EVELYN PLUHAR

Permsylvania State University

PHILOSOPHY -BEI'WEEN THE SPECIES 32

qUi .4 It 441 itt $

EVELYN PLUHAR

Permsylvania State University

PHILOSOPHY -BEI'WEEN THE SPECIES 32

qUi .4 It 441 itt $

iii mt'tt rD"'; TN '. at each follows.

Editors' Note: The following paper was presented at the Pacific Division meeting

of the Society for the Study of Ethics and Animals in san Francisco, March, 1987.

You and I are noral agents. We are capable of understanding and acting upon noral principles. [1] Provided that we do not

act under duress, we are responsible for what we do. As noral agents, we alone have noral obligations and can be held accountable for

flouting those obligations. All noral codes

are addressed to us. So are the following questions: ,Are we, as noral agents, all

~;rally considerable; i.e., are others (also noral agents) directly obligated to take our

interests into account when their actions would affect us? Are we all equally norally significant, entitled to be treated not, mere­

ly as means to further others' purposes?

Can a\1..~·' beings who are not nora! agents be norally considerable; i.e., are there any noral patients? [2] If there are noral pa­

tients, are they as norally significaat as we are? Are some noral patients more nora11y significant than others? These are all fun­damental, extraordinarily important ques­tions. They are also extraordinarily diffi ­cult to answer.

salle Candidates for Moral Patiency

Let us begin by considering some candi­dates for noral patiency. Those nost under discussion have been (1) agents who are not noral agents, (2) self-conscious beings, (3)

conscious beings, (4 ) living beings, or (in­creasingly under discussion (5) natural ob­

jects or systems. Philosophers have argued. that norally relevant similarities between beings in a given category and noral agents (who nost assume are norally considerable) justify the ascription of rights or rroral standing to thOSE; beings. Space limitations

prevent me frOOI considering the last two mentioned categories (living beings and na­

tural objects or systems) here. This limita­

tion is in no way meant to suggest that such

beings could not be norally considerable. We

will have quite enough complications if we

confine our attention to the first three

candidates for noral patiency. A brief look

33

(1) Agents need not be noral agents. Ckle may act purposefully without being capa­ble of comprehending noral principles. Be­

sides "agents," beings in this category can be called (like noral agents) "persons" in Joel Feinberg's sense and "subjects-of-lives"

in Tam Regan's sense. According to Feinberg, "In the commonsense way of thinking, persons are those beings who, arrong other things, are

conscious, have a concept and awareness of themselves, are capable of experiencing emo­

tions, can plan ahead, can act on their

plans, and can feel pleasure and pain."[3] "Person" is a notoriously slippery term.

Same use it interchangeably with "human" or mean by it no more than "norally considera­

ble. " Same would restrict it to noral agents. By contrast, I find Feinberg's fonn­ulation admirable, and whenever persons are referred to in the remainder of this paper, it will be in this sense only. However, the ambiguities and distracting emotive content one often finds in discussions of persons may well have led Regan to re-name the category: "individuals are subjects-of-a-life if they have beliefs and desires; perceptions, meroo­

ry, and a sense of the future, including

their own future; an emotional life together with feelings of pleasure and pain; prefe­

rence- and welfare-interests; the ability to initiate action in pursuit of their desires and goals; a psychophysical identity over time; and an individual welfare in the sense

that their experiential lives fare well or ill for them, logically independently of

their utility for others and logically inde­pendently of their being the object of anyone else's interests. [4] If there are noral

patients, these beings appear to be strong

candidates for that position.

(2) Other beings have lesser abilities.

Can a self-conscious being fail to be an agent, person, or subject-of-a-life? Yes, if that being's ability to act to fulfill goals is severely diminished or non-existent. The being may be physically or mentally incapable of agency but, have some awareness (perhaps rudimentary at best) of self. Possibly, such beings are noral patients. Many would deny

this.

(3) Beings who can be aware, in sane

sense, of their surroundings but not of them­

selves are often called "rroerely" conscious.

BEI'WEEN THE SPECIES

1 7"t'5t re'; TN -'It tt&

at each follows.

Editors' Note: The following pa.per waspresented at the Pacific Division meeting

of the Society for the Study of Ethicsand Animals in san Francisco, March,1987.

You and I are IlPral agents. We arecapable of understanding and acting uponIlPral principles. [1] Provided that we do not

act under duress, we are responsible for whatwe do. As IlPral agents, we alone have IlPralobligations and can be held accountable for

flouting those obligations. All IlPral codes

are addressed to us. So are the followingquestions: ,Are we, as IlPral agents, all

~;ra.lly considerable; i.e., are others (alsoIlPral agents) directly obligated to take our

interests into account when their actionswould affect us? Are we all equally IlPrallysignificant, entitled to be treated not. mere­

ly as means to further others' purposes?

Can an<~" beings who are not IlPral agents beIlPrally considerable; i.e., are there anyIlPral pa.tients?[2] If there are IlPral pa­

tients, are they as norally significaat as weare? Are some IlPral pa.tients more nora11ysignificant than others? These are all fun­damental, extraordinarily important ques­tions. They are also extraordinarily diffi­

cult to answer.

Some Candidates for Moral Patiency

Let us begin by considering some candi­dates for noral pa.tiency. Those nost underdiscussion have been (I) agents who are notnoral agents, (2) self-conscious beings, (3)

conscious beings, (4 ) living beings, or (in­creasingly under discussion (5) natural ob­

jects or systems. Philosophers have argued,that norally relevant s:imilarities between

beings in a given category and IlPral agents(who nost assume are IlPrally considerable)justify the ascription of rights or rroralstanding to thOSE; beings. Spa.ce l:imitations

prevent me frOOI considering the last twomentioned categories (living beings and na­tural objects or systems) here. This l:imita­

tion is in no way meant to suggest that such

beings could not be rrorally considerable. We

will have quite enough complications if weconfine our attention to the first three

candidates for noral pa.tiency. A brief look

33

(1) Agents need not be noral agents.Ckle may act purposefully without being capa.­ble of comprehending noral principles. Be­

sides "agents," beings in this category canbe called (like rroral agents) "persons" in

Joel Feinberg's sense and "subjects-of-lives"

in Tam Regan's sense. According to Feinberg,"In the commonsense way of thinking, personsare those beings who, arrong other things, areconscious, have a concept and awareness ofthemselves, are capable of experiencing erro­

tions, can plan ahead, can act on their

plans, and can feel pleasure and pain."[3]"Person" is a notoriously slippery term.

Same use it interchangeably with "human" ormean by it no more than "rrorally considera­

ble. " Same would restrict it to noralagents. By contrast, I find Feinberg's fonn­ulation admirable, and whenever persons arereferred to in the remainder of this pa.per,it will be in this sense only. However, theambiguities and distracting errotive contentone often finds in discussions of persons maywell have led Regan to re-name the category:"individuals are subjects-of-a-life if theyhave beliefs and desires; perceptions, merro­

ry, and a sense of the future, including

their own future; an errotional life togetherwith feelings of pleasure and pa.in; prefe­

rence- and welfare-interests; the ability toinitiate action in pursuit of their desiresand goals; a psychophysical identity overtime; and an individual welfare in the sense

that their experiential lives fare well orill for them, logically independently of

their utility for others and logically inde­pendently of their being the object of anyone

else's interests. [4] If there are rroralpatients, these beings appear to be strong

candidates for that position.

(2) other beings have lesser abilities.

Can a self-conscious being fail to be anagent, person, or subject-of-a-life? Yes, ifthat being's ability to act to fulfill goalsis severely diminished or non-existent. Thebeing may be physically or mentally incapableof agency but, have some awareness (perhapsrudimentary at best) of self. Possibly, such

beings are rroral pa.tients. Many would deny

this.

(3) Beings who can be aware, in sane

sense, of their surroundings but not of them­

selves are often called "merely" conscious.

BEI'WEEN THE SPECIES

-----

They too might be moral patients. The legi­timacy of this moral patiency candidate is

far more hotly disputed than the two others mentioned above. The further we depart from the qualities which characterize moral agen­

cy, the more contestable our candidates for

moral patiency becx:me. (Of course, one must not allow the relative p:Jpularity or unp:JpU­larity of our candidates to· decide the issue. It may not be coincidental that we see our­selves as the paradigms of moral considera­bility. One's decisions about who or what

"measures up" morally may be far from objec­tive. )

'!he categories above plainly cut across species lines. No one denies that there are

hillllaIls who are "merely" conscious, but it is also true that some agents are not hillllaIl. Agency as such requires self-consciousness,

desires, and some degree of reasoning abili­

ty; it does not require heights of intellec­tual sophistication in the way moral agency

does. For example, we do not require agents

to be skilled in deductive reasoning (let alone predicate logic), but we do require

some inductive abilities. Again, one can plan for the next five minutes, for the next hour, for the next week, or for years in advance and be an agent. You don't have to

plan for the next fifty years in order to be an agent. Some understanding is surely re­

quired for agency; the ability to read The New York Times is not. There is no reason to think that the skills required for purposeful action are restricted to members of the human species, and ample reason to think they are not. Recent ethological studies sUpp:Jrt this conclusion. (5]

Nevertheless, some humans continue to

express considerable skepticism about the p:Jssibility of nonhillllaIl agency. For example, the Cartesian view that thought requires

linguistic ability has been revived in a

highly sophisticated form by R. G. Frey. (6] He denies that nonhillllaIl animals can have

desires (and thus that they can act purpose­fully) on the grounds that beliefs are re­

quired for desires and linguistic ability is

required for beliefs (an ability he assumes nonhuman animals lack). This view has been

skillfully refuted by others on a number of grounds, (7] so I will not pursue it further

here, except to make one observation. Those who, like Descartes (but unlike Frey), refuse

to extend their skepticism about nonhuman

mentality to hillllaIl beings, would have us accept the following rather dubious contrast: a screaming, struggling child being dragged to the doctor for her second shot has beliefs about what is in store for her and desires to stay home instead, but a yelling, thrashing cat on the way to the vet for his second

visit is just exercising his limbs and vocal cords!

In addition to double-standard thinking,

one also finds the fallacy of false dilermna

cormnitted by some critics of nonhillllaIl agency.

For example, Michael A. Fox has said that animals cannot lead lives or be autonomous

because autonomy requires that one can "gene­rate a life plan" to guide one's life as a whole. (8] But surely this is to conflate

"highly autonomous" with "autonomous." No distinction is made between the purposeful agent who acts with a life plan in view and the purposeful agent who has shorter-range plans. As Regan has argued, many nonhuman animals may lack the high level of autonomy required for moral agency, but they do exhi­bit "preference autonomy."(9] '!his descrip­

tion fits many young and "defective" humans as well. All are agents.

Sad!y, there are other hillllaIls who do not

even have preference autonomy. Conscious hillllaIls exist who cannot act to satisfy very fundamental needs. Although at one p:Jint Regan claims that such hillllaIls are "subjects­of-lives,"(lO] they are not, since they rail to satisfy (at the very least) his action

requirement.

It is imp:Jrtant to remember these reali­

ties about humans and nonhillllaIls as we consi­

der abstract arguments about moral patiency.

But how do we even begin to determine whether

there are moral patients (i.e., beings who are morally considerable without being moral agents)? What characteristics must a being p:Jssess to be morally considerable? How can we defend the claim that, e.g., "mere" agency

is sufficient for moral considerability but

that "mere" consciousness is not? Or that

only moral agency suffices for moral consi­

derability? And how can we argue about de­

grees, if any, of moral significance? What would even count as a resolution of these problems?

