Locke, Empiricism, and Behaviorism
PM, pp. 18-20, 59-67, 75-79, 85-104
John Locke
1632-1704
PM, pp. 59-67
Introduction
A. 1632-1704
B. Essay Concerning Human Understanding (1690)
1. “underlaborer”
to
2. clearing away “rubbish” – meaningless jargon
C. Locke was to build an entirely different sort of foundation for our knowledge, one based on experience
D. also, of course, an important political philosopher
Book IV, ch. 3 Of the Extent of Human Knowledge (59)
Paragraph 6
A. in earlier paragraphs, he had argued that our knowledge is limited by the sort of ideas we are able to derive from our senses (q.v.)
B. here he argued that even when we have the ideas, our knowledge about them could still be limited (q.v.)
C. for example, we have ideas of circle, square, and equality, and yet may never know how to find a square equal (in area) to a circle
D. Also, we have the ideas of matter and thinking, but we may never know whether matter can think
1. merely by contemplating our own ideas, we cannot tell whether God:
a. gave matter the power to think
b. added to matter a separate substance that has the power to think (59-60)
2. there is no contradiction in God giving matter the power to think – although he will argue in ch. 10 that there is a contradiction in supposing that matter is the first Eternal thinking being (60)
E. how do we even know whether perceptions such as pleasure and pain could be in bodies themselves as well as in an immaterial substance as a result of certain motions in the body?
1. note the mechanical philosophy presupposed throughout
2. we have no idea how motions of physical particles can cause feelings of pleasure and pain, the colors, and sounds (q.v.)
3. so then what reason do we have to conclude that God could not have fixed things so that these sensations are produced in matter, which we think is incapable of sensation, as well as in mind, which we have no idea how motion operates upon? (q.v.)
F. Locke did not wish to be understood here as denying the immateriality of the soul
1. all he was arguing is that we do not have knowledge of this
2. all we have to go on here is probability and faith
G. religion and morality does not depend on the immateriality of the soul but on its immortality, so there’s no necessity to determine whether the soul is immaterial
H. For Locke, the dispute over the immateriality of the soul cannot be resolved (q.v.)
1. on one side, how can the materialists not allow the existence of anything that is not material?
2. or who, on the other side, could argue that God couldn’t have given the powers of thought and perception to solid matter?
3. the hypothesis of an unextended substance and the hypothesis of thinking, extended matter present equal difficulties (61, q.v.)
I. For Locke, there is no question that there is something in us that thinks; it is just that we are ignorant what sort of thing it is
J. But then all substances contain something that baffles us
K. but to return to his argument, knowledge not only is limited by the lack of ideas, but is even limited with respect to the ideas that we do have (61)
Book IV, ch. 10 Knowledge of the Existence of God (62)
Paragraph 10
A. it’s just as impossible to think that bare matter, by itself, could produce a thinking being, as that nothing by itself should produce matter
1. indeed, matter by itself could not produce motion
a. must be created by some other being
b. or eternal
2. but even if motion were eternal, matter and motion together could not produce thought
B. for Locke, none of these things can happen by themselves
1. producing matter from nothing in the first place
2. the beginning of motion if there is matter
3. the beginning of thought if there is matter and motion,
C. if matter were the first eternal thinking being, there would not be one but an infinite number of eternal thinking beings
1. because matter is not one thing but many particles
2. but this would never produce the order we see in the universe (62-63)
Locke’s first letter to Stillingfleet (63)
A. Stillingfleet, the Bishop of Worcester, had complained that in Locke’s philosophy, one cannot prove that there is a spiritual substance in us
B. Locke said he can prove it
1. we know by experience (experiment) that we think
2. thinking cannot exist by itself (self-subsistence) without being in a substance, which in Locke’s sense is spirit (63)
3. in Locke’s sense, any substance that thinks is spirit, regardless whether it is has the additional modification of being solid
C. if by spirit, Stillingfleet means an immaterial substance, Locke grants that he cannot prove there is such a thing, although he thought it highly probable
D. Locke added that if Stillingfleet could prove it, Locke would be happy to learn the proof from him
E. But again, he did not believe that religion and morality, or even the immortality of the soul, depend upon it (63-64)
Locke’s third letter to Stillingfleet (64)
A. the essence of matter is extended, solid substance
B. God can add any properties to matter he chooses without changing its essence
1. He can
a. give it motion
b. form it into plants
c. give it sensation and self-motion and form it into an elephant
2. but if one were to go further and suggest that God could also add thought to matter, some would limit his power and object that He could not do it, since this would change the essence of matter
3. however, Locke argued that if adding thought to matter changes its essence from extended solid substance, then why doesn’t making it into a plant or animal do the same? (q.v.)
