aristotelian sense, though I’ll

not directly argue that, assuming it will follow from the larger

argument.

I’ll ignore medieval/religious notions of the person as an

entity with a soul that exists after death, as these are not

relevant to this discussion. I doubt regret alone would be a

sufficient condition to establish this, though it would be an odd

deity that would allow moral regret to non-persons.

Notably, the modern period has a great deal of discussion of

personhood, beginning with Descartes’ self-definition as a

“thinking thing,” clearly an insufficient notion for legal

10 Nichomachean Ethics, book I, section 13, trans W. D. Ross, retrieved from classics.mit.edu


personhood, but one easily met by a cognitivist account of

regret.

More strongly, Locke directly addresses the term “person,”

saying “This being premised, to find wherein personal identity

consists, we must consider what PERSON stands for;—which, I

think, is a thinking intelligent being, that has reason and

reflection, and can consider itself as itself, the same thinking

thing, in different times and places; which it does only by that

consciousness which is inseparable from thinking, and, as it

seems to me, essential to it...11" This will be well-addressed by

regret, as regret, as I argue below, shows reason, self-

identification, and self-identification across time (and

presumably space.)

Kant addresses personhood in Groundwork for the Metaphysics of Morals,

“Now I say: man and generally any rational being exists as

an end in himself, not merely as a means to be arbitrarily

used by this or that will, but in all his actions, whether

they concern himself or other rational beings, must be always

regarded at the same time as an end… Beings whose existence

11 Essay Concerning Human Understanding Book II, Ch. XXVII, section 11


depends not on our will but on nature's, have nevertheless, if

they are irrational beings, only a relative value as means,

and are therefore called things; rational beings, on the

contrary, are called persons, because their very nature points

them out as ends in themselves, that is as something which

must not be used merely as means, and so far therefore

restricts freedom of action (and is an object of respect)...

rational nature exists as an end in itself. Man necessarily

conceives his own existence as being so; so far then this is a

subjective principle of human actions. But every other

rational being regards its existence similarly….12”

Understanding oneself as an end in oneself, that is, thinking

of oneself as containing a plan that aims at oneself, seems

central to regret. Again, I won’t argue this explicitly, but I

think it follows from the argument below that one regrets because

one understands oneself to be the, or one of the, objects of

one’s actions and the, or one of the sources of worth for those

actions. We would not regret, in this sense, if we did not think

we could have done better, and to think this is to instill

12 Section 2, trans. Thomas Kingsmill Abbott, retrieved from Gutenberg.org, 11/18/13


oneself with a degree of moral value, or the capacity to be

morally evaluated.

Versions of “person” have proliferated in the latter half of

the twentieth century up to the present. Frankfurt’s influential

notion of the person as entity with second-order desires is

probably a sufficient condition for meeting the standards of most

of the historical notions of person. Narrative accounts of

persons, such as those of Schechtman,13 Goldie,14 and Taylor,15

while varying enormously, accept that persons are essentially, in

non-damaged cases, entities that create self-narratives. Regret

is, essentially, a self-narrative, I will argue. Schechtman makes

a special exception for extremely damaged individuals to be

accepted as persons without the capacity for generating self-

narratives if they are accepted into “person space” be being

treated as persons by those who meet the stricter criterion or

criteria for personhood.16 I don’t accept this conclusion, but

even granting it, I’ll only be talking about persons in the 13 2005, “Personal Identity and the Past,” Philosophy, Psychiatry, & Psychology, 12: 9–2214 Peter Goldie, The Mess Inside, Oxford, UK: Oxford University Press, 201215 Charles Taylor, Sources of the Self: The Making of Modern Identity, Cambridge, MA: Harvard University Press 198916 Marya Schechtman, “Personhood and the Practical,” Theor Med Bioeth. 2010 Aug;31(4):271-83.


stricter sense, and only appealing to those definitions which

pick out entities that are worthy of praise or blame.

Similarly, I won’t address “animalist” accounts of personhood

which relate it only to species-type (note Wiggins Self Knowledge

and Self-Identity 22-25; Olson Was I Ever a Fetus, etc., Bernard Williams

Self and the Future where he thinks he’s his body after brain wipe)

These are clearly insufficient for establishing the sort of

entity that is worthy of praise or blame, and insufficient for

Locke’s “forensic notion.”

