5 bermudez thinking without words

JOSE LUIS BERMUDEZ

THINKING WITHOUT WORDS:AN OVERVIEW FOR ANIMAL ETHICSw

(Received and accepted: 3 April 2006)

ABSTRACT. In Thinking without Words I develop a philosophical framework fortreating some animals and human infants as genuine thinkers. This paper outlines

the aspects of this account that are most relevant to those working in animal ethics.There is a range of dierent levels of cognitive sophistication in dierent animalspecies, in addition to limits to the types of thought available to non-linguistic

creatures, and it may be important for animal ethicists to take this into account inexploring issues of moral signicance and the obligations that we might or might nothave to non-human animals.

KEY WORDS: animal cognition, animal ethics, thought without language

In Thinking without Words, I develop a philosophical frameworkfor treating (at least some) animals and human infants as genuinethinkers.1 A genuine thinker, I take it, is a creature that behaves inways that reect its thoughts about the environmentand hence acreature whose behavior needs to be explained in psychologicalterms. That many animals are genuine thinkers is taken for grantedby much research in cognitive ethology, but scientists andphilosophers have often been skeptical of what they take to beanecdotal evidence and tacit anthropomorphism. The aims of mybook are

to set out clear criteria for identifying when psychologicalexplanations are required for non-linguistic creatures;

to show how precise and determinate thoughts can be attrib-uted to non-linguistic creatures;

w I am grateful for comments on an earlier version from Robert Francescotti and

Clare Palmer.1 J. L. Bermudez, Thinking without Words (Oxford: Oxford University Press,

2003).

The Journal of Ethics (2007) 11:319335 Springer 2007DOI 10.1007/s10892-007-9013-8

to show how the psychological explanations that we give ofanimal and infant behavior are continuous with the psycholog-ical explanations that we give of language-using creatures;

to explore the dierences between thinking without words andlanguage-based thinking.

This paper outlines the aspects of this account that are of mostrelevance to those working in animal ethics. There is a range ofdifferent levels of cognitive sophistication in different animal species,in addition to limits to the types of thought available to non-linguisticcreatures, and it may be important for animal ethicists to take thisinto account in exploring issues of moral signicance and theobligations that we might or might not have to non-human animals.

1. PSYCHOLOGICAL AND NON-PSYCHOLOGICAL EXPLANATIONSOF BEHAVIOR

Psychological explanations come into play when non-psychologicalforms of explanation provide insufcient explanatory and predictiveleverage. Typical non-psychological forms of explanation appeal tomechanisms of associative conditioning and what are known asinnate releasing mechanisms.

Pavlovian or classical conditioning occurs when an association isreinforced between an unconditioned stimulus (e.g., food or pain) and aconditioned stimulus (e.g., the sound of a bell). The unconditionedstimulus typically generates an unconditioned response (e.g., saliva-tion). As a result of conditioning the unconditioned response isgenerated by the conditioned stimulus. Many forms of animal trainingare based on classical conditioning. It is classical conditioning thatmakes clicks and whistles eective rewards for dogs and dolphins. Ininstrumental (or operant) conditioning the process of reinforcement(or punishment) applies to actions rather than physiological responses.

According to behaviorist models of animal behavior, all animalbehavior is the product of either classical or instrumental condition-ing, and conditioning is certainly the type of animal learning mostfrequently studied in the laboratory. Ethologists, however, have alsoappealed to innate releasing mechanisms to explain behavior in thewild.2 Innate releasing mechanisms are xed and instinctive sequences

2 See, for example, N. Tinbergen, The Study of Instinct (Oxford: Oxford

University Press, 2003).

JOSE LUIS BERMUDEZ320

of movements. For example, when newly hatched herring gullsencounter stimuli matching the adult herring gull beak in color,length, and movement they respond by pecking. Innate releasingmechanisms have the following characteristics.3

They are triggered by specic stimuli. They always take the same form. They occur in all members of the relevant species. Their occurrence is largely independent of the individual

creatures history. Once launched they cannot be varied. They have only one function.

