Title: Rational Engagement, Emotional Response,
1Rational Engagement, Emotional Response, the
Prospects for Moral Progress in Animal Use
Debates
- Ethicists have increasingly turned their
attention to moral questions about the treatment
of non-human animals. Arguments from a range of
perspectives have been given for the conclusion
that routine uses of animals in agriculture, the
fashion industry, and experimentation are morally
wrong. - Defenses of these practices, however, have been
far fewer, and generally less developed, than the
cases in favor of animals. - My aim in this presentation is to encourage
development of stronger arguments in favor of
animal use and provide methodological guidance on
how to do so. - Nathan Nobis For Animal Research in Theory
Practice, - ed. Jeremy Garrett, Rice, Philosophy
2Harms Moral Justification
- Many fields and occupations involve harming
animals, making them worse off. - Animals for our purposes, mammals birds
least controversial cases for discussion. - Typically, people in these fields will agree that
animals are being harmed. - They claim, however, that these harms are morally
justified not all harms are wrong, and these
harms arent wrong (indeed, perhaps some are
morally obligatory).
3Common experimental procedures include
- drowning,
- suffocating,
- starving,
- burning,
- blinding,
- destroying their ability to hear,
- damaging their brains,
- severing their limbs,
- crushing their organs
- inducing
- heart attacks,
- cancers
- ulcers
- paralysis,
- Seizures
- forcing them to inhale tobacco smoke, drink
alcohol, and ingest various drugs, such as
heroine and cocaine.
4A few commonly overlooked observations about harm
- (1) Painless killing can be (and often is)
harmful for the one who is killed it is bad for
him/her. - Why? They are deprived of whatever goods they
would have experienced. No interests can be
satisfied. - Thus, the common if painlessly killed, then
humane, so nothing morally objectionable views
need defense.
- (2) Recent ethological research shows that just
being in a laboratory, and undergoing routine
procedures, is stressful (and thus harmful) for
animals.
5Balcombe JP, Barnard ND, Sandusky C, Laboratory
routines cause animal stress, Contemporary
Topics in Laboratory Animal Science, 2004, Nov,
43 (6)42-51
- Abstract Eighty published studies were appraised
to document the potential stress associated with
three routine laboratory procedures commonly
performed on animals handling, blood collection,
and orogastric gavage . . . - Significant changes in physiologic parameters
correlated with stress . . were associated with
all three procedures in multiple species in the
studies we examined. - The results of these studies demonstrated that
animals responded with rapid, pronounced, and
statistically significant elevations in
stress-related responses for each of the
procedures . . . - We interpret these findings to indicate that
laboratory routines are associated with stress,
and that animals do not readily habituate to
them. The data suggest that significant fear,
stress, and possibly distress are predictable
consequences of routine laboratory procedures,
and that these phenomena have substantial
scientific and humane implications for the use of
animals in laboratory research.
6Balcombe JP, Laboratory environments and
rodents behavioural needs A review, Laboratory
Animals (in press)
- Abstract Laboratory housing conditions have
significant physiological and psychological
effects on rodents, raising both scientific and
humane concerns. Published studies of rats, mice
and other rodents were reviewed to document
behavioural and psychological problems
attributable to predominant laboratory housing
conditions. - Studies indicate that rats and mice value
opportunities to take cover, build nests,
explore, gain social contact, and exercise some
control over their social milieu, and that the
inability to satisfy these needs is physically
and psychologically detrimental, leading to
impaired brain development and behavioural
anomalies (e.g., stereotypies). To the extent
that space is a means to gain access to such
resources, spatial confinement likely exacerbates
these deficits. Adding environmental
enrichments to small cages reduces but does not
eliminate these problems, and I argue that
substantial changes in housing and husbandry
conditions would be needed to further reduce them.
7Many ethicists have argued that its wrong to use
animals these ways theyve given reasons for
their views and defended them
- utilitarianism and other consequentialisms,
- rights-based deontologies,
- ideal contractarianisms (veil of ignorance,
Golden rule ethics), - virtue ethics,
- common-sense (least harm, needless harm)
moralities, - religious moralities, feminist ethics,
- and more indeed almost every major, influential
perspective in moral theory.
8Even Kants, Rawls, and other moral theories
have been modified to be friendly to non-rational
moral patients (not moral agents)
- Improve the theory so there are direct duties to
baby ( other non-rational powerless humans
shes of moral value not because others care
about her, despite her not being a
moral agent, rational, etc.
9If the theory is now not Bad for Baby (and other
vulnerable humans), it is now not Bad for
Animals?