BE'I'WEEN THE SPECIES 34

4 U u"~ 1 a Btu.1it J

They too might be moral patients. The legi­timacy of this moral patiency candidate is

far more hotly disputed than the two othersmentioned above. The further we depart fromthe qualities which characterize moral agen­

cy, the more contestable our candidates for

moral patiency becx:me. (Of course, one mustnot allow the relative p:Jpularity or unp:JpU­larity of our candidates to· decide the issue.It may not be coincidental that we see our­selves as the paradigms of moral considera­bility. One's decisions about who or what

"measures up" morally may be far from objec­tive. )

'!he categories above plainly cut acrossspecies lines. No one denies that there are

hillllaIls who are "merely" conscious, but it isalso true that some agents are not hillllaIl.Agency as such requires self-consciousness,

desires, and some degree of reasoning abili­

ty; it does not require heights of intellec­tual sophistication in the way moral agency

does. For example, we do not require agents

to be skilled in deductive reasoning (letalone predicate logic), but we do require

some inductive abilities. Again, one canplan for the next five minutes, for the nexthour, for the next week, or for years inadvance and be an agent. You don't have to

plan for the next fifty years in order to bean agent. Some understanding is surely re­

quired for agency; the ability to read TheNew York Times is not. There is no reason tothink that the skills required for purposefulaction are restricted to members of the humanspecies, and ample reason to think they arenot. Recent ethological studies sUpp:Jrt thisconclusion. (5]

Nevertheless, some humans continue to

express considerable skepticism about thep:Jssibility of nonhillllaIl agency. For example,the Cartesian view that thought requires

linguistic ability has been revived in a

highly sophisticated form by R. G. Frey. (6]He denies that nonhillllaIl animals can have

desires (and thus that they can act purpose­fully) on the grounds that beliefs are re­

quired for desires and linguistic ability is

required for beliefs (an ability he assumesnonhuman animals lack). This view has been

skillfully refuted by others on a number ofgrounds, (7] so I will not pursue it further

here, except to make one observation. Thosewho, like Descartes (but unlike Frey), refuse

to extend their skepticism about nonhuman

BE'I'WEEN THE SPECIES 34

mentality to hillllaIl beings, would have usaccept the following rather dubious contrast:a screaming, struggling child being draggedto the doctor for her second shot has beliefsabout what is in store for her and desires tostay home instead, but a yelling, thrashingcat on the way to the vet for his second

visit is just exercising his limbs and vocalcords!

In addition to double-standard thinking,

one also finds the fallacy of false dilermna

cormnitted by some critics of nonhillllaIl agency.

For example, Michael A. Fox has said thatanimals cannot lead lives or be autonomous-----because autonomy requires that one can "gene-rate a life plan" to guide one's life as awhole. (8] But surely this is to conflate

"highly autonomous" with "autonomous." Nodistinction is made between the purposefulagent who acts with a life plan in view andthe purposeful agent who has shorter-rangeplans. As Regan has argued, many nonhumananimals may lack the high level of autonomyrequired for moral agency, but they do exhi­bit "preference autonomy."(9] '!his descrip­

tion fits many young and "defective" humansas well. All are agents.

Sad!y, there are other hillllaIls who do not

even have preference autonomy. ConscioushillllaIls exist who cannot act to satisfy veryfundamental needs. Although at one p:JintRegan claims that such hillllaIls are "subjects­of-lives,"(lO] they are not, since they railto satisfy (at the very least) his action

requirement.

It is imp:Jrtant to remember these reali­

ties about humans and nonhillllaIls as we consi­

der abstract arguments about moral patiency.

But how do we even begin to determine whether

there are moral patients (i.e., beings whoare morally considerable without being moralagents)? What characteristics must a beingp:Jssess to be morally considerable? How canwe defend the claim that, e.g., "mere" agency

is sufficient for moral considerability but

that "mere" consciousness is not? Or that

only moral agency suffices for moral consi­

derability? And how can we argue about de­

grees, if any, of moral significance? Whatwould even count as a resolution of theseproblems?

4 U u"~ 1 a Btu.1it J

-- --

Rights Views, Perfectionism, and utilitarianism

We IlIUst first identify the structural features of different types of views on moral considerability and significance. I will be distinguishing rights views from perfection­istic and utilitarian views.

Rights views all propose characteristics which are claimed to be sufficient (and. pos­sibly also necessary)for moral considerabili­ty and. maximum moral significance. [11] "Rights" are spoken of here in Feinberg's sense of "valid claims." [12] It is basic moral rights which are relevant here, the rights to life, liberty, and. well-being. On

any of these views, morally considerable beings are due treatment which is cormnensur­

ate with their moral significance. Maximally morally significant beings are to be treated as ends in themselves rather than mere means. Less significant morally considerable beings, if any, may have their rights overridden in favor of those who are more morally signifi­cant. It is important to bear in mind that whatever characteristic (X) a given rights view proposes as sufficient, moral considera­bility is really a cluster of characteristics (t, u. v, ••• ). For example, if moral agency is the proposed characteristic, being a ~

son (which is itself a cluster of character­istics) who is capable of understanding and acting in accordance with moral principles would be the relevant constitutive features. Brief characterizations of the different types of rights views, along with convenient names for them, appear below.

1. The Total View: Being an X is suffi­cient and necessary for moral considerabili­ty. All X's are equally morally significant and. possess the full range of rights. Ac­cording to this view, being merely t or u or v or ••• , cannot be sufficient for moral

considerability. E.g., one might claim that being ~ moral agent is necessary and suffi­cient for moral considerability. Being a

person or agent who is not a moral agent

would not qualify one for moral considerabil­ity (although moral agents might still have indirect duties to one).

2. The Partial View: Being an X is sufficient for moral considerability and necessary for the whole range of rights.

Being t or u or v or ••• is sufficient for

moral considerability and. sufficient for restricted rights. E.g., one could claim that moral agents are due the full range of rights but that being a "mere" agent is suf­ficient for moral considerability and. only some rights. Concerning moral significance, partial views fall into three sub-categories:

a. The Equality View: All those who are morally considerable are equally morally significant. E.g., moral agents, "mere" agents, and. the "merely" self-conscious could be claimed to all be ends in themselves, although all would not have the same rights (such as the right to liberty).

b. The Additive View: Being t or u or v or ••• is sufficient for moral considerabili­ty. However, the more such characteristics one has, the more morally significant one becanes. Those who are maximally morally significant have all the relevant character­istics; i.e., they are X. E.g., the "merely" self-conscious may be claimed to be less morally significant than "mere" agents, who

are in turn less morally significant than moral agents.

c. The Combination View: A sub-cluster of the characteristics constitutive of being X are sufficient for moral considerability and equal moral significance with respect to X's. Possessing only some of the charac.1;er­istics of the sub-cluster is sufficient for moral considerability but results in a lesser degree of moral significance. E.G., one could claim that being a moral agent (X in this case) makes one maximally morally signi­ficant and that being a subject-of-a-life who

is 'not a moral agent is also sufficient for maximum moral significance, but that being a self-conscious nonagent or being "merely" conscious results in a lesser degree of moral significance.

No'N, I IlIUst introduce a further canpli-

Caro::' Belolln.er G,atton, Old-Faahlonl!d JIt.:llm/ll ~.

I'ev !o.rk:' Oo"e.l:, 1987 . . 35 BEIWEEN THE SPECIES

Rights Views, Perfectionism,and utilitarianism

We must first identify the structuralfeatures of different types of views on moralconsiderability and significance. I will bedistinguishing rights views from perfection­istic and utilitarian views.

Rights views all propose characteristicswhich are claimed to be sufficient (and pos­sibly also necessary)for moral considerabili­ty and maximum moral significance. [11]"Rights" are spoken of here in Feinberg'ssense of "valid claims." [12] It is basicmoral rights which are relevant here, therights to life, liberty, and well-being. On

any of these views, morally considerablebeings are due treatment which is cormnensur­

ate with their moral significance. Maximallymorally significant beings are to be treatedas ends in themselves rather than mere means.Less significant morally considerable beings,if any, may have their rights overridden infavor of those who are more morally signifi­cant. It is important to bear in mind thatwhatever characteristic (X) a given rightsview proposes as sufficient, moral considera­bility is really a cluster of characteristics(t, u. v, ••• ). For example, if moral agencyis the proposed characteristic, being a per­son (which is itself a cluster of character­istics) who is capable of understanding andacting in accordance with moral principleswould be the relevant constitutive features.Brief characterizations of the differenttypes of rights views, along with convenientnames for them, appear below.

1. The Total View: Being an X is suffi­cient and necessary for moral considerabili­ty. All XiS are equally morally significantand possess the full range of rights. Ac­cording to this view, being merely t or u orv or ••• , cannot be sufficient for moral

considerability. E.g., one might claim thatbeing ~ moral agent is necessary and suffi­cient for moral considerability. Being a

person or agent who is not a moral agent

would not qualify one for moral considerabil­ity (although moral agents might still haveindirect duties to one).

2. The Partial View: Being an X issufficient for moral considerability andnecessary for the whole range of rights.

Being t or u or v or ••• is sufficient for

35

moral considerability and sufficient forrestricted rights. E.g., one could claimthat moral agents are due the full range ofrights but that being a "mere" agent is suf­ficient for moral considerability and onlysome rights. Concerning moral significance,partial views fall into three sub-categories:

a. The Equality View: All those who aremorally considerable are equally morallysignificant. E.g., moral agents, "mere"agents, and the "merely" self-conscious couldbe claimed to all be ends in themselves,although all would not have the same rights(such as the right to liberty).

b. The Additive View: Being t or u or vor ••• is sufficient for moral considerabili­ty. However, the more such characteristicsone has, the more morally significant onebecomes. Those who are maximally morallysignificant have all the relevant character­istics; i.e., they are X. E.g., the "merely"self-conscious may be claimed to be lessmorally significant than "mere" agents, who

are in turn less morally significant thanmoral agents.

c. The Combination View: A sub-clusterof the characteristics constitutive of beingX are sufficient for moral considerabilityand equal moral significance with respect toXiS. Possessing only some of the charac.:ter­

istics of the sub-cluster is sufficient formoral considerability but results in a lesserdegree of moral significance. E.G., onecould claim that being a moral agent (X inthis case) makes one maximally morally signi­ficant and that being a subject-of-a-life who

is 'not a moral agent is also sufficient formaximum moral significance, but that being aself-conscious nonagent or being "merely"conscious results in a lesser degree of moralsignificance.

Now, I must introduce a further canpli-

Caro::' Belolln.er G,atton.Old-Faahlonl!d JIt.:l.lm/ll ~.I'ev !o.rk:' Oo"e.l:, 1987. .

BEIWEEN THE SPECIES

cation in my characterization of types of rights views. The total and partial views can each be IlDdified to allow potential X' s or those who are potentially t or u, etc., to be roorally considerable. Potential roorally counts just as much as actuality on what Feinberg calls "The strict Potentiality Cri­terion. " According to "The Gradualist Poten­tiality Criterion," the closer one comes to being the specified sort of being (i.e., the

roore one's potential is actualized), the roore roorally significant one becomes. [13] The different rights views could be IlDdified in either way, [14] although the gradualist ap­proach is roore plausible. Whatever is deci­ded, we must be careful to distinguish being a potential X (or t or u, etc.) fran being a near X (or t or u, etc.) who can never become X{or t or u, etc). On the IlDdified total view, potential XiS are roorally considerable, but near X' s who are not potential X's are not, even if they have the same actual char­

acteristics. On the IlDdified partial view, being t or u, etc., but falling short of being X, is sufficient for rooral considera­bility whether or not one is a potential X.

These rights views, IlDdified or unroodi­fied, are to be distinguished sharply fran perfectionism. According to this view, pos­sessing a given characteristic (e.g., intel­ligence) is sufficient for some degree of rooral significance but not for maximum rooral significance. One's rooral significance is claimed to increase as the degree to which one possesses the favored characteristic increases. Nietzsche is the roost famous advocate of this view.