I’ll give more detail in the next two sections on which senses

of “person” I’ll be appealing to, but the above is meant to point

out that on most accounts of “person,” regret should be

sufficient, and while I’ll be focusing on Dennett’s definition,

because I think it contains most of the above, the following

argument should be applicable, mutatis mutandis, to the other

definitions that appeal to the standard features of rationality,

accountability, narrative, etc. As well as Dennett, I’ll look

closely at Frankfurt’s definition, because it’s influence is

strong, and it’s scope is also unusually wide, such that a

version of “person” that didn’t imply, perhaps implicitly,


Frankfurt’s account would be a suspect version since it would

include a kind of person incapable of the sorts of self-

reflection that form the basis for moral agency.

3. Personhood and emotion

It has been noted that Dennett’s conditions for personhood

don’t include emotional content, and that emotion is probably a

necessary condition (Dennett, 1976). Indeed, one could

specifically argue that even an inability to feel empathy (think

of the role of mirror neurons in our general development) would

leave one with a highly diminished personhood, and the same could

be said of a lack of sorrow or joy.

Kant certainly downplays the importance of emotion, and does

not include it as a necessary condition of personhood. Indeed, to

behave morally is to behave without regard to the emotions except

for respect, which he holds is not quite an emotion (Groundwork).

Nonetheless, he discusses the role of the emotions, but they tend

to be given negative valence. The opposite is true for Nietzsche,

who held emotion to be essential to humanity (Gay Science, books II


and III)s. He thought it was ineliminable, even in cognitive

process, and so foreshadows D’Amascio’s discussion of the nature

of emotion and consciousness. One could include Kierkegaard,

Hume, and Schopenhauer among the emotion-as-essential-to-

personhood camp; Aristotle probably falls here too, with his

notion of right feeling as part of virtue (Nichomachean Ethics)

Arguably, Descartes and Plato see emotion as dispensable or

inessential to personhood; Descartes knows himself only as a

“thinking thing” and assumes emotions are bodily, not mental

(Meditations, Passions of the Soul), and Plato sees overcoming and

conquering the emotions as the philosophers goal, assuming this

can be completed in death (Phaedo).

The psychologist James Hillman (1979, among other works) wrote

a great deal about the role of the “shadow” emotions: shame,

guilt, depression, etc. All of these, too, seem necessary for

full personhood. Perhaps you can remember the first time your

heart was broken, and how you couldn’t explain it to a friend

who’d never been in love. That friend had not yet suffered

something that is, in some ways, essential to full personhood.

Our ability to interact with one another depends to a great deal


on our having a shared set of emotions, and to some extent,

personhood is predicated on this kind of communicability of

experience (this is, in part, what Dennett is talking about with

conditions 3, 4 and 5: that we take a stance toward an entity,

that it can return that stance, that it can communicate). Paul

Ekman (1971) listed five emotional states that he found that

people in every culture could recognize: anger, fear, sadness,

happiness and disgust_. This list has been expanded to also

include surprise and contempt (Mesquita & Frijda, 1992), and it

would be hard to imagine how we could speak with a people who

never felt any of these emotions.

Antonio Damascio (1999) claims that emotions are necessary for

a sense of self, and indeed for most of what we call

consciousness. While this claim may be overly strong, there’s at

least a basic level at which it’s hard to see why we would think

or do anything without some motivation, whether that’s desire,

fear, hope, etc.

Many philosophers, though, follow the psychologists in

understanding that without emotion, we are not persons.


Given, then that certain emotions are necessary for social

interaction, and perhaps necessary for thought, at least some

emotion seem interestingly necessary for personhood. Some

emotions, though, have such complexity that they act as

sufficient condition. Regret has the nice quality of necessarily

including memory, a connection to a prior self, and self-

reflection. Some forms of regret require a relation to others,

though that’s only very loosely necessary for regret to exist at

all (obviously, we have to have acquired language and sentential

judgments to have regret, or at least to know that we’re having

regret, and language is of course acquired socially.) So a

weakness of this condition is that it doesn’t necessitate seeing

oneself in relation to others. I actually think this is a large

weakness,17 but not a disqualifying one. Obviously, to regret we

have to have some idea of better and worse, some sort of judgment

about the quality of our actions, and these are at least partly

social. We could specify a kind of regret that involves others,

17 I’d hold that this is a weakness because I think selves are inherently social, and a condition that stressed that more strongly would be a better condition. Nonetheless, there are some minimal social aspects in regret.


but I’ll hold off on that for now. I think all by itself, regret

is still a sufficient condition.