10Thus, an abundance of ethical resources in
defense of animals.
- However, this hasnt made much of a difference in
thought or deed regarding uses of animals. - Possible explanations
- big changes are always slow trickle-down is
slow - philosophers (and other thinkers and authors)
typically just arent very influential, - personal, financial, legal, political,
institutional barriers to doing the right thing, - ???
11A competing explanation
- There are strong arguments that morally justify
(much of) the current treatment of animals. - Since these arguments are strong / sound / very
reasonable to accept, the defenses of animals are
weak / unsound / unreasonable. - Im going to suggest that this explanation is
unlikely, because these arguments are weak. - I encourage development of more and stronger
arguments in favor of, defending, animal use and
provide methodological guidance on doing so.
12Emotional responses to moral issues
- It sometimes appears that the quality of our
thought on a topic is inversely proportional to
the intensity of our emotions concerning that
topic. - -- Fred Feldman, Confrontations With the Reaper
A Philosophical Study of the Nature and Value of
Death (Oxford, 1994).
13Rational engagement of moral issues
- Identify some past instances of moral progress
in thought, attitude, deed - Hopefully, rational evaluation of arguments
contributed to this, somewhat! - We can identify some basic logical skills that
can help us improve the quality of our thought. - Apply these skills to some recent arguments
made by scientists and philosophers regarding
animals. - This is important because it seems that not
enough people consistently use these skills this
is not good.
14Formerly controversial issues and simple
arguments
- Women shouldnt be allowed to go to university
because women are so emotional that abstract
thought is so difficult for them. - "Slavery is morally right because we slave-owners
benefit greatly from slavery." - "Since animals are not rational, it's morally ok
to raise them to be killed and eaten." - These are arguments what are their faults?
15Women (1)
- Conclusion
- Women shouldnt be allowed to go to university.
- Why think that?
- Women are such emotional beings that abstract
thought is difficult for them. - Imprecise! SOME? or ALL?
- Some women are so emotional that abstract
thought is difficult. True, and true for some
men! - All women are so emotional False,
empirically indefensible claim, so unsound
argument
16Women (2)
- Some women are so emotional that abstract
thought is difficult. True, and true for some
men! - Therefore, no women should be allowed to go to
university. - But how do you get from (1) to (2)? Whats the
missing linking premise? - A question How would some womens emotionality
justify restricting educational opportunities
from all women? Not clear.
17Women (3)
- However, even if some or even all women are so
emotional and have difficulty with abstract
thought why would that justify denying any women
the opportunity to improve themselves through
education? - If my cup won't hold but a pint, and yours holds
a quart, wouldn't you be mean not to let me have
my little half measure full? - Sojourner Truth, Ain't I A Woman? 1851
18Slavery
- "Slavery is morally right because we slave-owners
benefit greatly from slavery. - (1) Slave-owners benefit from slavery. True
- (C) Therefore, slavery is morally right. ?
- --------------------------------------------------
------------ - How do you get from (1) to (2)?
- Whats the missing, assumed linking premise?
- Slave-owners benefit from slavery. True
- If some group benefits from some arrangement,
then that arrangement is right. ? - Therefore, slavery is morally right.
19Animals
- "Since animals are not rational, it's morally ok
to raise them to be killed and eaten. - Animals are not rational.
- Therefore, its OK to kill them
- Observations and questions
- (1) is imprecise some, or all, animals are not
rational? Which animals? - Ambiguity, lack of clarity what is meant by
rational? - Missing-link premise needed to make argument
logically valid If a being is not rational,
then its ok to kill it. False?
20Logical skills The (moral) value of basic
predicate logic
- Attending to the intended meanings of unclear or
ambiguous words - what do you mean?
- animal, human, being human, human being,
person, human person, humanity - Precision regarding , quantity some, all?
- Assumed, unstated premises that link stated
reason(s) to conclusion. (Logical validity).
21It seems these logical skills are generally
useful.
- A bioethicist disagrees about the value of these
skills for professional ethics - Frankly, science students would have very little
patience for the abstract argumentation and
reasoning that one finds in your paper and is
standard fare in philosophy journals.
22Apply these ( other) logical skills to some
recent arguments
- Scientists
- Stuart Derbyshire, Ph.D., U Birmingham UK (used
to be at Pitt) pain researcher. - Mark Mattfield, Ph.D., Research Defense Society,
UK - Colin Blakemore, Ph.D., Medical Research Council,
UK - Adrian Morrison, Ph.D., DVM, U Penn, sleep
disorders
- Philosophers
- Carl Cohen
- Neil Levy, Cohen Kinds A Response to Nathan
Nobis, JoAP) - Tibor Machan, Putting Humans First
- Matthew Liao, Virtually All Human Beings as
Rightholders A Non-Speciesist Approach
23The issue neednt be whether animals have
rights
- Moral or legal rights?