Perfectionism and the rights views must also be distinguished fran utilitarianism (nothing hinges here on the distinction be­tween rule- and act-utilitarianism). Unlike the rights views, utilitarianism is opposed to rights (although some utilitarians toler­ate a watered-down notion of rights); unlike perfectionism, it awards no increasing rooral significance to higher degrees of intelli­gence or other favored characteristics. In

an important respect, utilitarianism can be construed as denying that one has direct duties to any individual. One's duty on classical utilitarianism is to prcm:>te happi­ness as such; insofar as it is individuals who are the receptacles of happiness, one has

indirect duties to them. Nonhedonistic ver­sions have the same implication. [15]

Regan's Defense of Moral Patiency

These alternatives are radically differ­ent. Just how difficult it is to defem one's choice among them is well illustrated by the work of Tom Regan, one of the roost skillful proponents of noral patiency. The Case for Animal Rights is a magnificent achievement~owever, Regan's attack on certai.n views of rooral considerability and significance leaves some key objections un­forestalled. After raising these objections, I will offer a possible solution which has been inspired by the work of Alan Gewirth.

Regan defends a rights view. He agrees that rooral agency is sufficient for noral considerability and basic rooral rights but denies that it is necessary. Those humans and nonhurnans who, like moral agents, are subjects-of-lives but are not rooral agents, are moral patients. They are roorally consi­derable and just as roorally significant as rooral agents. He does not insist that being a subject-of-a-life is necessary for rooral considerability, although he suspects that it may be. He seriously doubts that any who are not subjects-of-lives could be as roorally significant as those who are. [16] Thus, Regan proposes a canbination rights view: being X,· where X refers to being a moral agent, is sufficient for rooral considerabili­ty and maximum moral significance, but a sub­cluster of the characteristics which consti­tute rooral agency, viz., being the subject­of-a-life, is also sufficient for rooral con­siderability and maximum rooral significance. Those who don't have all the characteristics in the sub-cluster (e.g., the "merely" con­scious) may be roorally considerable but not as roorally significant. Regan's strategy is to reject alternative views, to postulate the equal inherent value of rooral agents and certain others, and to propose "being the subject-of-a-life" as the relevant moral similarity between rooral agents and these others. They are due equal respect, he says.

How does one test alternative views on rooral considerability and significance? Re­gan argues that we require such a view to display adequacy of scope, precision, and

conf;rmity with our reflective intuitions. The last test is the critical one. It re­quires that we judge the view coolly, ration­ally, clearly, with ·as much infonnaqon as we can gather, and that we take an impartial

BE'IWEEN THE SPECIES 36

M : as: SA S;. iSt $.''M

cation in my characterization of types ofrights views. The total and partial viewscan each be IlDdified to allow potential X' sor those who are potentially t or u, etc., tobe roorally considerable. Potential roorallycounts just as much as actuality on whatFeinberg calls "The strict Potentiality Cri­terion. " According to "The Gradualist Poten­tiality Criterion," the closer one comes tobeing the specified sort of being (i.e., the

roore one's potential is actualized), the rooreroorally significant one becomes. [13] Thedifferent rights views could be IlDdified ineither way, [14] although the gradualist ap­proach is roore plausible. Whatever is deci­ded, we must be careful to distinguish beinga potential X (or t or u, etc.) fran being anear X (or t or u, etc.) who can never becomeX{or t or u, etc). On the IlDdified totalview, potential XiS are roorally considerable,but near X' s who are not potential X's arenot, even if they have the same actual char­

acteristics. On the IlDdified partial view,being t or u, etc., but falling short ofbeing X, is sufficient for rooral considera­bility whether or not one is a potential X.

These rights views, IlDdified or unroodi­fied, are to be distinguished sharply franperfectionism. According to this view, pos­sessing a given characteristic (e.g., intel­ligence) is sufficient for some degree ofrooral significance but not for maximum rooralsignificance. One's rooral significance isclaimed to increase as the degree to whichone possesses the favored characteristicincreases. Nietzsche is the roost famousadvocate of this view.

Perfectionism and the rights views mustalso be distinguished fran utilitarianism(nothing hinges here on the distinction be­tween rule- and act-utilitarianism). Unlikethe rights views, utilitarianism is opposedto rights (although some utilitarians toler­ate a watered-down notion of rights); unlikeperfectionism, it awards no increasing rooralsignificance to higher degrees of intelli­gence or other favored characteristics. In

an important respect, utilitarianism can beconstrued as denying that one has directduties to any individual. One's duty onclassical utilitarianism is to prcm:>te happi­ness as such; insofar as it is individualswho are the receptacles of happiness, one has

indirect duties to them. Nonhedonistic ver­sions have the same implication. [15]

BE'IWEEN THE SPECIES 36

Regan's Defense of Moral Patiency

These alternatives are radically differ­ent. Just how difficult it is to defemone's choice among them is well illustratedby the work of Tom Regan, one of the roostskillful proponents of noral patiency. TheCase for Animal Rights is a magnificentachievement~owever, Regan's attack oncertai.n views of rooral considerability andsignificance leaves some key objections un­forestalled. After raising these objections,I will offer a possible solution which hasbeen inspired by the work of Alan Gewirth.

Regan defends a rights view. He agreesthat rooral agency is sufficient for noralconsiderability and basic rooral rights butdenies that it is necessary. Those humansand nonhurnans who, like moral agents, aresubjects-of-lives but are not rooral agents,are moral patients. They are roorally consi­derable and just as roorally significant asrooral agents. He does not insist that beinga subject-of-a-life is necessary for rooralconsiderability, although he suspects that itmay be. He seriously doubts that any who arenot subjects-of-lives could be as roorallysignificant as those who are. [16] Thus,Regan proposes a canbination rights view:being X,· where X refers to being a moralagent, is sufficient for rooral considerabili­ty and maximum moral significance, but a sub­cluster of the characteristics which consti­tute rooral agency, viz., being the subject­of-a-life, is also sufficient for rooral con­siderability and maximum rooral significance.Those who don't have all the characteristicsin the sub-cluster (e.g., the "merely" con­scious) may be roorally considerable but notas roorally significant. Regan's strategy isto reject alternative views, to postulate theequal inherent value of rooral agents andcertain others, and to propose "being thesubject-of-a-life" as the relevant moralsimilarity between rooral agents and theseothers. They are due equal respect, he says.

How does one test alternative views onrooral considerability and significance? Re­gan argues that we require such a view todisplay adequacy of scope, precision, and

conf;rmity with our reflective intuitions.The last test is the critical one. It re­quires that we judge the view coolly, ration­ally, clearly, with ·as much infonnaqon as wecan gather, and that we take an impartial

M : as: SA S;. iSt $. I'M

"1iI1I -iIIiIlII - .*.. WJ:i!£1

attitude. Regan refers to impartiality in this context as "the fonnal principle of justice. " It enjoins us to treat similar cases similarly and dissimilar cases dissimi­larly. [17] Regan opposes views which deny moral considerability to those who are not moral agents or which assign them lesser moral significance. According to him, how do these views fail to be adequate?

Here is one important argument: "Nor

can we avoid recognizing that moral patients fall within the [respect] principle's scope

on the grounds that they have no inherent value or less inherent value than moral agents; this will not do because attempts to disenfranchise moral patients in this way will lay the groundwork for a perfectionist theory of justice, a theory that will either

sanction unjust treatment of some moral agents or avoid this--but only at the price of arbitrariness. [18] Regan is saying that those who favor moral agents, exclusively or to a greater degree, are either consistent or inconsistent perfectionists. Inconsistent perfectionists refuse to discriminate against moral agents who are less intelligent, re­flective, hapPy, etc., than oth¥ moral agents, but those who fall short of being moral agents are morally penalized for their lesser abilities. [19] Consistent perfection­ists, on the other hand, will have to assign lesser and greater moral significance to moral agents too, depending on their abili­ties. Such a view fails to provide an ade­quate interpretation of justice, Regan says.

Regan is right in holding that perfec­tionism would serve very -ill as a basis for a view which postulates the moral primacy of moral agents. Obviously, the inconsistent version fails the test of rationality. But two rrajor problems remain. First, the con­

sistent version needs to be shown to be in­adequat~. Second, Regan's criticisms do not touch the nonperfectionistic alternatives, most importantly the total view and the addi­

tive view.

Concerning the first problem, Regan does

offer an argument against the consistent version of perfectionism, certainly an impor­tant canpetitor of his rights view. He ar­gues that this view must be rejected as un­just because it bases moral significance on the presence or absence of abilities over

whose aa:.IUisition one has no control. Whe­

ther, and to what degree, one is intelligent, skillful, etc., depends on "the natural lot­

tery. " Those who come up short don't deserve to have lesser moral significance any more than those at the other end of the ability scale deserve more. [20] Such treatment is radically unfair.

This objection won't do, however. First, if Regan is arguing that moral consi­derability and significance must not be based on characteristics over which one has no control, his own view is in trouble. We do not choose to be subjects-of-lives any more than an anoeba chooses not to be one. I don't think Regan would want to say that there is no moral distinction between the

two. second, while I agree with Regan that perfectionism is unjust, at this point how is one to argue for this without begging the question? Perfectionists would claim that their view passes Regan's tests, including the test of impartial reflection. They cer­tainly do treat similar cases similarly and

dissimilar cases dissimilarly, given their criteria of moral significance. How can they be answered?

Moreover, what about rights views op­

posed to Regan's own which are not reducible to perfectionism? The total version of the prirracy of moral agency view holds that being a moral agent is necessary and sufficient for

moral considerability. Those "disenfran­chised" fran the moral canmunity are excluded because they are not moral agents, not be­Cause they are "lesser" moral agents. Simi­larly, the additive version of the primacy of moral agency view holds that the different characteristics which make one a moral agent

are each morally significant and that the more of these characteristics one possesses the more morally significant one is. One's status has nothing to -do with the ~ to which one possesses a given morally relevant characteristic. Is there a way to show these views wrong?

Regan does indeed offer arguments which

would apply to these views. He does an ex­cellent job arguing against particular moral agency views such as those held by and

Rawls. [21] The trouble is that these views could be amended in the light of several of his criticisms; he needs objections so funda­mental that ~ primacy of moral agency view would be subject to them. He presses just

37 BmWEEN THE SPOCIES

iT

attitude. Regan refers to impartiality inthis context as "the fonnal principle ofjustice. " It enjoins us to treat similarcases similarly and dissimilar cases dissimi­larly.[l7] Regan opposes views which denymoral considerability to those who are notmoral agents or which assign them lessermoral significance. According to him, how dothese views fail to be adequate?

Here is one important argument: "Nor

can we avoid recognizing that moral patientsfall within the [respect] principle's scopeon the grounds that they have no inherentvalue or less inherent value than moralagents; this will not do because attempts todisenfranchise moral patients in this waywill lay the groundwork for a perfectionisttheory of justice, a theory that will either

sanction unjust treatment of some moralagents or avoid this--but only at the priceof arbitrariness. [18] Regan is saying thatthose who favor moral agents, exclusively orto a greater degree, are either consistent orinconsistent perfectionists. Inconsistentperfectionists refuse to discriminate againstmoral agents who are less intelligent, re­flective, hapPy, etc., than oth¥ moralagents, but those who fall short of beingmoral agents are morally penalized for theirlesser abilities. [19] Consistent perfection­ists, on the other hand, will have to assignlesser and greater moral significance tomoral agents too, depending on their abili­ties. Such a view fails to provide an ade­quate interpretation of justice, Regan says.

Regan is right in holding that perfec­

tionism would serve very· ill as a basis fora view which postulates the moral primacy ofmoral agents. Obviously, the inconsistentversion fails the test of rationality. Buttwo major problems remain. First, the con­

sistent version needs to be shCMIl to be in­adequate. Second, Regan's criticisms do nottouch the nonperfectionistic alternatives,most importantly the total view and the addi­

tive view.