4. Basic Conditions

Dennett (1976) lists 6 conditions for personhood: (1)

rationality, (2) that intentional predicates apply, (3) that a

stance is taken toward that entity such that it is treated as an

entity capable of action (in the sense of the work of an agent),

(4) that it can reciprocate the stance, (5) that it can

communicate verbally, (6) that it is self-conscious or aware of

itself. It’s notable that there is no emotional condition

whatsoever: one could meet all of Dennett’s conditions without

ever having been sad, felt empathy for another, or had a desire.

Condition 3, stance, always seemed especially problematic. If

everyone ceased to treat some entity as a person, that entity

only ceases to be a person in a socially constructed way that I

think we’d like to say is wrong. For example, granting 3/5ths

personhood to slaves didn’t make slaves 3/5ths persons except in

a legal sense, and a legal sense that was rightly challenged as


failing to live up to some basic standards for what law should

be. If we accept a broad reading here, that an entity must merely

be capable of being treated in this way, the condition becomes

excessively broad, since we can have the stance toward virtually

anything. We cannot by fiat make a walrus or a doll into a

person.18

However, condition 3 drops out of we think of condition 6,

self-consciousness, as including treating oneself as an agent.

This is why regret, I think, is sufficient here: when we regret

we regard ourselves (if only perhaps our past selves) as persons,

because we regard those selves as taking an action that they

shouldn’t have taken, that is, we regard them as agents, and

therefore as beings with intentions, which is why we judge their

18 Though see Marya Schechtman’s “Personhood and the Practical” (2010), where she claims that persons become persons by occupying what she calls “person space” and being treated as persons, even if they lack rationality, language, etc. She claims this does not open the door to a pure conventionalism, but it’s hard to see why it doesn’t. In response to the objection that we could then treat dogs as persons and admit them into “person space,” she only offersthat human beings usually have person qualities whereas dogs and dolphins do not. Thus, she holds that humans who are seriously mentally disabled can be granted personhood by being admitted into person-space (i.e. by being treated as persons), but that we don’t need to worry about this extending to dogs or dolphins because that the humans who think these creatures are persons are “outliers” and that “humans as a group do not see things this way” (p. 281). It then seems feasible that dogs or dolphins could become persons merely by a change in attitude among humans, which seems problematic, as it should be at least partially something inherent in dogs and dolphins that admits them into personhood.


actions: we don’t regret accidents, though we may regret

intentional actions that led to accidents. (E.g., “I regret that

that girder fell on me” doesn’t seem like proper English, but “I

regret getting drunk and walking into the construction site,”

does.)

So regret may be sufficient for conditions 3 and 6. Clearly, if

we regret our past actions then we are applying intentional

predicates to ourselves (otherwise they wouldn’t be actions,

merely movements). What we regret must be actions otherwise we

would feel something other than regret; something like a wish

that some event had not occurred, rather than the sense that we

could have done something differently. So regret is sufficient for

condition 2. Condition 4, reciprocation is obvious here: the

entity regretting is engaging in the stance, and is not merely on

the receiving end of it. An entity that can regret is an entity

that is capable, in some sense, of reciprocating the intentional

stance, since it’s the entity attributing intentionality to

itself.19

19 Kierkegaard’s “the self is that relation that relates itself to itself…”(Kierkegaard, 1980b) could apply here: regret is an existential condition of personhood, and is probably sufficient for existential personhood.


It’s hard to imagine what regret would look like without some

kind of symbolic structure complex enough to represent to

ourselves that we did X, and should have done Y. The occurrence

of a subjunctive certainly implies verbal capacity, condition 5.

And since there’s a weighing of choices, insofar as we look at

some choice and compare it to another, non-instantiated choice,

and judge between them, regret clearly involves rationality,

condition 1, in a minimal sense.