Innate releasing mechanisms and conditioned responses are bothinvariant responses to stimuli. When the animal registers the relevantstimulus the appropriate response results in a way that can in generalbe fully understood, explained and predicted without any appeal to anintermediary between stimulus and response.4 Psychological explana-tions of behavior only become necessary when no such invariantinputoutput links can be identied. The essence of a psychologicalexplanation is that it explains behavior in terms of how the creature inquestion represents its environment, rather than simply in terms of thestimuli that it detects. Psychological explanations typically makereference to how the organism perceives its environment, to what itbelieves about the environment, and to what it desires to achieve.These beliefs, desires, and perceptions allow organisms to respondexibly and plastically to their environmentsthe same situations canaord dierent actions if a creature brings dierent beliefs and desiresto it, or perceives it in dierent ways.

How are we to determine which animals count as genuinethinkers? By identifying species whose members behave in ways thatdo not seem to be explicable in non-psychological terms. Any suchjudgment is provisional and defeasible, since it might always turn outthat we have been insufciently imaginative in thinking about thenon-psychological possibilities. What is not provisional and defeasi-ble, however, is the judgment that many species will prove to contain

3 S. E. G. Lea, Instinct, Environment, and Behavior (London: Methuen, 1984).4 But see A. Dickinson and B. Balleine, Actions and Responses: The Dual

Psychology of Behavior, in N. Eilan, B. Brewer, and R. McCarthy (eds.), SpatialRepresentation (London: Basil Blackwell, 1993) for examples of types of conditioned

response that do seem to require a psychological explanation.

THINKING WITHOUT WORDS 321

genuine thinkers. The weight of the evidence points strongly to theimpossibility of characterizing all animal behavior in non-psycho-logical terms.

There is a basic distinction, then, between creatures that behave inways that require psychological explanation and those that do not.This may mark a morally signicant dividing line. It is worth noting,however, that sentience seems to be required for some forms ofassociative learning. If the unconditioned stimulus is pleasure or pain,or anything whose status as reward/punishment is a function of itsphenomenal character, then only sentient creatures are capable oflearning through conditioning. If sentience is what matters for moralsignicance, then it is already built into some non-psychologicalmodels of explanation. Nonetheless, it is hard to imagine that moralsignicance is not a matter of degreeand even if it is not a matter ofdegree we will still need to make judgments of relative moralsignicance. Either way the distinction between merely sentientcreatures and thinking creatures is likely to be relevant.

2. PROPOSITIONAL AND NON-PROPOSITIONAL THINKINGIN NON-LINGUISTIC CREATURES

It might be accepted that animals of a certain species at a certainstage of development are thinkers, in the sense that they behave (atleast some of the time) in ways that require psychological explana-tion. But there are different types of thinking at the non-linguisticlevel. The basic distinction is between propositional and non-propositional thought.

According to Michael Dummett, the types of thinking available toanimals are just a subset of the central types of thinking available tolanguage-using creatures.5 Dummett accepts that there can be non-linguistic thoughts, which can be had both by animals and bylanguage-using creatures, but he calls them proto-thoughts. Theseproto-thoughts do not have the structure of verbally expressedthoughts; they are not full-edged thoughts; they cannot oatfree [of the environment], but can occur only as integrated withcurrent activity; and the vehicle of non-linguistic thought is spatialimages superimposed on spatial perceptions. There can be non-

5 Michael Dummett, The Origins of Analytical Philosophy (London: Duckworth,

1993).


linguistic thoughts, but these are not accurately expressible inlanguage.6 This is what I call the minimalist conception of non-linguistic thought.

According to the minimalist approach, all non-linguistic thinking is

context-bound; essentially pragmatic and dynamic; vehicled by spatial images superimposed on spatial perceptions; unstructured.

Proto-thoughts thus construed count as instances of thinking-howrather than thinking-that (to draw an analogy with Gilbert Ryleswell-known distinction between knowing-how and knowing-that7).Dummett explicitly assimilates them to complex behavioral skills.Their purpose is essentially the control of responses to the environ-ment, rather than the acquisition of information about it. They donot have a determinate content that can be put into words. In allthese respects they are fundamentally dierent from beliefs, desires,and other propositional attitudes.