C. one could object that we cannot conceive how matter could think
D. Locke granted this, but argued that God is not limited by what human beings can or cannot conceive
1. if adding properties to matter were to change its essence, there would hardly be any matter anywhere in the universe (64-65)
a. to begin with, God has added gravitational attraction to matter, which cannot be explained by the essence of matter (65)
b. and he has made some matter into plants, and some into animals
2. in each of these cases, adding properties to matter does not change its essence, unless there is some “repugnancy” between these properties and the essence of matter
a. the only argument for such a repugnancy is that we cannot conceive how matter could have these qualities unless they were added to it
b. but Locke sees no contradiction in God adding them to matter
E. to adopt the rule that whatever we cannot conceive, cannot possibly be, would be to wreak havoc in religion as well as philosophy (66)
1. one could then argue since we cannot conceive how matter attracts matter across empty space, God could not give it the power to do so
2. or since we cannot conceive how matter could sense or move itself, God could not give these powers to animals
3. if we apply this rule to the soul, we would have to say that since we cannot conceive how even a soul could think, God could not give it that power (The how is important here, q.v.)
F. Locke then has us imagine God creating both a solid extended substance and an immaterial substance and asked whether there is any power or property that God could give to one that he could not give to the other
1. consider the property of self-motion. (67)
a. If one were to argue that God must add it to matter because we cannot conceive how matter could move itself, Locke would argue that we cannot conceive how immaterial substance could move itself, either (q.v.)
b. nor does it do any good to argue that there is something in immaterial substance that we do not understand, because Locke would just reply that there is also something in material substance we do not understand: gravitational attraction
c. so God could give the property of self-motion to either matter or immaterial substance
2. similarly, God could give the property of thought either to matter or to immaterial substance
a. it’s equally beyond our capacity to understand how God could do either
b. but if that were a reason to deny that God could do something, it would be a reason to deny that he can create anything in the first place
Behaviorism
Introduction: Some intervening history
A. Locke is usually taken to be the founder of British Empiricism, followed by Berkeley and Hume
B. Hume (Heil’s introduction, p. 76)
1. took Locke’s empiricism a step further, rejecting the very idea of substance as meaningless
2. all knowledge divided into
a. matters of fact: claims about the world
b. relations among ideas
3. any claim about the world that could not be traced to observation was meaningless
4. hence not only metaphysical notions such as substance but theological notions were called into question as well
C. this rejection of metaphysics also characterizes Comtean positivism
D. which in the 20th century gets transformed into logical positivism
The Logical Analysis of Psychology (1935)
By Carl Hempel
Heil, pp. 85-95
I. Opens with the question whether psychology is a Naturwissenschaft or a Geisteswissenschaft (science of mind and culture) (85)
A. Hempel will represent
the
B. their method is one of the “logical analysis of the language of science,” that is, they treat philosophical questions as questions of meaning
II. the
A. On the accepted view, the sciences of mind and culture differ from the physical sciences with regard to both subject matter and method
1. the objects studied by psychology are mental objects that possess meaning
2. the proper method for studying such objects is one of sympathetic insight or introspection
B. example of the man who speaks:
1. physical sciences provide a causal account of the production of sound
2. sciences of mind concern themselves with the meanings of the words he uses (87)
C. the sciences of mind and culture are said to deal with meaning, while the objects studied by the physical sciences, on the other hand, do not carry any meaning
III. in psychology, this accepted view has been challenged by behaviorism
A. Behaviorism in psychology is characterized by the methodological postulate that scientific psychology concerns itself only with the bodily behavior of people and animals in response to changes in the environment
1. terms like “feelings,” “intentions,” “ideas,” etc. are unscientific and should play no role in a scientific psychology
2. see quotation from J B Watson on p. 75: behaviorism aims at the prediction and control of behavior
a. makes no distinction between human beings and animals
b. conscious thought plays no role
B. as Hempel understands it, behaviorism is attempting to show through its success that psychology deals with purely physical processes and there is no dividing line between psychology and physics (87)
1. However, Hempel thinks that the status of psychology is a philosophical question that cannot be settled by empirical research.