Concerning the first problem, Regan does

offer an argument against the consistentversion of perfectionism, certainly an impor­tant canpetitor of his rights view. He ar­gues that this view must be rejected as un­just because it bases moral significance onthe presence or absence of abilities over

whose aa;ruisition one has no control. Whe-

37

ther, and to what degree, one is intelligent,skillful, etc., depends on "the natural lot­tery. " Those who come up short don't deserve

to have lesser moral significance any morethan those at the other end of the abilityscale deserve more. [20] Such treatment isradically unfair.

This objection won't do, however.First, if Regan is arguing that moral consi­derability and significance must not be basedon characteristics over which one has nocontrol, his CMIl view is in trouble. We donot choose to be subjects-of-lives any morethan an anoeba chooses not to be one. Idon't think Regan would want to say thatthere is no moral distinction between the

two. Second, while I agree with Regan thatperfectionism is unjust, ~ this point how isone to argue for this without begging thequestion? Perfectionists would claim thattheir view passes Regan's tests, includingthe test of impartial reflection. They cer­tainly do treat similar cases similarly and

dissimilar cases dissimilarly, given theircriteria of moral significance. How can theybe answered?

Moreover, what about rights views op­

posed to Regan's CMIl which are not reducibleto perfectionism? The total version of theprimacy of moral agency view holds that beinga moral agent is necessary and sufficient formoral considerability. Those "disenfran­chised" fran the moral ccmnunity are excludedbecause they are not moral agents, not be­cause they are "lesser" moral agents. Simi­larly, the additive version of the primacy ofmoral agency view holds that the differentcharacteristics which make one a moral agent

are each morally significant and that themore of these characteristics one possessesthe more morally significant one is. One'sstatus has nothing to ·00 with the~ towhich one possesses a given morally relevantcharacteristic. Is there a way to show theseviews wrong?

Regan does indeed offer arguments which

would apply to these views. He does an ex­cellent job arguing against particular moralagency views such as those held by and

Rawls. [21] The trouble is that these viewscould be amended in the light of several ofhis criticisms; he needs objections so funda­mental that ~ primacy of moral agency viewwould be subject to them. He presses just

BEl'WEEN THE SPOCIFS

.. "..,

such an objection against Kant (who holds a total vielY'). Those who deny that nonrational beings are norally considerable fail to pass the test of impartiality, he urges. Let's grant that it would be wrong to torture a

noral agent for fun. Now, imagine torturing someone who isn't a noral agent, like a human child (to avoid the potentiality canplica­tion, let's stipulate that the child is se­verely retarded). FeIY' would deny that the child suffers just as a noral agent would. But this, and this alone, Regan says, is the norally relevant similarity between the two

in this hypothetical case: "The issue con­cerns their shared capacity for suffering,

not their differing abilities, otherwise we flaunt the requirement of fOl:mal justice: we allow dissimilar treatment of relevantly· similar cases. II [22] The same kind of criti­cism would apply to the additive view, ac­cording to which the child's suffering would

not count ~ RUlch ~ the suffering of a nora! agent. Regan claims that all such vielY's are arbitrary: the kind of hann inflicted is the same, regardless of whether one is a noral agent.

This is similar to Regan's criticism of perfectionism, and, unfortunately, it too fails to forestall a very difficult reply. The total and additive vielY' advocates would emIiIatically deny that their vielY'S fail to be impartial. The cases of the noral agent and the retarded child are relevantly dissimilar, they would say: to claim that suffering is the sole issue is to beg the question against these views.

The trouble is that impartiality is a formal requirement which all but the non­

universal ethical egoist (whose vielY' has plenty of other problems) would embrace. This principle cannot decide the issue. Re­

gan is aware of this, he point out that a nonnative interpretation of justice is re­quired to spell out what counts as a norally

relevant similarity or dissimilarity. [23] But how are we to decide which interpretation of justice to adopt? By applying the same tests of scope, precision, and conformity to our reflective intuitions, which must be cool, clear, rational, infonned, and ~

tial, in the formal sense. The total and additive vielY' advocates, as well as the per­fectionist and the utilitarian[24] would all say that their views do pass these tests. How

is one to reply to them? To show, as Regan

does in Chapter 7, that the subject-of-a-life

view passes the tests just isn't enough. It seems we have reached a relativistic im­passe. [2S]

A Possible Solution

At this point, I would like to suggest a possible way out. I am going to use a line of reasoning developed by ethical theorist Alan Gewirth to sketch a defense of a cc:xnbi­nation vielY' very like Regan's. If this line of reasoning is correct, it will illuminate

what is wrong with the alternative vielY'S. Gewirth himself believes his vielY' to have very different implications, but I will argue that this is not the case.

All noral codes are action guides ad­dressed to those of us capable of understand­ing and acting on them. In his important book, Reason and Morality, [26] Gewirth argues that those of us trying to determine which noral code, if any, to follow RUlst begin by

asking what is required for action itself. Doing so will provide us with reasons to becane active noral agents and will in addi­tion allow us to reject a number of ethical vielY's.

cnly agents are capable of action. A­

gents are able to control their own behavior, have knowledge of the relevant proximate circumstances of their actions, and have goals or purposes they wish to fulfill. [27]

As reflective agents (the only agents who could ever be IOOral agents), we are able to identify the necessary preconditions of our agency. We require (l) the ability to have purposes and (2) the freedan which would allow us to pursue those purposes. The abil­ity to have purposes itself has precondi­tions: life and minimal mental and physical capabilities. Beyond these basic conditions,

a certain quality of life is required for purposiveness: one RUlSt not have to fear constantly losing what has already been at­

. tained, and one RUlSt have the opportunity to increase one's gains. The life and quality of life conditions are canbined by Gewirth under the heading of "well-being." This tenn designates the abilities and condilions ne­cessary for agents to maintain and obtain

what they desire in general. [28] Freedan and

well-being are, then, necessary for agents to achieve their goals.

BE'IWEEN THE SPECIES 38

L ia it 2.1

such an objection against Kant (who holds atotal vielY'). Those who deny that nonrationalbeings are norally considerable fail to passthe test of impartiality, he urges. Let'sgrant that it would be wrong to torture a

noral agent for fun. Now, imagine torturingsomeone who isn't a noral agent, like a humanchild (to avoid the potentiality canplica­tion, let's stipulate that the child is se­verely retarded). FeIY' would deny that thechild suffers just as a noral agent would.But this, and this alone, Regan says, is thenorally relevant similarity between the two

in this hypothetical case: "The issue con­cerns their shared capacity for suffering,

not their differing abilities, otherwise weflaunt the requirement of fOl:mal justice: weallow dissimilar treatment of relevantly·similar cases. II [22] The same kind of criti­cism would apply to the additive view, ac­cording to which the child's suffering would

not count ~ RUlch ~ the suffering of a nora!agent. Regan claims that all such vielY's arearbitrary: the kind of hann inflicted is thesame, regardless of whether one is a noralagent.

This is similar to Regan's criticism ofperfectionism, and, unfortunately, it toofails to forestall a very difficult reply.The total and additive vielY' advocates wouldemIiIatically deny that their vielY'S fail to beimpartial. The cases of the noral agent andthe retarded child are relevantly dissimilar,they would say: to claim that suffering isthe sole issue is to beg the question againstthese views.

The trouble is that impartiality is aformal requirement which all but the non­

universal ethical egoist (whose vielY' hasplenty of other problems) would embrace.This principle cannot decide the issue. Re­

gan is aware of this, he point out that anonnative interpretation of justice is re­quired to spell out what counts as a norally

relevant similarity or dissimilarity. [23]But how are we to decide which interpretationof justice to adopt? By applying the sametests of scope, precision, and conformity toour reflective intuitions, which must becool, clear, rational, infonned, and~tial, in the formal sense. The total andadditive vielY' advocates, as well as the per­fectionist and the utilitarian[24] would allsay that their views do pass these tests. How

is one to reply to them? To show, as Regan

BE'IWEEN THE SPECIES 38

does in Chapter 7, that the subject-of-a-life

view passes the tests just isn't enough. Itseems we have reached a relativistic im­passe. [2S]

A Possible Solution

At this point, I would like to suggest apossible way out. I am going to use a lineof reasoning developed by ethical theoristAlan Gewirth to sketch a defense of a cc:xnbi­nation vielY' very like Regan's. If this lineof reasoning is correct, it will illuminate

what is wrong with the alternative vielY'S.Gewirth himself believes his vielY' to havevery different implications, but I will arguethat this is not the case.

All noral codes are action guides ad­dressed to those of us capable of understand­ing and acting on them. In his importantbook, Reason and Morality, [26] Gewirth arguesthat those of us trying to determine whichnoral code, if any, to follow RUlst begin by

asking what is required for action itself.Doing so will provide us with reasons tobecane active noral agents and will in addi­tion allow us to reject a number of ethicalvielY's.

cnly agents are capable of action. A­

gents are able to control their own behavior,have knowledge of the relevant proximatecircumstances of their actions, and havegoals or purposes they wish to fulfill. [27]

As reflective agents (the only agents whocould ever be IOOral agents), we are able toidentify the necessary preconditions of ouragency. We require (l) the ability to havepurposes and (2) the freedan which wouldallow us to pursue those purposes. The abil­ity to have purposes itself has precondi­tions: life and minimal mental and physicalcapabilities. Beyond these basic conditions,

a certain quality of life is required forpurposiveness: one RUlSt not have to fearconstantly losing what has already been at-

. tained, and one RUlSt have the opportunity toincrease one's gains. The life and qualityof life conditions are canbined by Gewirthunder the heading of "well-being." This tenndesignates the abilities and condilions ne­cessary for agents to maintain and obtain

what they desire in general. [28] Freedan and

well-being are, then, necessary for agents toachieve their goals.

L ia it 2.1

The next step in Gewirth' s transcenden­tal arguII¥3I1t is to say that all reflective agents at least implicitly claim freedom and well-being as rights. These claims are pre­scriptions which mean at the mininn.nn that others should not interfere with one's free­dom and well-being. [29] Any reflective agents who do not make such claims at least implicitly could not desire to fulfill their purposes-which would mean that they are not agents at all. [31] All reflective agents llUlSt, on pain of contradiction, make these claims. [ 31 ]

At this point, reflective agents are not yet active m:>ral agents because they are considering only their own interests. The extension to others (and thus the transforma­tion to m:>ral agency in the active sense) occurs when those agents realize that what justifies their rights claims is the fact that they ~ agents. The fact that one is Arabella WOpenschm:i.dt, white American };hysi­cist, is irrelevant to one's claims. Agents who claim freedom and well-being, claims that are justified by their purposive natures, who deny the freedom and well-being of other agents contradict themselves. This is the heart of Gewirth' s argument for what he calls "the supreme principle of m:>rality: Act in accord with the generic rights of your reci­pients as well as yourself" (the principle of generic consistency [PGC]). [32] Further re­flection on our part indicates that some recipients of our action are due m:>re than

our noninterference. We have positive obli­gations too: to help them avoid hann when doing so causes us no canparable losses and to aid those who, through no fault of their own, cannot achieve well-being. [33]

Following this line of reasoning, it is

clear that a perfectionistic view is unjusti­fied. Agents with m:>re abilities are not to be favored. Arabella wopenschm:i.dt may be a m:>re intelligent, thoughtful agent than Hulk Hogan, but that gives her no m:>ral premium. What counts--and what is sufficient for the full range of basic m:>ral rights, other things being e:rual--is that both are purpose­

ful beings. Utilitarianism is also undercut by this reasoning: we owe agents respect for their rights as agents. The PGC enjoins us to respect other purposive beings as we do ourselves1 it does not impose on us the over­riding duty to maximize happiness or other noIlIlDral goods. Gewirth appeals to us as

rational, reflective agents who wish both to be consistent and to achieve our purposes when he leads us to reject these views.