- Which moral rights? (be specific)
- Rights conflicts right to smoke, right to
a smoke-free environment - Rights appeals can conceal details.
- Common invalid argument
- If animals have rights, then serious change is
needed. But they dont have rights, so change
isnt needed. - Logically invalid conclusion doesnt follow
and avoids the concrete questions.
24The issue neednt be whether animals have
rights
- Better to consider
- (1) whether various (specific) uses of animals
are morally permissible or not, whether any ways
of treatment are morally obligatory and - (2) why or why not.
- Keep this the focus on these deontic categories
is helpful for many practical and theoretical
reasons.
25The issue also neednt be whether animals are
equal to humans
- Are any animals equal to humans? Are all humans
equal? Hard to answer - What is meant by equal? Not obvious.
- Which humans, which animals? (What is meant by
humans and animals?). (fetus, baby, adult,
100 y/o?) - Common invalid argument avoids the concrete
questions. - If animals are equal to humans, then serious
change is needed. But they arent equal, so
change isnt needed. - Equal consid. vs.No consid.vs. mid-level
consid? - Again, ideal Qs are about moral permissibility.
26Objection An abundance of resources is a
philosophical embarrassment?
- Many philosophers argue that animals are treated
wrongly, but disagree on why (e.g., Peter Singer
demolishes Tom Regan and Regan demolishes
Singer). Therefore, there is no justification for
thinking that animals are treated wrongly. - Adrian Morrison Richard Vance, JAMA
27A parallel argument
- Many thinkers argue that animals are not treated
wrongly, but disagree on why (e.g., Carl Cohen
demolishes Jan Narveson Narveson demolishes
Cohen). Therefore, there is no justification for
thinking that animals are not treated wrongly.
28The false, unstated assumption
- If you believe p, and for reasons X, Y, Z, but
others believe p for reasons A, B, C, etc. and
these reasons are logically incompatible (and you
recognize this), then either you have no (good)
reason to believe p or there is no good reason to
believe p. - At the very least, this principle isnt one
typically accepted or universally applied (e.g.,
global warming is bad).
29Appeals to evolution / biological perspectives
- Morrison to refrain from exploring nature in
every possible way would be an arrogant rejection
of evolutionary forces. - Evolution has endowed us with a need to know as
much as we can. (Nicoll, Russell). - Humans evolved therefore, morally we should .
Does not follow. - Constraints on using other humans to advance our
own genetic line, when its in our interest?
30Benefits Arguments / Arguments from Necessity
- animal experiments are vital to the future
well-being of humans and, as long as they are
conducted to high ethical standards, they are
entirely justifiable. Mark Matfield - The argument Benefits for humans justify animal
experimentation (and other uses) - The are necessary.
31Is animal use necessary? (1)
- Depends on what you mean by necessary.
- In one sense, yes!
- To do animal experiments, it is necessary to do
animal experiments. To make these exact
scientific discoveries using animals, it is
essential to use animals if animals werent
used, the experiments would be different.
32Is animal use necessary? (2)
- In other senses, perhaps not. Is animal necessary
for making medical progress and for, more
generally, bettering human welfare? - Necessary for the well-being of humans, but
which humans? A few? (Maybe!). Everyone? Doubtful
that every human benefits from (every) animal
experiment. - There are other ways of bringing about goods for
humans - clinical research, epidemiology, in vitro
research, uses of technology, autopsies,
prevention, etc. - feeding people, getting existing medical care to
them, etc.. Its been argued that these would
yield greater human utility.
33Defenses of the low (human) utility of animal
experimentation
- RC Greek N Shanks, Animal Research in Light of
Science (2006? Rodopi) - N Shanks LaFollette, Brute Science (Routledge
1997) - RC Greek J Greek DVM, Sacred Cows Golden
Geese (Continuum 2000), Specious Science (2002),
What Will We Do if We Dont Experiment on
Animals? (2004) - They argue that other methods of research are
more effective at addressing human needs.
34Benefits argument
- Animal experiment yields some benefits.
- If some action benefits someone (or some group),
then that action is right. false needs
refinement and serious defense - Therefore, animal experimentation (and other
uses) are right. - What about direct harms (to animals, to humans,
esp. indirect harms from opportunity costs)? How
are these weighed? A careful methodology would be
nice, at least is necessary for serious defense.
35Want benefits?