One of the central claims of Thinking without Words is that theminimalist conception cannot be a complete account of non-linguisticthought. Non-linguistic thought goes beyond perception, becausethere are forms of animal behavior that we can only explain bythinking of the creatures performing those actions as having full-edged beliefs and desires. By this I mean that these are beliefs anddesires that represent the world in ways that can be accuratelyreported in sentences of something not too dissimilar to Englishbutnot identical to English, since we will need a vocabulary that reectsthe dierences between how we carve up the world into objects andhow the environment is perceived by dierent types of animal.Psychological explanations of this type are propositional attitudeexplanations.

The types of behavior that most obviously pose problems for theminimalist conception are those that go beyond the here and now.When animals represent contingencies between actions andoutcomes, perhaps in thinking about how to tailor means to ends,they are going beyond the sensorimotor schemas envisaged by theminimalist conception. Similarly, when they engage in tool use andother forms of long-range planning. Wild chimpanzees, for example,

6 Dummett, The Origins of Analytical Philosophy, pp. 122123.7 Gilbert Ryle, The Concept of Mind (London: Hutchinson, 1949), Chapter 2.


make two different types of wands for dipping into ant and termitenests from different types of branches.8 They make wands for dippinginto ant swarms by taking a stick several feet long and stripping theside leaves and leafy stem. For dipping into termite nests, on theother hand, they use wands made from vines or more exible twigsthat are considerably shorter and that have a bitten end, unlike theant wands.

Of course, as with the initial determination of whether we aredealing with thinkers at all, careful experimental work is required toidentify when propositional explanations are required. This isprobably the most intensively studied and controversial area ofanimal cognition. It is also potentially the area of most interest toanimal ethics. This is also the area where most philosophical work isrequired to explain the truth-conditions of the thoughts ascribed tonon-linguistic creatures and how to go about attributing suchthoughts.9

If it is argued that moral signicance depends upon the capacityfor genuine thought, then it is natural (but not, of course,compulsory) to think that there are degrees of moral signicancecorrelated with degrees of cognitive sophistication. The dividing linebetween thinking of the minimalist kind and thinking of thepropositional kind may well be important.

3. THE LIMITS OF NON-LINGUISTIC THOUGHT:INTENTIONAL ASCENT AND SEMANTIC ASCENT

As far as animal ethics is concerned, what animals cannot do is likelyto be just as important as what animals can do. Many discussions ofmoral signicance make it contingent upon particular types ofcognitive achievement. Some of these cognitive achievements couldbe implicated in the types of non-linguistic thinking we have alreadydiscussed. Suppose, for example, that moral signicance werethought to be restricted to creatures capable of a concern for theirown future. One might take the ability to engage in certain types oflong-range planning to be evidence for such concern.

8 See R.W. Byrne, The Thinking Ape (Oxford: Oxford University Press, 1995),p. 97.

9 See Bermudez, Thinking without Words, Chapters 4 and 5, and, for a shorteroverview, J. L. Bermudez, Ascribing Thoughts to Non-Linguistic Creatures, Facta

Philosophica 5 (2003), pp. 313334.


There is a number of ways of thinking about moral considerabil-ity, however, that cannot in principle be applied to non-linguisticcreaturesor, at least, not if one of the central arguments of Thinkingwithout Words is sound. In Chapter 8, I argued that higher-orderthought (thinking about thinking) is language-dependent. In thissection, I shall present a revised version of the basic argument. In thefollowing sections, I draw out some of the limitations that theunavailability of higher-order thoughts has upon animal cognition.

By a higher-order thought I mean a thought that takes anotherthought as its object. Thoughts about anothers mental states countas higher-order thoughts, for example, as does reection on onesown mental states. W. V. O. Quine once described semantic ascent asthe shift from talking in certain terms to talking about them.10 Byanalogy we can characterize intentional ascent as the shift fromthinking in certain ways to thinking about those ways of thinking.My argument, in eect, is that intentional ascent requires semanticascentthat we can only think about thoughts through thinkingabout words.