Next, Gewirth' s line of reasoning leads us to reject the view that m:>ral agency is necessary for m:>ral considerability and maxi­

1I1I.IlII m:>ral significance. The PGC applies to all who have purposes they want to fulfill, for they require freedcm and well-being just as we do. They do not have to claim freedcm and well-being as rights in order to be m:>r­ally considerable. Not all agents have the ability to conceptualize the PGC. They are still, ~ agents, due m:>ral rights. [34] (Of

course, their right to freedcm may have to be abridged when they threaten the freedcm or well-being of others. That holds for m:>ral agents too.) According to this line of rea­soning, those with "preference autonomy" are fully included in the m:>ral ccmmunity: as m:>ral patients rather than as m:>ral agents.

What about those who are not ~

agents? Do they lack m:>ral considerability? Or are they of lesser m:>ral significance? Gewirth's short answer is that as. long as they have purposes or desires, they are m:>r­

ally considerable. Having purposes one wants to fulfill or have fulfilled is the basis for the claim to the right of well-being. Having the right to freedcm is based on another aspect of agency, the ability to fulfill purposes. Those who lack that ability will have no right to unrestricted freedcm but will retain the right to well-being. More­over, Gewirth holds that potential agents are due "preparatory" rights to aid them in achieving agency. [35] He believes (rightly, I think) that potential agency is m:>rally relevant, but he also holds that those who are only potentially purposive are less m:>r­ally significant than those who already have even rudimentary desires. Thus, he believes that infants, who have rudimentary desires, are due a full right to well-being, whereas fetuses (who can have no desires, according to his possibly mistaken view) are not due the same full protection (i.e., in a conflict between their lives and the life or well­being of a purposive being, they should be sacrificed if there is no other way to re­solve the conflict).[36]

Those beings with desires or purposes who are not yet able to carry out their own purposes do not have the full range of rights due an agent, but they are m:>st emJ;hatically

39 BElWEEN THE SPOCIES

The next step in Gewirth' s transcenden­tal arguIlEI1t is to say that all reflectiveagents at least implicitly claim freedom andwell-being as rights. These claims are pre­scriptions which mean at the mininn.nn thatothers should not interfere with one's free­dom and well-being. [29] Any reflectiveagents who do not make such claims at leastimplicitly could not desire to fulfill theirpurposes-which would mean that they are notagents at all. [31] All reflective agentsIlU.1St, on pain of contradiction, make theseclaims. [ 31 ]

At this point, reflective agents are notyet active IOOral agents because they areconsidering only their CMIl interests. Theextension to others (and thus the transforma­tion to IOOral agency in the active sense)occurs when those agents realize that whatjustifies their rights claims is the factthat they ~ agents. The fact that one~Arabella wopenschmidt, white American physi­cist, is irrelevant to one's claims. Agentswho claim freedom and well-being, claims thatare justified by their purposive natures, whodeny the freedom and well-being of otheragents contradict themselves. This is theheart of Gewirth' s argument for what he calls"the supreme principle of IOOrality: Act inaccord with the generic rights of your reci­pients as well as yourself" (the principle ofgeneric consistency [PGC]). [32] Further re­flection on our part indicates that somerecipients of our action are due IOOre than

our noninterference. We have positive obli­gations too: to help them avoid hann whendoing so causes us no canparable losses andto aid those who, through no fault of theirCMIl, cannot achieve well-being. [33]

Following this line of reasoning, it is

clear that a perfectionistic view is unjusti­fied. Agents with IOOre abilities are not tobe favored. Arabella wopenschmidt may be aIOOre intelligent, thoughtful agent than HulkHogan, but that gives her no IOOral premium.What counts--and what is sufficient for thefull range of basic IOOral rights, otherthings being equal--is that both are purpose­

ful beings. Utilitarianism is also undercutby this reasoning: we owe agents respect fortheir rights as agents. The PGC enjoins us

to respect other purposive beings as we doourselves; it does not impose on us the over­riding duty to maximize happiness or othernoIlIlDral goods. Gewirth appeals to us as

39

rational, reflective agents who wish both tobe consistent and to achieve our purposeswhen he leads us to reject these views.

Next, Gewirth' s line of reasoning leadsus to reject the view that IOOral agency isnecessary for IOOral considerability and maxi­

mum IOOral significance. The PGC applies toall who have purposes they want to fulfill,for they require freedom and well-being justas we do. They do not have to claim freedomand well-being as rights in order to be IOOr­ally considerable. Not all agents have theability to conceptualize the PGC. They arestill, ~ agents, due IOOral rights. [34] (Of

course, their right to freedom may have to beabridged when they threaten the freedom orwell-being of others. That holds for IOOralagents too.) According to this line of rea­soning, those with "preference autonomy" arefully included in the IOOral canmunity: asIOOral patients rather than as IOOral agents.

What about those who are not ~agents? Do they lack IOOral considerability?Or are they of lesser IOOral significance?Gewirth's short answer is that as. long asthey have purposes or desires, they are IOOr­

ally considerable. Having purposes one wantsto fulfill or have fulfilled is the basis forthe claim to the right of well-being. Havingthe right to freedom is based on anotheraspect of agency, the ability to fulfillpurposes. Those who lack that ability willhave no right to unrestricted freedom butwill retain the right to well-being. More­over, Gewirth holds that potential agents aredue "preparatory" rights to aid them inachieving agency. [35] He believes (rightly,I think) that potential agency is rrorallyrelevant, but he also holds that those whoare only potentially purposive are less IOOr­ally significant than those who already haveeven rudimentary desires. Thus, he believesthat infants, who have rudimentary desires,are due a full right to well-being, whereasfetuses (who can have no desires, accordingto his possibly mistaken view) are not duethe same full protection (i.e., in a conflictbetween their lives and the life or well­being of a purposive being, they should besacrificed if there is no other way to re­solve the conflict).[36]

Those beings with desires or purposeswho are not yet able to carry out their CMIl

purposes do not have the full range of rightsdue an agent, but they are IOOSt emphatically

BEIWEEN THE SPECIES

equally I1Drally considerable on Gewirth's view. We I1Dral agents have I1Dre ];Ositive

obligations to them than we do to other I1Dral agents: we should act to further their well­being whenever we reasonably can. Thus,

"deficient" humans are on equal I1Dral footing with nonnal humans, so long as they have even primitive desires. [37] We see that Gewirth rejects what I have called the additive view; fewer rights don "t translate into decreased I1Dral significance.

I disagree with Gewirth about sane of the implications he believes his view to have. For example, he does not believe that nonhuman animals can occupy the same I1Dral ];Osition as nonnal and deficient humans. His belief is in part based on false empirical assumptions and in part based on a perva&;i.ve haoocentric bias evident in his work. On the empirical side, he doubts that animals can be agents; he even assumes that severely defi­cient humans are far closer to agency than

nonhuman animals. [38] The truth is the oppo­site. Animals in the wild, apart fran the extremely young, far eclipse sane humans in their mental developnent and their capslcity to achieve their goals. Gewirth I S hanocen­

trism is evident too: he takes himself to be defending the view that basic rights are

human, [39] that "all humans are actual, pros­pective, or ];Otential agents"[40] (contrary to what he later admits about defective hu­mans), that having rights is necessarily con­nected to being human, [41] and that "for human rights to be had one IlRlst only be human."[42]

When his views are separated fran mista­

ken assumptions and the inconsistencies of haoocentrism, we see that what remains is that (1) all I1Dral agents are I1Drally consi­derable and maxi.ma.lly I1Drally significant, (2) all agents, human and nonhuman, are like­wise maxi.ma.lly significant, (3) pw:posive beings, human and nonhuman, with a restricted or nonexistent ability to carry out their pw:poses are equally I1Drally significant, although they cannot be said to have the right to freedcm, (4) ];Otential agents who are not yet pw:posive are m:>rally considera­ble but not as m:>rally significant as the others above. This view is an example of what I have called the canbination view. If the reasoning here is correct, the alterna­tive rights and nonrights views are unaccep­table to the rational agent.

On the face of it, this Gewirthian can­bination view goes further than Regan's rights view. Regan argues for the equal inherent worth of all subjects-of-lives, which as he has defined it means "all agents. " Beings with desires who are unable by their very nature to carry out those de­sires would appear to be excluded. Yet, Regan notes that severely deficient humans, who cannot satisfy even "basic needs and

correlative desires," are subjects-of­lives. [43] This puzzle is resolved when one realizes that Regan has given us another, less restrictive definition of his key rroral notion: "A sufficient condition of being arred such duties [of justice] is that one have a welfare--that one be the experiencing subject of a life that fares well or ill for one as an individual--independently of whe­ther one has a conception of what this is. " [44] Deficient humans are included under this definition, and rightly so. According to this "welfare criterion," one need not be an agent to be fully I1Drally significant.

But what status, on such a view, does the so-called "merely" conscious being have? D:> we have direct duties to such a being, or is it merely "a sacred symbol of the real

BE'IWEEN THE SPECIES 40

IX .L. t iZtL £. IS.

equally I1Drally considerable on Gewirth'sview. We I1Dral agents have I1Dre ];Ositive

obligations to them than we do to other I1Dralagents: we should act to further their well­being whenever we reasonably can. Thus,

"deficient" humans are on equal I1Dral footingwith nonnal humans, so long as they have evenprimitive desires. [37] We see that Gewirthrejects what I have called the additive view;fewer rights don "t translate into decreasedI1Dral significance.

I disagree with Gewirth about sane ofthe implications he believes his view tohave. For example, he does not believe thatnonhuman animals can occupy the same I1Dral];Osition as nonnal and deficient humans. Hisbelief is in part based on false empiricalassumptions and in part based on a perva&;i.vehaoocentric bias evident in his work. On theempirical side, he doubts that animals can beagents; he even assumes that severely defi­cient humans are far closer to agency than

nonhuman animals. [38] The truth is the oppo­site. Animals in the wild, apart fran theextremely young, far eclipse sane humans intheir mental developnent and their capslcityto achieve their goals. Gewirth I S hanocen­

trism is evident too: he takes himself to bedefending the view that basic rights are

human, [39] that "all humans are actual, pros­pective, or ];Otential agents"[40] (contraryto what he later admits about defective hu­mans), that having rights is necessarily con­nected to being human, [41] and that "forhuman rights to be had one IlRlst only behuman."[42]

When his views are separated fran mista­

ken assumptions and the inconsistencies ofhaoocentrism, we see that what remains isthat (1) all I1Dral agents are I1Drally consi­derable and maxi.ma.lly I1Drally significant,(2) all agents, human and nonhuman, are like­wise maxi.ma.lly significant, (3) pw:posivebeings, human and nonhuman, with a restrictedor nonexistent ability to carry out theirpw:poses are equally I1Drally significant,although they cannot be said to have theright to freedcm, (4) ];Otential agents whoare not yet pw:posive are m:>rally considera­ble but not as m:>rally significant as theothers above. This view is an example ofwhat I have called the canbination view. Ifthe reasoning here is correct, the alterna­tive rights and nonrights views are unaccep­table to the rational agent.

BE'IWEEN THE SPECIES 40

On the face of it, this Gewirthian can­bination view goes further than Regan'srights view. Regan argues for the equalinherent worth of all subjects-of-lives,which as he has defined it means "allagents. " Beings with desires who are unableby their very nature to carry out those de­sires would appear to be excluded. Yet,Regan notes that severely deficient humans,who cannot satisfy even "basic needs and

correlative desires," are subjects-of­lives. [43] This puzzle is resolved when onerealizes that Regan has given us another,less restrictive definition of his key rroralnotion: "A sufficient condition of beingarred such duties [of justice] is that onehave a welfare--that one be the experiencingsubject of a life that fares well or ill forone as an individual--independently of whe­ther one has a conception of what thisis. " [44] Deficient humans are included underthis definition, and rightly so. Accordingto this "welfare criterion," one need not bean agent to be fully I1Drally significant.