- Whatever benefits animal experimentation is
thought to hold in store for us, those very same
benefits could be obtained through experimenting
on humans esp. vulnerable ones instead of
animals. Indeed, given that problems exist
because scientists must extrapolate from animal
models to humans, one might think there are good
scientific reasons for preferring human
subjects. - Philosopher Ray Frey
36Why not use these humans?Blakemores answer
- The only firm line to make moral distinctions
on genetic and morphological grounds is between
our own species and other species. - Suggested if something is of our species, then
it is more morally valuable than any animals. - But he says a human embryo, certainly before the
nervous system begins to develop, is just a
bundle of cells. - Suggested being of our species does not
necessarily confer moral value. - We should have a special attitude toward other
humans, so crucial to this argument is how we
define a person. He did not do this.
37Why not use these humans?Derbyshires answer
- Animals lack the capacity for reflection (and
therefore an inner world) and the capacity for
reasoning (So do many humans!!) - Its remarkable that we have to consider the
question. - Not remarkable if someone suggests that whats
required for a presumption against harm are
properties that many, many human beings lack. - Society cares about vulnerable humans.
- All of them? What about secret experiments? What
if they could be re-educated? Why do they care?
(Harms)
38Avoiding objections from non-rational human
beings.
- A common claim
- Its wrong to seriously harm a being only if that
being is rational, autonomous, makes moral
choices, is creative, intelligent, contributes to
society, etc. - OK, animals arent like that, but neither are
lots of (conscious, feeling) humans. This
principle suggests its not seriously wrong to
harm them. Is this principle correct?
39Some odd inferencesCohen, Levy Kinds
- Cohen NEJM Moral rights depend on moral
agency, the ability to respond to moral claims.
A being has rights only if its a of a kind
characterized by moral agency. - Finnis to be a person is to belong to a kind of
being characterized by rational (self-conscious,
intelligent) nature. - Scanlon the class of beings whom it is possible
to wrong will include at least all those beings
who are of a kind that is normally capable of
judgment-sensitive attitudes.
40Cohen, Levy Kinds
- Cohen All humans are of a kind capable of moral
agency, but - animals are not beings of a kind capable of
exercising or responding to moral claims. Animals
therefore have no rights, and they can have
none. - What kind are animals? How are humans who are not
moral agents of the kind moral agent? Cohen
doesnt explain.
41Cohens possible answer?
- Humans who are non-moral agents are of this kind
because they are members of a set e.g., the
kind, a species some of which are moral agents.
- Response But animals are also members of a set
e.g., the kind, sentient beings some of whom
are moral agents also! They have rights too, on
Cohens account! - Humans and animals are of many kinds, some
overlapping, some not. Inconsistent conclusions
follow from Cohen-esque reasoning.
42Levys attempt to find the right kind the
narrowest natural kind
- If (1) an individual A is a member of some
species S and (2) some, most or all of the other
members of that species have some property C and
(3), on the basis of having property C, they have
moral property R, then individual A has moral
property R as well, even though A lacks property
C.
43If (1) an individual A is a member of some
species S and (2) some, most or all of the other
members of that species have some property C and
(3), on the basis of having property C, they have
moral property R, then individual A has moral
property R as well, even though A lacks property
C.
- C non-moral property of "having doneno serious
crimes - R "not deserving lifeimprisonment."
- Implications for lone criminal?
- C "intelligent" and "aware
- R "being such that one ought to be allowed to
make decisions to direct one's own life." - Implications for young children and others?
44Machans Arguments from Whats Normal
- A being has moral rights (presumably making it
wrong to harm it) only if it a moral nature, a
capacity to see the difference between right
and wrong and choose accordingly. - It is this moral capacity that establishes a
basis for rights, not the fact that animals, like
us, have interests or can feel pain. - Humans are of the kind of being that have such
a moral nature and animals are not thus humans
have rights and animals do not.
45What about humans who seem to lack this moral
capacity?
- We must consider humans as they exist normally,
not abnormally and focus on the healthy cases,
not the special or exceptional or borderline
ones. - We do need to deal with borderline cases. But we
can do so only by applying and adapting the
knowledge we acquire from the normal case. We
cant start with the exception and infer the
rule.
46The suggested argument
- Humans who lack moral capacities are human.T!