We should distinguish rst-order target thoughts from the higher-order thoughts that might be directed at them. My belief that p is atarget thought. It is the object of my higher-order belief that I believethat p. Target thoughts must be represented to be the objects ofhigher-order thoughts. There are all sorts of things going on belowthe threshold of consciousness when we think (perhaps thinkinginvolves manipulating sentences in a subpersonal language ofthought, for example). But these subpersonal events are not whatwe think about when we think about our own thoughts. There is adierence between thinking about thoughts and thinking about themachinery of thinking. So the question is: How must target thoughtsbe represented in order for them to be the objects of higher-orderthoughts?

There are two possibilities. On the one hand representation mightbe secured symbolically through the complex symbols of a naturallanguage. A thought would be represented, therefore, through itslinguistic expression and would appear as a potential object ofthought qua linguistic entity. On the other hand representation mightbe secured in an analog manner, through some kind of pictorialmodel. On this conception of the vehicles of thought, which we nd

10 W.V.O. Quine, Word and Object (Cambridge: Harvard University Press, 1960),

p. 271.


developed in dierent ways in mental models theory in the psychol-ogy of reasoning, and in the conception of mental maps put forwardby D. Braddon-Mitchell and F. Jackson, the vehicle of a thought is apictorial representation of the state of aairs being thought about.11

The argument in favor of public language sentences and againstpictorial models rests upon considerations of structure and inferentialrole. I am assuming that thoughts are individuated at least in part bytheir inferential role. What makes a given thought the thought that pis partly a matter of the inferential relations in which it stands toother thoughts. Some of these relations are entailment relations (thethoughts that entail p and the thoughts that p entails), but they alsoinclude evidential relations (the thoughts whose holding true wouldbe good evidence for thinking that p holds true, and the thoughts thatwould be judged more likely to be true if p were true). Any thinkercapable of thinking a higher-order thought directed at a targetthought must, almost by denition, have some grasp of theindividuation conditions of the target thought. He must have somegrasp of what it is that he is thinking about. There is nothing peculiarhere to higher-order thoughts. This is just an application of the verygeneral requirement that to think about anything one must havesome sort of cognitive access that enables one to pick that thingout. It follows that a higher-order thinker must have some sort ofgrasp of the entailment and evidential relations in which the targetthought stands.12

At least some of these entailment and evidential relations are afunction of the structure of the thought that p. In order to understandthe inferential role of a thought we need to be able to view it as madeup of distinguishable components that can feature in further thoughtsand, moreover, we need to be able to view it as made up from thosecomponents in a way that determines its semantic value (therebycapturing the dierence between the true thought Bogota is the capitalof Colombia and the false thought Colombia is the capital of Bogota).

11 The theory of mental models was rst proposed in K. Craik, The Nature ofExplanation (Cambridge: Cambridge University Press, 1943), and is most compre-hensively developed in P. Johnson-Laird, Mental Models (Cambridge: Cambridge

University Press, 1983). For mental maps see D. Braddon-Mitchell and F. Jackson,The Philosophy of Mind and Cognition, 2nd Edition (London: Blackwell Publishers,2006), Chapter 10.

12 The thinker who merely thinks such thoughts (as opposed to thinking aboutthem) does not have to grasp these entailment and evidential relations. They simply

have to think in ways that respect them.


We may say, therefore, that the structure of the thought must beperspicuous in the consciously accessible representation that is thetarget of the higher-order thought.

The nal step in the argument is that the structure of a thoughtcannot be perspicuous in the right sort of way in thoughts that arerepresented in a pictorial manner. The qualication is important,since pictorial representation in mental maps and mental models doesdepend upon a notion of structural isomorphism between the models/maps and what they represent. The relations holding betweenelements of the mental model/map can be mapped onto the relationsholding between objects in the represented state of aairs. This comesacross very clearly in the following passage from Braddon-Mitchelland Jackson.

There is no natural way of dividing a map at its truth-assessable representationaljoints. Each part of a map contributes to the representational content of the whole

map, in the sense that had that part of the map been different, the representationalcontent of the whole would have been different. Change the bit of the map of theUnited States between New York and Boston, and you change systematically whatthe map says. This is part of what makes it true that the map is structured. How-

ever, there is no preferred way of dividing the map into basic representationalunits. There are many jigsaw puzzles you might make out of the map, but no sin-gle one would have a claim to have pieces that were all and only the most basic

units.13

We might gloss this as follows. Pictorial representations do not havea canonical structure. Their structure can be analyzed in manydierent ways (corresponding to the jigsaw puzzles that one canconstruct from it), but none of these can properly be described asgiving the structure of the state of aairs.