2. Note well that although the logical positivists insisted that psychology is one of the natural sciences, they maintained that philosophy is something separate.
IV. The physicalist conception of the subject matter of psychology (88)
A. in order to determine whether the subject matter of psychology and physics is the same or different, we must first clarify the concept of the subject matter of a science
B. For Hempel, the subject matter of a science consists of propositions (88)
C. Thus, to determine whether there is a difference between the propositions of physics and psychology, we must determine what it is that gives these propositions content or meaning
1. using the example of a proposition about temperature
a. he shows how such a proposition can be translated into a proposition about physical test conditions in which the word “temperature” no longer appears – and without changing its meaning! The statement containing the term “temperature” is just an abbreviation for all these test procedures.
b. this example also shows that two different propositions can have the same meaning
2. he then concludes that “the meaning of a proposition is established by the conditions of its verification”
a. two propositions that are both true or both false under the same conditions have the same meaning (88-89)
b. a proposition for which there are no conditions under which it would be verified is meaningless, a “pseudo-proposition” (89)
D. the problem of the status of psychology then becomes the question concerning the difference between the conditions under which psychological propositions are verified and those under which physical propositions are verified
1. using the example of the proposition “Paul has a toothache”
a. he argues that all the circumstances that verify it are expressed in physical terms (q.v.)
b. the word “pain” is just an abbreviation for passing all those test procedures
c. note that these procedures include “speech behavior of such and such a kind” – for my objection, see discussion at end
d. the proposition containing the word “pain” can be translated without loss of content or meaning into other propositions that are expressed in purely physical terms that do not contain the term “pain” (89)
2. this result contradicts the claim that there is an unbridgeable gap between the natural sciences and psychology
3. further, this result can be generalized to all meaningful psychological propositions (89-90)
a. even those that are said to concern “deeper psychological strata”
b. for example, the claim that Mr. Jones suffers from feelings of intense inferiority can be confirmed or falsified only by his behavior, inc. bodily behavior, physiological behavior, verbal behavior (90)
E. the physicalist conception of psychology: that all meaningful psychological statements can be translated without change of meaning into propositions that use only the concepts of physics and no psychological terms (90)
F. the results of this analysis bear an affinity with behaviorism
V. Some objections:
A. first objection: that these physical test sentences merely describe the outer symptoms from which one infers inner mental states and processes
B. Hempel replies
1. the method of understanding or any other psychological method depends on observable physical data, including the answers people give to questions.
2. There is no other information available to us.
3. consider Mr. Jones again: even if the conclusion that he’s suffering from inferiority is achieved through sympathetic understanding, the only information one has is behavioral – usually the answers he gives to certain questions (91, q.v.)
C. Second objection: people can fake their behavior, such as the criminal on trial who fakes insanity
D. Hempel replies that this fraud can fake only some of the conditions that verify the statement that he is mentally unbalanced.
1. for instance, we could go into the behavior of his nervous system
2. for Hempel, it makes no sense to say that this person could satisfy all the conditions that verify this proposition and yet still be mentally healthy
3. again, psychological terms are just abbreviations for physical responses
E. he turns to an analogy with a watch
1. argues that it would be absurd to say that the movement of its hands is just a symptom of the running of the watch, which cannot be captured by physical means
2. psychological concepts are like “running” and “temperature:” they serve merely to help us abbreviate physical test statements (92)
F. although psychological terms may be useful as abbreviating symbols, there is the danger that their repeated use will lead us to think that they designate some sort of psychological objects
1. this then leads to meaningless metaphysical questions about how these psychological objects relate to physical objects
2. the mind-body problem is thus a pseudo-problem for Hempel, like asking how the running of the watch relates to the movement of its hands (q.v.)