- If someone is human, then they have all the
(moral) properties that normal, healthy,
typical humans have. - Therefore, these humans have moral capacities,
and so they have rights. - Reply 2 is, at least, unsupported, and is an
instance of a generally false principle for moral
non-moral properties. (e.g., 4 limbs Ted Bundy)
47Matthew Liao, Virtually All Human Beings as
Rightholders The Species-Norm Account
- to be a rightholder (a being with the highest
moral status), something need not - be a moral agent
- have the potential to be a moral agent
- be of the kind (species) that normally is a moral
agent - be actually sentient, conscious, etc. or even
have the potential, i.e., that its possible in
some sense - Could be tinkered into a pro-animal exper. view.
48The correct answer is
- A being has rights iff the entity has
incorporated into it the genetic basis for the
species capacity for moral agency (i.e. the
relevant bits of DNA that normally allow for
moral agency) or the functional equivalent
thereof (e.g. software and/or hardware that would
normally allow for moral agency in an artificial
being). The intrinsic value that resides in the
relevant genetic bits grounds rightholding even
when that genetic material is blocked from
developing and cannot allow for moral agency. - If X is like that, then X has moral rights.
49Liaos reasoning in favor of the view, it seems
- There are moral duties only if there are moral
agents. T - There are moral agents only if there are beings
with the genetic basis for moral agency. OK
accept this for sake of argument - Therefore, there are moral duties only if there
are beings with the genetic basis for moral
agency. - Therefore (?), any being with the genetic basis
for moral agency is a rightholder.
50A parallel argument
- There are moral duties only if there are living
beings, or beings that can perceive, or . T - There are living beings, or beings that can
perceive only if there are beings with the
genetic basis for life, perception, etc. OK - Therefore, there are moral duties only if there
are beings with the genetic basis for life,
perception, etc. - Therefore (?), any being with the genetic basis
for life, perception, is a rightholder.
51Objections from Chris Grau, FIU
- If the species-norm account is true, then
- A cabbage that has "integrated" the relevant
genetic bits but is damaged such that the
capacity for moral agency is permanently blocked.
(this cabbage has rights even though it lacks
moral agency and the potential for it.) - A (future) computer with the relevant hardware.
software "integrated" but blocked. This computer
has rights even though it would lack both moral
agency and the potential for moral agency. - Cabbage or computer vs. sentient animals and
sentient humans lacking the relevant genetic
material for moral agency? - The species norm account seems entirely ad hoc.
52Conclusions / Summary
- Presented a basic method for thinking about moral
issues demonstrated its use applied it to some
recent arguments defending current animal use
and/or criticizing pro-animal arguments. - Suggested that these arguments are weak.
- My hope since these methods are generally
useful, perhaps future defenders of current uses
of animals will utilize them for better
arguments. - To make moral progress and contribute to
reasonable debate it is important that this is
done.
53For an overview of the recent literature on
ethics and animals issues, see Angus Taylors
Ethics Animals An Overview of the
Philosophical Debate (Broadview, 2003). For
arguments from utilitarianism, see, among other
sources, Peter Singers Practical Ethics, 2nd
Edition (Cambridge UP, 1993) and his Animal
Liberation, 3rd Edition (Harper, 2001) although
the former is, strictly speaking, not an argument
from utilitarianism. From rights-based
deontology, see, among other sources, Tom Regans
The Case for Animal Rights, 2nd Edition (U
California Press, 2004), as well as his more
accessible Empty Cages Facing the Challenge of
Animal Rights (Rowman Littlefield, 2004) for
Rawlsian-style ideal contractarianism, see among
other sources, Mark Rowlands Animals Like Us
(Verso, 2002) from virtue ethics, see among
other sources, Rosalind Hursthouses Ethics,
Humans and Other Animals (Routledge, 2000), from
common-sense morality, see, among other sources,
Mark Bernsteins Without a Tear Our Tragic
Relationship With Animals (U Illinois Press,
2004) and David DeGrazias Animal Rights A Very
Short Introduction (Oxford, 2002) for religious
moralities, see, among other sources, Matthew
Scullys Dominion The Power of Man, the
Suffering of Animals, and the Call to Mercy (St.
Martins, 2003) from feminism, see Carol Adams
and Josephine Donavan (eds.) Beyond Animal
Rights A Feminist Caring Ethic for the Treatment
of Animals (Continuum, 1996).
54Stuart Derbyshire,an animal experimentation
advocate
- It is not possible to advocate animal welfare
and at the same time give animals untested drugs
or diseases, or slice them open to test a new
surgical procedure. . . - The Scientist, 3/06, Time to Abandon the Three
Rs - Submitting to refinement, reduction, and
replacement risks the future of animal research
- Once the perspective of the animal is adopted,
it is inevitable that all experimentation will be
seen negatively, as no animal experiments are in
the interest of the animal - - Why Animals Rights Are Wrong (p. 39)