I think regret also clearly meets Frankfurt’s condition of

personhood, in that it implies second order volitions. Regret is

essentially wishing I had chosen some other action, which, while

not forward-looking in the way of Frankfurt’s second-order

volition, is at least a backward-looking second order volition (I

would that my will had been different). Which perhaps implies

that, given another chance, I’d do it differently, and, if that’s

true, then I think it is also a forward-looking second-order

volition. In other words, proper regret involves at least wanting

to do something a certain way in the future, were a certain set

of events or conditions to arise again, and, I assume, it


involves desiring that my will be a certain way (“next time, I’ll

have the courage to tell him off if he says that to me again…”).

Further, it’s clear that the life that contains regret

necessarily and inherently includes the capacity for moral

judgment and the ability to judge one’s actions, something

Frankfurt thinks is definitive of personhood (1971). We don’t

have to like or agree with someone’s moral opinions to hold that

regret involves thinking that things could have been done better,

that in some way, when we regret, we look at ourselves as having

failed to live up to some ideal. So it’s a sort of moral judgment

if Joe regrets not killing Charlie when he had the chance; he at

least is using a higher-order reflection upon some desire. In any

event, in making the judgment in regret we tie ourselves together

in a larger sense; we hold that we have specific ideals, that we

have at least one value, and that that value should be used in

determining our worth across time, insofar as we failed to evince

that value at some time, and believe we should have done

otherwise, which is a sort of minimal, non-judgmental model of

what’s necessary for something to be an ethic.


So regret is sufficient for Dennett/Frankfurt-style personhood.

But I don’t think that Dennett’s list is sufficient for

personhood (it leaves out to much in not noting the necessary

emotional elements of personhood, though I think some are implied

by 3.) Further, though I won’t detail it here, I think regret

meets Charles Taylor’s (1985) criteria in that it implies that we

hold our actions to be significant, it is sufficient for at least

a basic narrative notion of the self (which I cover below,) and

it certainly meets any of the minimal criteria of personhood

which rely on self-awareness and continuity of memory.

So my sense is, while there are several different notions of

personhood, on virtually any of them regret is a sufficient

condition.

5. Identity across time: regret, memory and narrative

Notably, the most popular theories of the continuity of identity

include memory conditions. This makes sense within personhood,

too, in that it’s hard to imagine a person as an entity that

exists only for a moment, with no tie to past or future. We could


have hunger or pain in a moment, but those clearly aren’t

sufficient for personhood. They may be sufficient, as Mary

Midgley notes (1985), for being an entity that deserves moral

consideration; they may be sufficient, according to Galen

Strawson (1999), for being a self. But I think we’d be hard-

pressed to say that an entity with little to no continuity of

identity would really count as a person; I won’t argue that here,

though Marya Schechtman makes a case in “Self-Expression and

Self-Control” (2005) and “Stories, Lives and Basic Survival”

(2007).

Regret’s memory condition is necessarily tied to identity. We

only regret things we ourselves (think ourselves to) have done.20

So in regret, we establish a connection across time. We tie

ourselves to a prior being, and create an extended life. Further,

regret is not just tied to identity, it’s identity-making,

understanding identity in the common sense of “one’s identity”

20 I’m thinking here of what Williams calls “agent-regret” (1981, p. 27). I doubt that in modern English it actually makes sense to say “I regret that Hitler invaded Poland,” so I’m not convinced that we need a separate term for regret that is felt for our own actions. I’ll later note that “our own” can beexpanded to include actions taken by identity groups to which we belong; for example, as a U.S. citizen, I think I can regret that the U.S. invaded Iraq, but only because I can say “I regret that we invaded Iraq.” I think this locution makes sense, or would be acceptable, even though I personally did notinvade Iraq, and would have voted against doing so, had there been a vote.


(as opposed to the bare or strict sense of identity in pure

metaphysics,) in other words, what we identify ourselves as, or

the identity that others would attribute to us, a collection of

personality traits and perhaps a personal history. In that sense,

part of our identity is, at least subjectively, derived from our

recurring thoughts and our attitudes toward what we believe

ourselves to be. Such thoughts and attitudes are person-making;

they constitute, in part, who we are. In regret, we establish an

opinion about some entity that we take to be ourselves, and we

return to and rehearse that opinion, consistently re-tying us to

that self and that attitude toward that self.