Yet, in order to understand the inferential role of a thought onedoes need to understand the canonical structure of that thought(what is often termed its logical form). This canonical structure isperspicuous, although not always perfectly perspicuous, whenthoughts are expressed in public language sentences. It is becauseof this that higher-order thought is language-dependent. Only publiclanguage sentences can make the canonical structure of a targetthought available to thinkers in a way that allows them to grasp theinferential role of the target thought. The conclusion of the argument,then, is that thinking about thinking is only available to language-using creatures.

13 Braddon-Mitchell and Jackson, The Philosophy of Mind and Cognition, p. 171.


4. WHAT THE ARGUMENT DOES NOT Show:Sentience and Higher-Order Thoughts

Some philosophers have proposed higher-order thought theories ofconsciousness.14 According to these theories a mental state isconscious if and only if it is the object of a higher-order thought.Given that sentience just is the capacity to have conscious experi-ences, higher-order thought theories of consciousness restrict sen-tience to creatures capable of thinking higher-order thoughts. Anysuch conclusion is potentially very important for animal ethics, giventhe weight that is standardly put on animal suering in thinkingabout our obligations to non-human animals.

It should be stressed, however, that this conclusion does not in anysense follow from the argument from intentional ascent to semanticascent. The argument presupposes a theory of consciousness. It doesnot set out to provide one and it is perfectly compatible with the viewthat non-human animals can not only have conscious experiences(and so be sentient) but also have conscious beliefs and desires. Theobject of a conscious belief is a state of affairs in the world (or, in thecase of a false belief a merely possible state of affairs) and theargument from intentional ascent to semantic ascent applies only tothoughts that have other thoughts as their objects. As such it offersno direct support to arguments that animals cannot be sentientbecause they are not capable of having higher-order thoughts.15

5. INTENTIONAL ASCENT AND UNDERSTANDING OTHER MINDS

Is it possible for non-linguistic creatures to participate in practices ofattributing psychological states to their conspecics or indeed to anyother creatures? In the light of the preceding discussion it is not hardto see why a very broad class of psychological attributions should be

14 See, for example, D. Rosenthal, Two Concepts of Consciousness,Philosophical Studies 49 (1986), pp. 329359.

15 It may provide indirect support, however, on some ways of developing higher-order theories of consciousness. Authors such as P. Carruthers have argued that thetype of higher-order thoughts required for consciousness are only available to

creatures possessing a theory of mind and as we will see in more detail in the nextsection, the argument from intentional ascent to semantic ascent does rule out the-ories of mind at the non-linguistic level [P. Carruthers, Natural Theories of Con-

sciousness, European Journal of Philosophy 6 (1998), pp. 203222].


unavailable to non-linguistic creatures. To attribute a belief, forexample, to another creature is essentially to view that creature asstanding in a particular relation to a thoughtthe relation ofbelieving the thought to be true. Clearly, therefore, the attribution ofa belief requires thinking about a thought. It is a canonical form ofintentional ascent that requires being able to hold a thought inmind.

This has potential implications for animal ethics, on any view thatlinks moral signicance to the capacity to engage in certain types ofreection about the mental states of conspecicsor to the capacityto engage in types of behavior (perhaps caring behavior) thatpresupposes and involves such reection. It is important to recognize,however, that the argument from intentional ascent to semanticascent does not leave non-linguistic creatures completely mind-blind. There are types of mental state that can be comprehended andattributed by non-linguistic creatures.

To explain this further, we need to distinguish two ways ofthinking about desire.16 One can desire a particular thing, or one candesire that a particular state of aairs be the case. This is thedistinction between goal-desires and situation-desires. At the level ofverbalizable thought, the distinction can be marked in terms of twodierent ways of completing the sentence X desires . A sentenceascribing a goal-desire is completed by the name of an object or bythe name of a kind of stu (e.g., food). But when a sentenceascribes a situation-desire, it is completed by a that clause inwhich the blank is lled by a complete sentence specifying the state ofaairs in question.