VI. Hempel now compares “physicalism” or “logical behaviorism” with psychological behaviorism and materialism (92)
A. Logical Behaviorism is not denying that feelings, minds, etc. exist
1. rather, it says the question whether they exist is a pseudo-problem, since the legitimate use of these terms is only for abbreviating physicalistic statements
2. nor is he saying that we can know only the physical side of psychological processes and that the mental side is beyond the scope of science and must be left to faith (93)
a. every meaningful question must be capable of a scientific answer
b. in the case of the mind-body problem, that which is considered as an object of belief cannot be stated as a fact. Nothing can be an object of faith that cannot be known. (q.v.)
B. Unlike psychological behaviorism, logical behaviorism is not saying that psychological research should restrict itself to studying responses to stimuli
1. it is not offering a theory in psychology, but a logical theory about the meaning of propositions in science
2. on this theory, it is not imposing a limitation on psychology, but saying that it is logically impossible for psychology to do otherwise than make physicalistic statements
C. in order for logical behaviorism to be acceptable, it is not necessary that we first have detailed knowledge of the human body
VII. Physicalism generalized to other sciences of mind and culture
A. the same considerations can be extended to sociology and in fact to all the sciences of mind and culture (93-94)
1. every meaningful sociological statement is about behavior of groups of individuals (q.v.)
2. that is, every meaningful sociological statement is a physicalist statement (94)
3. Neurath calls this “social behaviorism”
B. Hempel thus arrives at the thesis of the unity of science: all the sciences are really branches of physics (94)
VIII. Sums up: every meaningful statement is translatable without change of meaning into a statement expressed in purely physicalistic terms
A. Objection: that physicalism is denying all the richness of life
B. Hempel replies
1. this is a misunderstanding of physicalism
2. this philosophy is not making any factual claims about the world but is concerned only with analyzing the forms of scientific statements and the relationships between them
3. it is these analyses that lead to the conclusion that certain traditional philosophical problems are only pseudo-problems
a. there may be people committed to such metaphysical problems for various emotional reasons (95)
b. but if philosophy is to be a science, it must break with the past
Discussion:
A. can meaning be captured in physicalist terms alone?
B. let’s back up and re-examine his example:
1. consider the second test sentence on p. 89, in which Paul is asked a question and he answers it. Is this expressed in purely physical terms?
2. Hempel anticipates this objection and tried to answer it in parentheses. But can being asked “What is the matter?” or Paul’s reply truly be equated with the propagation of vibrations in air?
3. similarly, on page 90, he writes of “speech movements aroused by certain physical stimuli (the asking of questions).” Can we define question-asking in purely physical terms?
C. in reply, one may argue that there must be something physical in common to all expressions of the same words in different voices, or we wouldn’t be able to understand each other
D. But Paul could also write his response, or type it on a computer keyboard, or convey it in many other ways that would all have the same meaning
Brains and Behavior (1965)
Hilary Putnam
In Heil, pp. 96-104
I. Introduction
A. opens with a contrast between materialism and dualism
1. materialists: all that exists are material processes
2. dualists: argues that although pains may be correlated with brain processes, they have different properties, such that it makes no sense to ascribe properties of brain events to pains (q.v.)
B. then characterizes logical behaviorism as a third alternative that arose in the 1930s
1. he explains it through an analogy with Russell and Whitehead’s treatment of numbers:
a. these philosophers treated numbers as logical constructions upon sets in order to be able to dispense with embarrassing entities (numbers)
b. in the same way, mental events were regarded as logical constructions on behavior – although in this case, the reduction was never actually carried out (97, q.v.)
2. thus, for the logical behaviorist, talk about mental events is translatable into talk about behavior
C. Logical Behaviorism combines the advantages of dualism and materialism
1. it can adopt the dualist’s argument that brain events are not what we mean when we are talking about pains
2. but it can also adopt the materialist’s tough-minded rejection of unnecessary entities, such as a Cartesian mental substance (97)
D. But Putnam wants to know whether logical behaviorism is just one more alternative to be added to materialism and dualism, or whether we can somehow decide that it is right or wrong
1. its extreme thesis that talk about mental events is translatable into talk about physical events has proved to be false
2. is there anything that can be “saved” from logical behaviorism?
a. in the last 30 years, a weaker version of logical behaviorism has emerged, which is limited to the following claims:
1.) that there are entailment relations between mind-statements and behavior-statements (q.v.)