It’s important to note that this does not mean that the things

we regret are things we actually did; rather, regret is identity

making because we believe ourselves to have done certain things,

and we have an emotional connection to those beliefs and we

define ourselves partly in relation to those believed events and

our emotional relation to them and the judgment we pass on them.

So, if I truly believed I was Chamberlain, I might spend a great

deal of time regretting the Munich Treaty, and thinking about

what I should have done differently. Regret’s identity-making


aspect doesn’t mean that I then am Chamberlain, but rather that

my personality, the traits I exhibit, my tendencies, my

obsessions, the content of my thoughts, and much of what makes me

who I am, both in my private moments of reflection and (I would

assume) in my interactions with others, comes from this gnawing

regret about the Munich Treaty. Perhaps I am overly cautious not

to appease the orderlies at the mental hospital where I live.

Perhaps each time someone offers a compromise I recoil and then

steel my resolve. Perhaps each time I hear about a Nazi atrocity

I am personally filled with shame and guilt. In other words, my

person, my personality, and the content of my consciousness are

shaped by this regret, even if in fact I did not do the thing I

regret having done.

Now, many contemporary notions of personhood, instead of

Dennett and Frankfurt’s fairly synchronic conditions, call for

narrative as an essential (and perhaps sufficient) condition.

That is, a person is one who lives life with at least some self-

conscious awareness of the narrative quality of experience.


Part of the purpose of this move (I’ll advert mostly to

Schechtman’s conception of the narrative self here (2007; 2010;

2011), as I think it’s among the most sophisticated and

Schechtman has taken account of virtually all other leading

narrative theories) is to tie personhood in to continuity of

identity. If persons are self-narratizers, then persons are

automatically trans-temporal. Regret, it seems, is sufficient for

at least a partial self-narrative.

If I regret doing X then I tie myself into a past self, and see

the actions of that past self as affecting my current self, and

see that action as in some way mine. If I didn’t, I couldn’t

really regret it, at least as I understand regret; I might wish

John and Keisha had never done something, but I’m not sure it

makes sense to say that I regret what John and Keisha did, at

least in this sense of “regret.” I might tell John and Keisha,

“you’ll regret that!” but I won’t say in regard to an action

they’re about to take “I’ll regret that!” (unless it somehow

draws me into an action.)


I also think regret does something that Schechtman, notably

among narrative theorist, does in her version of narrativity

theory, which is tie in a necessary emotional content to the

narrative. I hold hers to be, in many regards, the best of the

narrativity theories because of the way it brings emotion in, not

as a supplement, but as an integral part of what makes the

narrative my narrative.

While Schechtman (2003) rightly emphasizes empathic and

sympathetic connections to the narrative of my former human being

in order for us to be the same person or same self, this is

clearly not a sufficient condition21. One can have empathic and

sympathetic access to a narrative that is not the narrative of my

life; otherwise, as Marcia Eaton notes (1982), I wouldn’t get so

much out of reading certain sorts of stories. I tend to empathize

and sympathize with the heroes and heroines of certain works of

fiction, and of narrative versions of factual accounts.

But I don’t regret the actions of Jane Eyre or John Shade. I

might wish they had acted differently, I might be heartbroken at 21 I had, upon first reading “Empathic access: The missing ingredient in personal identity,” thought that it was claiming that empathic access was a sufficient condition; I’m grateful to Prof. Schechtman for clarifying this point. We agree, I think, that it is a necessary condition for at least a richsense of continuity of identity.


their choices, in fact, if I don’t have a series of interesting

emotional responses to their actions then, at least in some

narrative forms, the writing fails at its goal.

I can only regret actions I believe that I was responsible for.

And I don’t think this is a linguistic oddity explicable by

saying that “regret” just happens to be “wishing some X didn’t

occur” which, by convention, includes in its definition “and that

X was something I did.” That would only give us a circular path

to personhood purely dependent on some linguistic quirk. If

“regret” didn’t exist I could always just invent a word that

meant “an attitude toward one’s own past actions” and this would

be a cheap tactic for establishing personhood.