Goal-desires are more basic than situation-desires. The contrast iseffectively between desire construed as a propositional attitude(in situation-desires, which are attributed via that-clauses picking outthe thought that is the object of desire) and the more fundamentalgoal-desires that are directed not at thoughts but rather at objects orfeatures. There is no reason why non-linguistic creatures should notbe able to attribute goal-desires to other agents. The argument fromintentional ascent cannot get a grip, since goal-desires are relationsbetween a subject and an object/feature, rather than between asubject and a proposition.

The ability to attribute goal-desires goes hand in hand with a basicunderstanding of intentional, that is to say goal-directed, behavior.

16 See Bermudez, Thinking without Words, pp. 4849.


Although of course there will be many different degrees of complexityin goal-directed behavior, depending on the richness of the desiresand beliefs by which it is driven, a creature capable of attributinggoal-desires will be able to make the basic distinction betweenpurposeful behaviors, on the one hand, and random movements andinstinctive reactions on the other. A purposive action is an action forwhich a motivating goal-desire can be identied.

Goal-desires cannot be the only mental states that can beidentied and attributed by non-linguistic creatures. It is hard tosee, for example, how a goal-desire can be attributed to a creaturewithout some evidence of the information that the creature possessesabout its environment. At the bare minimum this information will beperceptual. To know what goal-desire might be motivating a creatureat a given moment a creature needs to know, rst, what end it ispursuing and, second, how it might reasonably expect that end to berealized by its current behavior. Both of these require knowing towhich features of its environment the creature is perceptuallysensitive. If, therefore, a non-linguistic creature is to be able toattribute goal-desires to a fellow creature it must be able to formulatehypotheses about what that creature is perceiving.

Here too we can distinguish two ways of thinking about seeing byfollowing F. Dretske, making a distinction between simple seeing andepistemic seeing. According to Dretske, what we see in simple seeing(or what he calls non-epistemic seeing) is a function solely of whatthere is to see and what, given our visual apparatus and theconditions in which we employ it, we are capable of visuallydierentiating.17 In contrast, epistemic seeing involves standing in arelation to a proposition (a thought). Epistemic seeing involves seeingthat something is the case.

The argument from intentional ascent shows that non-linguisticcreatures are not capable of understanding epistemic seeing, since thisinvolves thinking about the perceivers relation to a thought. But thisis perfectly compatible with non-linguistic creatures being capable ofthinking about the direct perceptual relations in which othercreatures stand to objects. This allows non-linguistic creatures toengage in a primitive form of psychological explanation. A creaturethat knows what a conspecic or predator desires and has some senseof its perceptual sensitivity to the environmental layout (as well as an

17 F. Dretske, Seeing and Knowing (London: Routledge, 1969), p. 76.


understanding of its motor capabilities) can expect to be able topredict its behavior with some success.

This restrictive interpretation of the mind-reading abilities ofsome non-linguistic creatures is compatible with much recent researchinto the extent to which non-human primates can properly bedescribed as possessing a theory of mind. There are well-docu-mented examples of primate behavior that some prominent studentsof animal behavior have thought can only be interpreted as examplesof interpersonal deception.18 But the consensus opinion amongprimatologists is that a more parsimonious interpretation of thesebehaviors is to be preferred.19 Many examples of what has come to betermed tactical deception can be understood as the manipulation, notof anothers propositional attitudes, but simply of their visualperspective. Here is an example of a tactical deception in a troupeof baboons in Ethiopia that lends itself to such an interpretation:

An adult female spent 20 min gradually shifting in a seated position over a dis-tance of about 2 m to a place behind a rock about 50 cm high where she began togroom the subadult male follower of the groupan interaction not tolerated by

the adult male. As I was observing from a cliff slightly above [the animals] I couldjudge that the adult male leader could, from his resting position, see the tail, backand crown of the females head, but not her front, arms and face: the subadultmale sat in a bent position while being groomed, and was also invisible to theleader. The leader could thus see that she was present, but probably not that shegroomed.20

The behavior of the female baboon, assuming that it is indeed tocount as an instance of tactical deception, does not seem to requirerepresenting the beliefs of the alpha male. What she is doing isproting from an understanding of the alpha males visual take onthe situation to escape detection. The female baboon needs only toappreciate the alpha males line of sight and the fact that he would beprevented from seeing the subadult male by the intervening rock. Thisseems rmly at the level of simple seeing rather than epistemic seeing.