But what status, on such a view, doesthe so-called "merely" conscious being have?D:> we have direct duties to such a being, oris it merely "a sacred symbol of the real

. t iZtL £. IS. IX .L

(

.. .- III III ··-11 ~ ~,; .,....,."....."". t7mJei!it2tls;j;ji'ti1!!!imt@'W'_t!tjtttgffi

tiling," as Feinberg would say, [45] to be pro­tected on utilitarian grounds alone? Regan suggests this may be so when he speculates that such a being may be a mere receptacle for intrinsic value rather than being valua­

ble in its own right. [46] (It is fascinating that he and Singer, who differ on so IlUJch else, reach the same conclusion here; both treat the self-conscious as irreplaceable and

the ''merely'' conscious as replaceable recep­tacles of value.)

I can only speculate here about this very troubling issue. Is self-consciousness a prerequisite for having preferences, de­sires, or purposes (i.e., is it a precondi­tion for having a welfare)? If it is, then I have serious doubts about whether anything living is merely conscious. The newborn has preferences, although it is not self-con­scious in the fully reflexive sense of the tenn. Might it not be self-conscious in another sense, having something akin to what Sartre calls a "horizon" surrounding its consciousness [47]--being thus very simply self-aware? If this is denied-if fully reflexive self-consciousness is declared to be the only kind--then it seems that one does not have to be self-conscious to have prefe­rences. Having preferences enables one to have a welfare, and this-following the rea­soning sketched above--is sufficient for noral considerability. [47] The only con­scious being I can conceive of who would not be norally considerable would be one who could not care about what states it under­goes, who could not have preferences, even potentially. Perhaps a highly soI;histicated robot could fit this description, but the humans and nonhumans we have contact within this world do not.

If conscious beings who are not, and never will be, reflexively self-conscious are norally considerable, can they be said to be as norally significant as other noral pa­tients and agents? Or do they just have the right not to be tortured wantonly? I want to

,suggest that they are owed the same respect­ful treatment due other noral patients and agents. Since they care about what happens to them (without knowing that they do), it would be wrong to treat them as mere weans. However, this does not necessarily mean that, in a case of unavoidable conflict between their lives and the lives of noral agents or reflexively self-conscious noral patients, we

should flip a coin to determine who dies. Suppose that a "merely" conscious being and a reflexively self-conscious being must have

our assistance in oroer to live, but we only have the resources to save one. Now suppose that either being would be harmed by death, a·~_ least in the sense that any further

experiences would be precluded for that being. Regan has argued that if we IlUJst choose between hanning one IIDrally considera­ble being rather than another in such a situ­ation and if one of the beings would be banned less by our actions than the other, we ought to bann that being rather than the other. Such a decision would in no way imply a lack of equal respect for the two beings. He goes on to suggest that loss of life may be a IlUJch greater bann to one being than to another. [49] Perhaps--we IlUJst be very cau­tious here-a reflexively self-conscious being would be nore banned by death than one who is not. If this can be made out, and if it would be wrong to cause one being IIDre bann than another when that can be avoided, and if both would otherwise die,· then the "merely" conscious being should die instead of the reflexively self-conscious being. Note, however, that this line of reasoning would not justify experimenting on the fonner being in oroer to save the life of the lat­ter. We would not be shcMing equal respect for the two if we did this, since we would be treating the fonner as a mere weans.

If what has been said in the last sec­tion of this paper is correct, there are a great many moral patients we IIDral agents are obligated to consider. Determining the ex­tent to which we should not interfere with those beings and the extent to which we should assist them is no easy matter. One hopes that one day we will cxxoe IlUJch closer to giving other IIDrally considerable beings the respect which is due them, whatever our

specific obligations may be. OUr recoro so far has been nothing short of dismal.

Notes

1. Some I;hilosophers would dispute this way of characterizing IIDral agency. Lawrence Johnson, for example, has argued that one need not understand ethical principles, or indeed possess IIDral concepts at all, in order to be a IIDral agent (see his "can Animals Be Moral Agents?," Ethics ~ Animals 4/2 (1983) ; 50-61 ) • However, if we were to broaden the category of noral agent in this

41 BE'IWEEN THE SPECIES

mrs -n tt' t t WM t br

(

tldng," as Feinberg would say, [45] to be pro­tected on utilitarian grounds alone? Regansuggests this may be so when he speculatesthat such a being may be a mere receptaclefor intrinsic value rather than being valua-

ble in its own right. [46] (It is fascinatingthat he and Singer, who differ on so IlUJchelse, reach the same conclusion here: bothtreat the self-conscious as irreplaceable and

the ''merely'' conscious as replaceable recep­tacles of value.)

I can only speculate here about thisvery troubling issue. Is self-consciousnessa prerequisite for having preferences, de­sires, or purposes (i.e., is it a precondi­tion for having a welfare)? If it is, then Ihave serious doubts about whether anythingliving is merely conscious. The newborn haspreferences, although it is not self-con­scious in the fully reflexive sense of theterm. Might it not be self-conscious inanother sense, having something akin to whatSartre calls a "horizon" surrounding itsconsciousness [47]--being thus very simplyself-aware? If this is denied-if fullyreflexive self-consciousness is declared tobe the only kind--then it seems that one does

not have to be self-conscious to have prefe­rences. Having preferences enables one tohave a welfare, and this-following the rea­soning sketched above--is sufficient forrooral considerability. [47] The only con­scious being I can conceive of who would notbe roorally considerable would be one whocould not care about what states it under­goes, who could not have preferences, evenpotentially. Perhaps a highly soI;histicatedrobot could fit this description, but thehumans and nonhumans we have contact withinthis world do not.

If conscious beings who are not, andnever will be, reflexively self-conscious areroorally considerable, can they be said to beas roorally significant as other rooral pa­tients and agents? Or do they just have theright not to be tortured wantonly? I want to

,suggest that they are owed the same respect­ful treatment due other rooral patients andagents. Since they care about what happensto them (without knowing that they do), itwould be wrong to treat them as mere means.However, this does not necessarily mean that,in a case of unavoidable conflict betweentheir lives and the lives of rooral agents orreflexively self-conscious rooral patients, we

41

should flip a coin to determine who dies.Suppose that a "merely" conscious being and areflexively self-conscious being must have

our assistance in order to live, but we onlyhave the resources to save one. Now supposethat either being would be harmed by death,a·~. least in the sense that any further

experiences would be precluded for thatbeing. Regan has argued that if we IlUJstchoose between harming one xoorally considera­ble being rather than another in such a situ­ation and if one of the beings would bebanned less by our actions than the other, weought to bann that being rather than theother. Such a decision would in no way implya lack of equal respect for the two beings.He goes on to suggest that loss of life maybe a IlUJch greater bann to one being than toanother. [49] Perhaps--we IlUJst be very cau­tious here-a reflexively self-consciousbeing would be roore harmed by death than onewho is not. If this can be made out, and ifit would be wrong to cause one being xoorebann than another when that can be avoided,and if both would otheIWise die, - then the"merely" conscious being should die insteadof the reflexively self-conscious being.Note, however, that this line of reasoningwould not justify experimenting on the fonnerbeing in order to save the life of the lat­ter. We would not be shcMing equal respectfor the two if we did this, since we would betreating the fonner as a mere means.

If what has been said in the last sec­tion of this paper is correct, there are agreat many rooral patients we xooral agents areobligated to consider. Determining the ex­tent to which we should not interfere withthose beings and the extent to which weshould assist them is no easy matter. Onehopes that one day we will cane IlUJch closerto giving other xoorally considerable beingsthe respect which is due them, whatever our

specific obligations may be. OUr record sofar has been nothing short of dismal.

Notes

1. Some I;hilosophers would dispute thisway of characterizing xooral agency. LawrenceJohnson, for example, has argued that oneneed not understand ethical principles, orindeed possess xooral concepts at all, inorder to be a xooral agent (see his "canAnimals Be Moral Agents?," Ethics & Animals4/2 (1983): 50-61) • However, if ~ were tobroaden the category of rooral agent in this

BETWEEN THE SPECIES

way, many who now believe it is prima facie Wrong to harm noral agents \I1Ould simply re­tract and refonnulate this view. I prefer to follOW' Steve F. sapontzis here. He argues that animals are not "noral beings" (noral agents in the reflective sense) but can in­deed be 'virtuous (see his "Are Animals Moral Beings?," American Philosophical Quarterly 17/1 (1980): 45-52).

2. It is convenient for the prrposes of this paper to define ''noral agent" and "nora! patient" in IlD..ltual1y exclusive tenns. Tan

Regan does this too (The Case for Animal Rights (Berkeley: University of California Press, 1983): 151-6). However, I depart from his sense of ''noral patient" in t\l1O respects. Regan' s "noral patient" applies to a very specific group, viz., subjects-of-lives. I prefer not to restrict the tenn this way. Moreover, in his sense, being a noral patient carries no implication that anyone has noral obligations to one. It seems nore natural to me to build in this implication.

3. Joel Feinberg, "Abortion," In Tan

Regan (ed.), Matters of Life and Death, se­cond Edition (New York: Random House, 1986), pp. 261-2.

4. Regan, £12. cit., 243.

5. see Donald Griffin, The QIestion of Animal Awareness (New York: Rockefeller University Press, 1981) and his Animal Think­~ (cambridge: Harvard University Press, 1984). Also, see Stephen Walker, Animal Thought (London: Routledge and Kegan Paul, 1983) and Stephen Clark, The Nature of the Beast (Oxford: Oxford University Press, 1982).

6. Interests and Rights (Oxford: The Clarendon Press, 1980).

7. For excellent critiques, see Regan, .912. cit., 38-49, and S. F. sapontzis, "Inte­rests and Animals, Needs and Language, "

~~ Animals 4/2 (1983): 38-49.

8. Michael A. Fox, The ~ for Animal Experimentation (Berkeley: University of California Press, 1986): 28-9.

9. Regan,.912. cit., 84-6.

10. Ibid., 244. But they do have a

BETWEEN THE SPEX:;IES 42

welfare: their lives "fare well or ill for them logically independently" of their value for others. That does not make them "sub­jects-of-lives" as he has defined the notion, however. I will argue later that Regan real­ly offers two criteria for noral considera­bility, both of which he calls "being the subject-of-a-life."

11. In this context, "necessary" means "required for one to be justified in attri­buting noral considerability" and "suffi­cient" means "justifying ground for the at­tribution of noral considerability. " What

sort of justification one can get depends on which metaethical theory is correct. All the major theories except one, enotivism, imply

that some sort of justification is possible, and enotivism is rightly not much defended these days. (For a nore detailed discussion of ethical justification, see my "The Justi­fication of an Envirorunental Ethic," Environ­mental Ethics 5/1 (1983): 56-60.)

12. Joel Feinberg, "The Nature and Value of Rights," The Journal of Value Inquiry 4 (1970). This is the sense Regan adopts tOOl see Chapter 8 of ~ Case for Animal Rights.

13. Feinberg, "Abortion," 266-7.

14. Feinberg and others have argued that potentiality views oommit a fatal logical

error. The strict view is said to confuse being potentially qualified for rights with having rights. The gradualist view is said to confuse being alnost qualified for rights with being partially qualified for rights or being qualified for weak rights (Feinberg, ibid., 267, 269.). The criticism of gradual­ism would also apply to the additive view.

I believe this criticism misconstrues potentiality views. The strict view claims that being an actual or potential X is suffi­cient for noral rights, not that being a potential rights-bearer is equivalent to being a rights-bearer. Nor does the gradual­ist view claim that being alnost qualified for rights is equivalent to being qualified for partial rightsl it takes rights claims to be ordered on a continuum, increasing in strength as one's potential is actualized.