Rather, I would hold that regret is a very particular emotional

state, that people in general feel, and that is felt only in

regard to one’s own actions. I don’t think it’s a purely

invented, conventional category. Rather, I think there is a

particular emotional state picked out by “regret,” as much as

there is by, say, “grief.” I think regret is as distinct from,

and as similar to, disappointment, as love and pity are to each

other. I won’t make an argument for that here (and I could


certainly be wrong) but I think reflection will show a very

particular character to wishing I hadn’t done something that is

distinct from wishing something hadn’t happened. Probably, this

involves feelings of shame or guilt as well as the desire that

things had been different, and the shame or guilt are part of the

unique character of regret.

6. Regret, identification, alienation, and collective identity

You could conceivably mistakenly regret something that you

didn’t do, but believe you did do. You could believe that someone

else’s action was your own, for example. But a key feature of the

feeling of regret, and what makes it distinct from sorrow over a

past event that you had no control over, is that in regret

there’s a sense of self-blame. It’s one thing to hate someone for

making a decision that affected you negatively, or to forever

wish that some event had not occurred, but the quality of looking

at yourself and thinking that you are responsible for some bad

that happened is distinct. It combines blame and the feeling of


being sorry. It’s as though you wish to apologize to yourself,

and this is what makes it so essential to selfhood.

It’s a confrontation with oneself that both establishes a

connection to the past and sets us at odds with ourselves. In

this way I think it importantly answers to the intuition that we

are our past selves, and to the intuition that we are not our past

selves, as it combines in one moment an attitude toward some

entity, and a receiving of that attitude. Its basic, seeming

contradiction, that I’m mad at me, gets at something that

distinguishes persons from non-persons in a large sense. It

includes self-consciousness, but also the splitting of the self

into subject and object while maintaining the connection between

the two by means of emotion. This is somewhat cognate with

Frankfurt’s (1971) notion of the person, which requires that we

obtain a distance from our own desires and at times treat them as

objects, and even as objects that we wish weren’t parts of

ourselves, but which we accept are parts of ourselves.

Similarly, in regret, I look at an entity I believe to be

myself, but in the past, and feel that I am inescapably that

entity (If I could escape from the identification I wouldn’t feel


regret, but something like, maybe, the redemption that people

feel upon being born again), and also see that entity as behaving

in a way that, in some sense, I think I wouldn’t behave, or at

least as behaving against what I believe are my best judgments

and proper opinions and valuations. So it’s a me that I feel as

other but also as myself.

Bernard Williams points out something important: we can feel

regret for something that, strictly speaking, we cannot be held

legally responsible for, nor, in a strict Kantian sense, morally

responsible for. Williams gives the example of the truck driver

who, “through no fault of his own, runs over a child” (1981, p.

28). The truck driver could (and, according to Williams, should,

if he is fully human) feel regret about this, because “there is

something special about his relation to this happening, something

which cannot merely be eliminated by the consideration that it

was not his fault” (p. 28).

I think this has to do with the fact that his is an event that

is tied in to the truck driver’s identity; it forms part of his

narrative. It’s true, let us say, that it wasn’t his fault, but


if he had done something differently, this wouldn’t have

happened. He has at least some access to a story wherein he could

have made a difference. Further, the very structure of the

narrative implicates him: he was driving the truck. He hit the

child.

This is why I think it is acceptable to have regret toward

actions taken by a group of which you were a part, even if you

yourself would not have authorized such actions nor were directly

responsible. I think it’s reasonable English for a U.S. citizen

to say, “I regret that we invaded Iraq,” even if the citizen in

question had no direct say in the matter.

And this is because a U.S. citizen identifies as a U.S. citizen.

Just as the truck driver must identify as the driver of the

truck, the citizen identifies as a citizen, and then can feel

regret for actions taken on his or her behalf. I imagine that the

shame that many Germans felt after the war was like this: perhaps

some individual could have done nothing to stop the atrocities,

but they were done in that individual’s name.

Still, I think that someone who actively fought against the

Nazis, a true resister, would not feel regret about the atrocities


(no doubt he or she would feel horror, revulsion, etc.), because

such a person would have identified very strongly against the

Nazi regime. She or he would have staked an identity in

opposition, and thus would not feel that a “we” that included him

or her had engaged in the atrocities.

Some of this plays on uses of terms, but I think it points to a

deeper sense of the connection between identity and regret.