18 See, for example, the papers in R. W. Byrne and A. Whiten (eds.), Machia-vellian Intelligence (Oxford: Oxford University Press, 1995); D. Premack andG. Woodru. Does the Chimpanzee Have a Theory of Mind? Behavioral andBrain Sciences 1 (1978), pp. 515526; and F. de Waal, Chimpanzee Politics (London:

Jonathan Cape, 1982).19 E.g., D. Povinelli, Chimpanzee Theory of Mind, in P. Carruthers and P. K.

Smith (eds.), Theories of Theory of Mind (Cambridge: Cambridge University Press,1996); and M. D. Hauser, Wild Minds (London: Penguin Books, 2000).

20 Report by Hans Kummer quoted in Byrne, The Thinking Ape, p. 106.


This example (and the discussion of animal mind-reading moregenerally) shows that the argument from intentional ascent tosemantic ascent is compatible with taking non-linguistic animals tohave fairly sophisticated cognitive abilities. The argument requirestheorists to think critically about some contemporary research inanimal cognition. It places limits on the conceptual abilities that canbe attributed to non-linguistic creatures, but in so doing opens upnew ways of interpreting the behaviors revealed by observation andexperiment.

6. REASONING, RATIONALITY, AND LOGICAL THINKING

Explaining animal behavior in psychological terms forces us toconsider questions of rationality and reasoning. Psychologicalexplanations work because they identify beliefs and desires in thelight of which the action being explained makes sense from the agentsperspective. To say that an action makes sense in the light of anagents beliefs and desires is to say that it is the rational thing to do(or, at least, a rational thing to do) given those beliefs and desires.And that in turn means that, in at least some cases, a creature mightreason her way from those beliefs and desires to acting in the relevantway. Reasoning and rationality are correlative notions. How shouldwe make sense of those notions at the non-linguistic level?

Here is one way in which we cannot make sense of them. Theargument from intentional ascent stands squarely in the way oftreating animals as thinking logically. We can illustrate this with themost basic form of logical thinkingthe form of thinking codied inthe propositional calculus and involving the basic logical connectives,such as disjunction (or), conjunction (and), and the materialconditional (if... then ...). Consider a conditional thought of thesort that might be expressed in the sentence if A then B. Toentertain such a thought is to understand that two thoughts arerelated in a certain waynamely, that the second thought cannot befalse if the rst thought is true. But this means that understandingtruth-functional compound thoughts is a form of intentional ascent.One cannot think about the truth-values of thoughts withoutthinking about thoughts and this, by the earlier argument, requiressemantic ascent.

Logical thinking depends upon language, therefore, because itpresupposes the capacity for intentional ascent, which in turn


depends upon semantic ascent. This poses an obvious challenge forhow we think about reasoning in animals. The challenge is to identifyforms of reasoning at the non-linguistic level and then explain themwithout assuming that the animal (or prelinguistic infant) isdeploying elementary logical concepts. I will illustrate how thischallenge can be met for a very basic form of reasoning. This isstraightforward conditional reasoning of the type formalized asmodus ponensreasoning that is standardly thought to exploit thevalidity of the inference from if A then B and A to B. Thedetection of patterns of behavior is closely bound up with thepossibility of conditional reasoning. A creature that knows that if thegazelles see the lion they will run away and that recognizes (perhapson the basis of its understanding of the gazelles visual perspective)that the lion will shortly be detected by the gazelles, is in a position topredict that the gazelles will soon take ight.

In Thinking without Words, I propose looking for the sources ofconditional reasoning in a primitive form of causal reasoning.Whereas conditional reasoning (in the sense codied in the propo-sitional calculus) exploits truth-functional relations between completethoughts, causal reasoning exploits causal conditions holdingbetween states of aairs. Since causal relationships do not holdbetween complete thoughts, an understanding of causality presup-poses no intentional ascent, and hence does not require language.