(this view is particularly plausible given Feinberg's analysis of rights as valid claims. ) The critics interpret potentiality views as presupposing their very denials 1

La A , t, g itJJ _

way, many who now believe it is prima facieWrong to harm noral agents \I1Ould simply re­tract and refonnulate this view. I prefer tofollOW' Steve F. sapontzis here. He arguesthat animals are not "noral beings" (noralagents in the reflective sense) but can in­deed be 'virtuous (see his "Are Animals MoralBeings?," American Philosophical Quarterly17/1 (1980): 45-52).

2. It is convenient for the prrposes ofthis paper to define ''noral agent" and "nora!patient" in IlD..ltual1y exclusive tenns. Tan

Regan does this too (The Case for AnimalRights (Berkeley: University of CaliforniaPress, 1983): 151-6). However, I depart fromhis sense of ''noral patient" in t\l1O respects.Regan' s "noral patient" applies to a veryspecific group, viz., subjects-of-lives. Iprefer not to restrict the tenn this way.Moreover, in his sense, being a noral patientcarries no implication that anyone has noralobligations to one. It seems nore natural tome to build in this implication.

3. Joel Feinberg, "Abortion," In Tan

Regan (ed.), Matters of Life and Death, se­cond Edition (New York: Random House, 1986),pp. 261-2.

4. Regan, £12. cit., 243.

5. see Donald Griffin, The QIestion ofAnimal Awareness (New York: RockefellerUniversity Press, 1981) and his Animal Think­~ (cambridge: Harvard University Press,1984). Also, see Stephen Walker, AnimalThought (London: Routledge and Kegan Paul,1983) and Stephen Clark, The Nature of theBeast (Oxford: Oxford University Press,1982).

welfare: their lives "fare well or ill forthem logically independently" of their valuefor others. That does not make them "sub­jects-of-lives" as he has defined the notion,however. I will argue later that Regan real­ly offers two criteria for noral considera­bility, both of which he calls "being thesubject-of-a-life."

11. In this context, "necessary" means"required for one to be justified in attri­buting noral considerability" and "suffi­cient" means "justifying ground for the at­tribution of noral considerability. " What

sort of justification one can get depends onwhich metaethical theory is correct. All themajor theories except one, enotivism, imply

that some sort of justification is possible,and enotivism is rightly not much defendedthese days. (For a nore detailed discussionof ethical justification, see my "The Justi­fication of an Envirorunental Ethic," Environ­mental Ethics 5/1 (1983): 56-60.)

12. Joel Feinberg, "The Nature and Valueof Rights," The Journal of Value Inquiry 4(1970). This is the sense Regan adopts tOOlsee Chapter 8 of ~ Case for Animal Rights.

13. Feinberg, "Abortion," 266-7.

14. Feinberg and others have argued thatpotentiality views oommit a fatal logical

error. The strict view is said to confusebeing potentially qualified for rights withhaving rights. The gradualist view is saidto confuse being alnost qualified for rightswith being partially qualified for rights orbeing qualified for weak rights (Feinberg,ibid., 267, 269.). The criticism of gradual­ism would also apply to the additive view.

6. Interests and Rights (Oxford: TheClarendon Press, 1980).

7. For excellent critiques, see Regan,.912. cit., 38-49, and S. F. sapontzis, "Inte­rests and Animals, Needs and Language, "

~~ Animals 4/2 (1983): 38-49.

8. Michael A. Fox, The~ for AnimalExperimentation (Berkeley: University ofCalifornia Press, 1986): 28-9.

9. Regan,.912. cit., 84-6.

10. Ibid., 244. But they do have a

BETWEEN THE SPEX:;IES 42

I believe this criticism misconstruespotentiality views. The strict view claimsthat being an actual or potential X is suffi­cient for noral rights, not that being apotential rights-bearer is equivalent tobeing a rights-bearer. Nor does the gradual­ist view claim that being alnost qualifiedfor rights is equivalent to being qualifiedfor partial rightsl it takes rights claims tobe ordered on a continuum, increasing instrength as one's potential is actualized.

(this view is particularly plausible givenFeinberg's analysis of rights as validclaims. ) The critics interpret potentialityviews as presupposing their very denials 1

La A , t, g itJJ _

---- --this is unwarranted.

15. An interesting apparent exception to this is Peter Singer's version of preference utilitarianism (see his "Killing Humans and Killing Animals," Inquiry 22/1-2 (1979): 145­56). He holds that the basic moral principle is to maximize preference satisfaction. He argues that reflexively self-conscious indi­viduals are not "mere receptacles of value" as the "merely" conscious are said to be; they are said to be prima facie irreplacea­ble, unlike the ,latter, because of their preference to live. This argument fails for a variety of reasons. Most important, self­conscious beings turn out to be just as re­placeable on this form of utilitarianism as on any other. I discuss this argument in my

"On Replaceability," Ethics ~ Anlinals 3/4 (1982): 96-105. See Tom Regan's excellent dissection of Singer's view in The Case for Animal Rights, 206-11.

16. Regan, ibid, 246.

17. Ibid, Chapter 4.

18. Ibid., 260-1.

19. Ibid., 240.

20. Ibid., 234.

21. Ibid., Chapter 5, especially sec­tions 5.4 and 5.5.

22. Ibid., 183. Another argument against the total view which Regan could have pressed (but wisely did not) is that moral agents would not deny basic moral rights to those less fortunate than themselves (i.e.,

'. those who cannot be moral agents). The argu­ment is that denial of rights in these cases

is unjust, that such beings, who had no choice about their condition, do not deserve

such treatment. The problem is that this argument presupposes that those who are not moral agents are morally considerable. The principle of justice, of treatment according to desert, does not apply to those without moral considerability. Thus, this argument

simply begs the question against the total view. (This very same argument is at' the heart of some recent defenses of speciesism. see Michael Wreen, "In Defense of Speciesism, Ethics ~ Anlinals 5/3 (1984): 47-60. Michael A. Fox also employed this argument in The

Case for Anlinal Experimentation, 61. I have

criticized Wreen's article in Ethics & Ani­mals 5/4 (1984). For further responses and

replies between us, see the 1986 issues of

Between the Species.)

23. Regan, £E. cit., 232.

24. utilitarians are apt to dismiss appeals to intuitions, but they have no quar­

rel against rationality, coolness, conceptual clarity, being informed, and being impartial.

25. One might think that a Rawls-type approach to justice would help here, but it would not. Rawls' view, plrged of its arbi­trariness (Regan rightly points out that those behind the veil of ignorance should not be allowed to knCM that they will be human,

moral agents in the society whose rules they are deciding (£E. cit., 170-3», would seem to ensure an egalitarian outcome for moral agents and patients, but it does not. Apart fram the dubious assmnption that self-inte­rest lies at the basis of ethics, this (otherwise nonarbitrary) Rawlsian view cannot exclude the dedicated perfectionists, Kan­

tians, and others we have been discussing. These people would interpret justice in the ways their theories prescribe, regardless of the consequences they might suffer when in­

carnated in their chosen society.

26. Alan Gewirth, Reason and Morality

(Chicago: University of Chicago Press, 1978) • A much briefer description of his theory appears in Part 1 of his ~ Rights: Essays on Justification and Applications (Chicago: University of Chicago Press, 1982).

27. Reason and Morality, 27.

28. Ibid., 64.

29. Ibid., 78.

30. Gewirth argues that those agents who deliberately act to limit their own freedom

or well-being are still, as they act (i.e., ~ agents), claiming the right to noninter­ference. (He further notes that mental, physical, or economic constraints are gener­

ally the reasons for such behavior) (ibid., 264-7).

31. Gewirth cautions us that he is not

43 BElWEEN THE SPECIES

------------------------_.__ .._._ .

this is unwarranted.

15. An interesting apparent exception tothis is Peter Singer's version of preferenceutilitarianism (see his "Killing Humans andKilling Animals," Inquiry 22/1-2 (1979): 145­

56). He holds that the basic moral principleis to maximize preference satisfaction. Heargues that reflexively self-conscious indi­

viduals are not "mere receptacles of value"as the "merely" conscious are said to be;they are said to be prima facie irreplacea­

ble, unlike the ,latter, because of their

preference to live. This argument fails fora variety of reasons. Most important, self­conscious beings turn out to be just as re­

placeable on this form of utilitarianism ason any other. I discuss this argument in my

"On Replaceability," Ethics ~ Anllnals 3/4(1982): 96-105. See Tom Regan's excellentdissection of Singer's view in The Case forAnimal Rights, 206-11.

16. Regan, ibid, 246.

17. Ibid, Chapter 4.

18. Ibid., 260-1.

19. Ibid., 240.

20. Ibid., 234.

21. Ibid., Chapter 5, especially sec­tions 5.4 and 5.5.

22. Ibid., 183. Another argumentagainst the total view which Regan could havepressed (but wisely did not) is that moral

agents would not deny basic moral rights to

those less fortunate than themselves (i.e.,those who caIUlOt be moral agents). The argu­

ment is that denial of rights in these cases

is unjust, that such beings, who had nochoice about their condition, do not deserve

such treatment. The problem is that thisargument presupposes that those who are notmoral agents are morally considerable. Theprinciple of justice, of treatment according

to desert, does not apply to those without

moral considerability. Thus, this argument

simply begs the question against the totalview. (This very same argument is at' theheart of some recent defenses of speciesism.See Michael Wreen, "In Defense of Speciesism,

Ethics ~ Anllnals 5/3 (1984): 47-60. MichaelA. Fox also employed this argument in The

43

Case for Anllnal Experimentation, 61. I have

criticized Wreen's article in Ethics & Ani­mals 5/4 (1984). For further res~es and

replies between us, see the 1986 issues of

Between the Species.)

23. Regan,.9I2. cit., 232.

24. utilitarians are apt to dismissappeals to intuitions, but they have no quar­

rel against rationality, coolness, conceptualclarity, being informed, and being impartial.

25. One might think that a Rawls-type

approach to justice would help here, but itwould not. Rawls' view, purged of its arbi­

trariness (Regan rightly points out thatthose behind the veil of ignorance should not

be allowed to knCM that they will be human,

moral agents in the society whose rules theyare deciding (9.E. cit., 170-3», would seemto ensure an egalitarian outcome for moralagents and patients, but it does not. Apartfram the dubious assumption that self-inte­rest lies at the basis of ethics, this(otherwise nonarbitrary) Rawlsian view caIUlOtexclude the dedicated perfectionists, Kan­

tians, and others we have been discussing.These people would interpret justice in the

ways their theories prescribe, regardless ofthe consequences they might suffer when in­

carnated in their chosen society.

26. Alan Gewirth, Reason and Morality

(Chicago: University of Chicago Press,

1978) • A much briefer description of histheory appears in Part 1 of his~ Rights:

Essays on Justification and Applications(Chicago: University of Chicago Press,1982).

27. Reason and Morality, 27.

28. Ibid., 64.

29. Ibid., 78.

30. Gewirth argues that those agents whodeliberately act to limit their own freedom

or well-being are still, as they act (i.e.,

~ agents), claiming the right to noninter­ference. (He further notes that mental,

physical, or economic constraints are gener­

ally the reasons for such behavior) (ibid.,264-7).

31. Gewirth cautions us that he is not

BElWEEN THE SPECIES

r-'~" .

claiming that "X needs freedan and well-being in order to act" logically necessitates "X has the right to freedan and well-being. " The necessity arises within the agent's view­point: fran "X regards freedan and well­being as necessary goods" it does follow that "X. rationally claims the rights to freedan and well-being" (ibid., 160-1). The distinc­tion is crucial; this is Gewirth's way of not being trapped by the "is/ought" problem.

32. Ibid., 112. For Gewirth's extended

statement and defense of the PGC, see Reason and Morality, Chapter 3. A briefer accOlIDt is given in "The Basis and Content of Human Rights," Essay 1 in his ~ Rights.