7. Conclusion

Regret, then seems sufficient not only for synchronic notions

of personhood, but for those concepts based on narrative or

memory. Further, regret is person-making: it provides us with the

material to become Frankfurtian persons, and to become persons in

the narrative sense. It has the benefit of being an emotional

condition, and a rich emotional condition: it includes desire and

either shame or guilt. It includes a relation between oneself and

oneself that fulfills existential conditions for personhood. It

includes inherently the capacity to evaluate, to make higher-

order judgments, and to direct those judgments toward behavior.


It is self-shaping (and one assumes a self is a key part of

personhood) in that it instills a desire to be different, and it

ties us to a sense of self, a self-story, that we cannot help but

identity with. It works both with highly individualistic notions

of self, and with more communal notions of self, in that we can

regret what was done on our behalf or in our name (indeed,

responsibility often demands that we have just these sorts of

regrets even if we didn’t directly initiate the action we are

regretting.)

Of course, there are many uninteresting sufficient conditions

for personhood. Being me, being an entity that has rights and

obligations under the law, etc. But what makes regret interesting

is both that it could serve as a condition for an entity that we

don’t already know to be a person. If we could determine that our

computers had begun to feel regret, we would have to begin take

them seriously as person-candidates. Further, unlike dread or

anxiety, which had been the standard emotional condition for

personhood given by the existentialists, regret requires a great

deal of higher-order activity and a linguistic-narrative sense of

self. One could imagine a deer feeling dread, but I doubt one


ever felt regret, and if we find that deer do feel regret, I

would hope that we would drastically change our attitudes toward

them.

References

Damasio, A. R. (1999). The feeling of what happens: Body and emotion in the

making of consciousness. New York: Harcourt Brace.

Dennett, D. C. (1976). Conditions of personhood. In A. O. Rorty

(Ed.), The identities of persons (pp. 175-196). Berkeley, CA:

University of California Press.

Eaton, M. M. (1982). A strange kind of sadness. Journal of Aesthetics &

Art Criticism, 41(1), 51-63.


Ekman, P. (1971). Universals and cultural differences in facial

expressions of emotion. Nebraska Symposium on Motivation, 19, 207-

283.

Frankfurt, H. G. (1971). Freedom of the will and the concept of a

person. Journal of Philosophy, 68(1), 5-20.

Hillman, J. (1979). The dream and the underworld. New York: Harper &

Row.

Kierkegaard, S. (1980a). In Thomte R., Anderson A. (Eds.), The

concept of anxiety: A simple psychologically orienting deliberation on the dogmatic

issue of hereditary sin. Princeton, NJ: Princeton University Press.

Kierkegaard, S. (1980b). In Hong H. V., Hong E. H. (Eds.), The

sickness unto death: A christian psychological exposition for upbuilding and

awakening. Princeton, NJ: Princeton University Press.

Mesquita, B., & Frijda, N. H. (1992). Cultural variations in

emotions: A review. Psychological Bulletin, 112(2), 179-204.

Midgley, M. (1985). Persons and non-persons. In P. Singer (Ed.),

In defense of animals (pp. 52-62) Blackwell.

Schechtman, M. (2003). Empathic access: The missing ingredient in

personal identity. In R. Martin, & J. Barresi (Eds.), Personal

identity (pp. 238-259). Malden, MA: Blackwell.


Schechtman, M. (2005). Self-expression and self-control. In G.

Strawson (Ed.), The self? (pp. 45-62). Malden, MA: Blackwell.

Schechtman, M. (2007). Stories, lives, and basic survival: A

refinement and defense of the narrative view. Royal Institute of

Philosophy Supplement, 60, 155-178.

Schechtman, M. (2010). Personhood and the practical. Theoretical

Medicine and Bioethics, 31(4), 271-283.

Schechtman, M. (2011). Memory and identity. Philosophical Studies,

153(1), 65-79.

Strawson, G. (1999). The self and the SESMET. Journal of Consciousness

Studies, 6(4), 99-135.

Taylor, C (1985). The Concept of a Person, Philosophical Papers. Volume

1. Cambridge: Cambridge University Press, 98-102

Williams, B. (1981). Moral luck. Moral luck: Philosophical papers, 1973-

1980 (pp. 20-39). Cambridge, England: Cambridge University

Press.

Regret as a sufficient condition of personhood; new version.

Documents