One might expect on both experimental, observational, andevolutionary grounds that some capacity for causal cognition is verywidespread among animals and available at a very early stage inhuman development.21 The ability to detect certain types of causalregularity and to distinguish genuine causal relations from accidentalconjunctions has obvious survival value. Causal dependence relationsare directly observable, highly salient and pragmatically signicant ina way that no other dependence relations are.

How might causality be understood by non-linguistic animals? Itseems plausible that the core of the understanding of causation at thenon-linguistic level is sensitivity to regularities in the distal environ-ment. A basic sensitivity to environmental regularities must be part ofthe innate endowment of any creature capable of learning about theenvironment, and one might expect any creature to be peculiarlysensitive to regularities between its own actions and ensuing changesin its immediate environment (which is why instrumental condition-

21 See D. Sperber (ed.), Causal Cognition (Oxford: Oxford University Press, 1995).


ing works as well as it does). Of course, as regularity theories ofcausation have been forced to acknowledge, there are many regular-ities that are not causal, and it is in the capacity to distinguishgenuinely causal regularities from accidental regularities that onemight expect differences between different species of non-linguisticcreature and, for that matter, different stages of development withinany given species. The regularities to which non-linguistic creaturesare sensitive (unlike those usually stressed in regularity analyses ofcausation) need not be exceptionless. No creature that only acted onexceptionless regularities would fare well in evading predators andobtaining food.

Proto-causal understanding tracks relationships, which can beeither deterministic or probabilistic, between states of affairs. Thismakes possible a (primitive) grasp of causation at the non-linguisticlevel. It also explains why primitive versions of certain fundamentalinference forms are available at the non-linguistic level. We can termthis proto-conditional reasoning. Instead of treating animals asexploiting full-edged conditionals (i.e., truth-functional compoundsof thoughts) we can think of them as tracking causal relationsbetween states of affairs. I call these proto-conditionals. Conditionalreasoning in animals can be understood in terms of a proto-conditional together with an understanding, which may take the formof a perception or a memory, that the antecedent holds. Theconsequent will straightforwardly be detached.

We see, therefore, that the initial argument for the language-dependence of logical thinking does not rule out the possibility ofnon-linguistic reasoning. We cannot, of course, understand non-linguistic reasoning as involving logical concepts (or any form ofintentional ascent). But we can identify at the non-linguistic levelforms of inference that are analogues of canonical logical inferenceforms and that can be deployed in practical reasoning without anymastery of logical concepts or capacity for higher-order thinking.

7. CONCLUSION

The study of animal cognition offers a rich eld for theorists ofanimal ethics. There are signicant continuities between the cognitivelife of some non-linguistic animals and the cognitive life of humananimals. Some species of animal are genuine thinkers in much thesame way that humans count as genuine thinkers. That is, they


behave in ways that reect their desires and their beliefs about theenvironment. Others are genuine thinkers in a weaker sensethesense characterized by what I have called the minimalist conceptionof non-linguistic thought. Even at the minimalist level we aredealing with forms of behavior that cannot be explained purely interms of conditioning or innate releasing mechanisms. Ethicists whothink that the moral signicance of animals is a function of their levelof cognitive sophistication will need to take account of the subtlegradations between different types of thinking without words. Theywill also need to take on board the limits to non-linguistic thoughtimposed by the argument from intentional ascent. There are seriousconsequences to making the moral signicance of animals dependupon the capacity for higher-order thought (thinking aboutthinkingor metarepresentation). Nonetheless, the types of cognitiveactivity that are ruled out by the argument from intentional ascentare more limited than might immediately appear. As I brought outwith reference to non-linguistic mind-reading and non-linguisticreasoning, non-linguistic animals can get a long way without thinkingabout thinking!

Philosophy-Neuroscience - Psychology programWashington UniversitySt. Louis, MO, 63130, USAE-mail: [email protected]


/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org?) /PDFXTrapped /False

/Description >>> setdistillerparams> setpagedevice

5 bermudez thinking without words

Documents