33. Reason and Morality, 312-7.

34. Gewirth is inconsistent on this very inp:>rtant point. His definition of "agent" at the outset of his argument ("a being with purposes he or she wants to fulfill, who has control over his or her actions, and who

knCMS the relevant proximate circumstances of those actions") includes no reference to the ability to reflect on the abstract precondi­tions of actions. He also makes a point of arguing that agents with very low-level abil­ities are nevertheless agents with full moral rights (ibid., 140) • He repeats this in his later book ~ Rights, arguing that "mini­mal rationality" in the sense spelled out above is all that is required for agency and

the rights to freedan and well-being (p. 8) • Beings who are only preferentially autoncm:>us fit this description. But in other passages in which agency is characterized, he interpo­lates the requirement that one be capable of reflecting on the preconditions of action and

of claiming these as rights (pp., 120, 138). See Human Rights, 11: "For a person to have human rights, then, is for him to be in a position to make morally justified stringent, effective demands on other persons that they not interfere with his having the necessary goods of action and that they also help him

to attain these goods when he cannot do so by his own efforts." In ~ and Morality, this interpolation is not found; see pages 124, 133, and 180.

The interpolation should never have been made. First, the reasoning which leads to the PGC hinges on the reflective agent's recognizing that he or she has the rights to freedan and well-being because these are the

preconditions for action. All beings like the agent in the respect that they too need freedan and well-being in order to fulfill their purposes are then accorded the same rights by that agent. Plainly, it is not necessary that these others also be capable of conceptualizing the PGC. Second, as we shall see, Gewirth holds that some beings who are not even preferentially autonomous (i.e., who are not even agents in the rn:i..nimal sense) are due the right to well-being. These beings (e.g., children and the mentally de­fective) need have no concept of the right to well-being (let alone be able to claim it as a right) in order to have that right. This is an extremely important part of Gewirth' s overall view. If he wants to retain it, he must drop his repeated insistence that only intellectually sophisticated moral agents can have rights.

35. ReasOn and Morality, 142.

36. Ibid., 142-3

37. Ibid., 141-2.

38. see his description of their respec­tive abilities on pages 141-2, 144.

39. Ibid., 317. He does, however, argue that nonhuman animals have the right not to have pain wantonly inflicted on them (p. 144).

40. Ibid, 64.

41. Ibid., 103.

42. Ibid., 317. As the title of his more recent book, Human Rights, tells us, he has not retreated fran a hcm:>centric posi­tion.

43. Regan,~. cit., 244.

44. Ibid., 171.

45. Feinberg, "Abortion," 273.

46. Regan, op. cit., 246.

47. Jean-Paul sartre, The Transcendence of the .!E (New York: Noonday Press, 1971): 75.

48. One philosopher who seems to think

.BEIWEEN THE SPECIES 44

aiijjLiT-~CJ. -----­J2M @Q

r-'~" .

claiming that "X needs freedan and well-beingin order to act" logically necessitates "Xhas the right to freedan and well-being. "The necessity arises within the agent's view­point: fran "X regards freedan and well­being as necessary goods" it does follow that"X. rationally claims the rights to freedanand well-being" (ibid., 160-1). The distinc­tion is crucial; this is Gewirth's way of notbeing trapped by the "is/ought" problem.

32. Ibid., 112. For Gewirth's extended

statement and defense of the PGC, see Reasonand Morality, Chapter 3. A briefer accOlIDtis given in "The Basis and Content of HumanRights," Essay 1 in his~ Rights.

33. Reason and Morality, 312-7.

34. Gewirth is inconsistent on this veryinp:>rtant point. His definition of "agent"at the outset of his argument ("a being withpurposes he or she wants to fulfill, who hascontrol over his or her actions, and who

knCMS the relevant proximate circumstances ofthose actions") includes no reference to theability to reflect on the abstract precondi­tions of actions. He also makes a point ofarguing that agents with very low-level abil­ities are nevertheless agents with full moralrights (ibid., 140) • He repeats this in hislater book~ Rights, arguing that "mini­mal rationality" in the sense spelled outabove is all that is required for agency and

the rights to freedan and well-being (p. 8) •Beings who are only preferentially autoncm:>usfit this description. But in other passagesin which agency is characterized, he interpo­lates the requirement that one be capable ofreflecting on the preconditions of action and

of claiming these as rights (pp., 120, 138).See Human Rights, 11: "For a person to havehuman rights, then, is for him to be in aposition to make morally justified stringent,effective demands on other persons that theynot interfere with his having the necessarygoods of action and that they also help him

to attain these goods when he cannot do so byhis own efforts." In~ and Morality,this interpolation is not found; see pages124, 133, and 180.

The interpolation should never have beenmade. First, the reasoning which leads tothe PGC hinges on the reflective agent'srecognizing that he or she has the rights tofreedan and well-being because these are the

.BEIWEEN THE SPECIES 44

preconditions for action. All beings likethe agent in the respect that they too needfreedan and well-being in order to fulfilltheir purposes are then accorded the samerights by that agent. Plainly, it is notnecessary that these others also be capableof conceptualizing the PGC. Second, as weshall see, Gewirth holds that some beings whoare not even preferentially autonomous (i.e.,who are not even agents in the rn:i..nimal sense)are due the right to well-being. Thesebeings (e.g., children and the mentally de­fective) need have no concept of the right towell-being (let alone be able to claim it asa right) in order to have that right. Thisis an extremely important part of Gewirth' soverall view. If he wants to retain it, hemust drop his repeated insistence that onlyintellectually sophisticated moral agents canhave rights.

35. ReasOn and Morality, 142.

36. Ibid., 142-3

37. Ibid., 141-2.

38. see his description of their respec­tive abilities on pages 141-2, 144.

39. Ibid., 317. He does, however, arguethat nonhuman animals have the right not tohave pain wantonly inflicted on them (p.144).

40. Ibid, 64.

41. Ibid., 103.

42. Ibid., 317. As the title of hismore recent book, Human Rights, tells us, hehas not retreated fran a hcm:>centric posi­tion.

43. Regan,~. cit., 244.

44. Ibid., 171.

45. Feinberg, "Abortion," 273.

46. Regan, op. cit., 246.

47. Jean-Paul sartre, The Transcendenceof the .!E (New York: Noonday Press, 1971):75.

48. One philosopher who seems to think

J2M @Q aiijjLiT-~CJ. ------

that only the re£lexively self-conscious can have a genuine welfare is the Michael A. Fox of The case for Animal Experimentation. He there claims that it .:is "completely spurious" to speak of the lives of those who are not re£lexively self-conscious as "more or less 'full,' 'satisfying,' and so forth," since these beings cannot "re£lectively evaluate the quality of their lives and find them a cause of satisfaction or regret" (pp. 28-9). Because of their lack of reflection, Fox says that their "pleasurable experiences are not valuable to them" (27), and in general "their lives also cannot have intrinsic value or value to themselves" (48). Fox's view im­plies that, in a crucial sense, such beings cannot fare well or ill because they are unable to care about what hapPenS to them. (Fox does continue to use the word "welfare" on occasion when referring to those he be­lieves are not re£lexively self-conscious. He also grants that "it would be meaningful to say that it is in their interest (accords with their observed pre£erences) to have pleasurable experiences repeat themselves" (27) • But as the implied reductionism in this quote, in addition to the previous quotes, indicates, this is "welfare" and "interest" in a very attenuated sense of these tenns at best.)

Although Fox makes all the above claims about nonhuman animals, they apply rather more accurately to very young or impaired humans. Before we decide that the lives of babies and the severely senile cannot be "more or less satisfying," let's consider their "observed preferences" when they are cuddled or beaten. Don't they give every indication that they are satisfied or greatly distressed by what is being done to them? Must they be able to tell themselves "my

life-quality has now taken a distinct down-

NUMBER 87

Might have been a rabbit,

or maybe a white rat.

Just another specimen,

but I recall it's number was "S7."

I think we poured something into its eyes,

or tried to give it cancer.

Funding is getting tighter,

a back-up disease is good business.

People believe we cure them

by what goes on here.

I guess we il..m akin to gods.

I've forgotten our original theory,

but it still pays my salary.

Lucky that animals don't have feelings,

or we might be held accountable

for numberS7.

Kathleen Malley

ward turn" in order to be dissatisfied by a beating? Apart fran language,a tortured baby and a tortured normal human adult display remarkably similar "observedpre£erences." How can Fox account for this? Doesn't the principle of parsimony suggest that both care about what is happening to them?

49. Regan, Ope cit., 324-5. I strongly disagree with the way Regan applies this principle, however. He says that "no reason­able person would deny" that the de{ith of a normal human is a gr~ter hann to that human than the death of a dog is to the dog. He suggests that we ought to throw the dog out of the proverbial lifeboat if there isn't sufficient roan on board for all the humans and the dog. This seems to be a totally uncharacteristic touch of human chauvinism on Regan's part. For a thoughtful discussion of this problem, see Henry Cohen's review of The

case ~ ~ Rights in Ethics ~ Animals 5/1 (1984): 11-4.

45 BEIWEEN THE SPECIES

that only the reflexively self-conscious canhave a genuine welfare is the Michael A. Foxof The case for Animal Experimentation. Hethere claims that it .:is "completely spurious"to speak of the lives of those who are notreflexively self-conscious as "more or less'full,' 'satisfying,' and so forth," sincethese beings cannot "reflectively evaluatethe quality of their lives and find them acause of satisfaction or regret" (pp. 28-9).Because of their lack of reflection, Fox saysthat their "pleasurable experiences are notvaluable to them" (27), and in general "theirlives also cannot have intrinsic value orvalue to themselves" (48). Fox's view im­plies that, in a crucial sense, such beingscannot fare well or ill because they areunable to care about what happens to them.(Fox does continue to use the word "welfare"on occasion when referring to those he be­lieves are not reflexively self-conscious.He also grants that "it would be meanin<jfulto say that it is in their interest (accordswith their observed preferences) to havepleasurable experiences repeat themselves"(27). But as the implied reductionism inthis quote, in addition to the previousquotes, indicates, this is "welfare" and

"interest" in a very attenuated sense ofthese tenns at best.)

Although Fox makes all the above claimsabout nonhuman animals, they apply rathermore accurately to very young or impairedhumans. Before we decide that the lives ofbabies and the severely senile cannot be"more or less satisfying," let's considertheir "observed preferences" when they arecuddled or beaten. Don't they give everyindication that they are satisfied or greatlydistressed by what is being done to them?Must they be able to tell themselves "my

life-quality has now taken a distinct down-

45

NUMBER 87

Might have been a rabbit,

or maybe a white rat.

Just another specimen,

but I recall it's number was "87."

I think we poured something into its eyes,

or tried to give it cancer.

Funding is getting tighter,

a back-up disease is good business.

People believe we cure them

by what goes on here.

I guess we .s.m akin to gods.

I've forgotten our original theory,

but it still pays my salary.

Lucky that animals don't have feelings,

or we might be held accountable

for number 87.

Kathleen Malley

ward turn" in order to be dissatisfied by abeating? Apart fran language,a tortured baby

and a tortured no:cnal human adult displayremarkably similar "observed preferences. "How can Fox account for this? Doesn't theprinciple of pirsimony suggest that both careabout what is happening to them?

49. Regan, Ope cit., 324-5. I stronglydisagree with the way Regan applies thisprinciple, however. He says that "no reason­able person would deny" that the death of ano:cnal human is a gr~ter hann to that humanthan the death of a dog is to the dog. Hesuggests that we ought to throw the dog outof the proverbial lifeboat if there isn'tsufficient roan on board for all the htnnansand the dog. This seems to be a totallyuncharacteristic touch of human chauvinism onRegan's part. For a thoughtful discussion ofthis problem, see Henry Cohen's review of Thecase ~ ~ Rights in Ethics ~ Animals5/1 (1984): 11-4.

BEIWEEN THE SPEX::IE'S