2.) although these entailments do not provide the basis for an actual translation, this is because mind-talk is much more ambiguous than behavior-talk (98, q.v.)
b. Putnam will argue that even this weaker form of behaviorism is incorrect
-- in footnote 3, he mentions a fourth alternative, functionalism, which we will get to later in this course
II. Logical Behaviorism and the meaning of “pain”
A. logical behaviorists argue that we do not learn the meanings of words like “pain” in the same way that we learn the meanings of words like “red,” that is, by pointing to it (98-99, q.v.)
1. all one can point to is behavior (99)
2. one cannot point to someone else’s pain, and ask you to compare your feeling to theirs, and say that if it is the same thing, you are feeling a pain (q.v.)
3. thus, it is not true that the meaning (intension) of pain is a certain quality that I feel (q.v.)
B. but this does not refute dualism, which says only that you feel the pain, not that the meaning of the word “pain” depends on this feeling
C. for Putnam, “pain” is a cluster concept (99)
1. that is, there are lots of criteria for it, all of which are synthetic
2. the logical behaviorist might agree that it stands for a cluster of observable behavior
3. but Putnam has something else in mind
III. Putnam’s polio argument
A. when doctors discovered that polio was caused by a virus, they could say that some diseases they previously thought were cases of polio were not, since the virus was not present
1. the same would happen if another disease, such as MS, were discovered to be caused by a virus. This would not be falsified if:
a. in a few cases, some patients had similar symptoms without the virus
b. or if this virus also caused some additional symptoms not previously recognized
2. thus, the meaning of “polio” or “MS” could not be identified with a cluster of symptoms (99-100)
3. rather, one would define such diseases as “those that are normally responsible for such symptoms” (100)
B. some philosophers might want to argue that polio used to mean a cluster of symptoms, but that now the meaning has changed
C. Putnam does not want to go this route, as it violates common sense and linguistic practice
1. on the change-of-meaning view
a. if a doctor in the past had said “I think polio is caused by a virus,” he would have been wrong, because as the word was used in the past, polio was not always caused by a virus (100)
b. only what we mean now by polio is caused by a virus (100)
2. Putnam’s way of defining diseases leaves it open whether there may be more than one cause, whether someone could have the disease without manifesting the symptoms, etc.
D. one could accept Putnam’s way of defining diseases but still think that there may be analytic entailments between diseases and symptoms that allow translation of disease talk into symptom talk – but Putnam will reject this.
E. For example, one could say “Normally people who have MS have the following symptoms” (q.v.)
1. However, it does not follow that one can translate statements about diseases into statements about symptoms – the “normally” gives it away
2. not because disease talk is too ambiguous
3. but because diseases are not logical constructions on symptoms; causes are not logical constructions on their effects
IV. Mental states, again
A. here he draws the analogy between diseases and mental states:
1. the dualist and materialist could agree that that what we mean by a word like “pain” is not a cluster of responses, but something that normally caused those responses
2. but even if that were what we meant by the word “pain,” that is, even if the following were necessary truths:
Normally, when one says “ouch,” one has a pain.
Normally, when one has a pain, one says “ouch.”
that would shed no light on what pain is (101)
a. nor would it follow that pain talk can be translated into behavior talk
b. because pains are not logical constructs on behavior
B. the dualist might want to go even further and argue that these two statements are not necessary truths (101)
C. Putnam agrees with the dualist: there can be worlds in which there are pains without pain behavior
1. we should not confuse conditions for saying someone is in pain – pain behavior – with conditions for the existence of pain
2. doesn’t help if in addition to “John says he’s in pain” we add “John speaks sincerely,” because unless we want to claim that sincerity is a construction out of behavior! (102, q.v.)
3. pains are not identical with clusters of responses, even if – in the normal case – they are the causes of clusters of responses. To say this much does not rule out the possibility that the causes may be present without their usual effects
4. first he has us imagine communities of “super-spartans” or “super-stoics,” who are able to completely suppress all outward expressions of pain
a. one might try to argue that they would still avoid painful things, but the problem is that people also avoid things that are disgusting, frightening, boring, etc. (102-3)
b. even if they never showed that they were in pain, one would not want to deny that they can feel pain (103)
5. then he has us imagine another world where instead of the relationship between pain and its effects being altered, the relation between pain and its causes is changed
a. for example, people who feel pain caused by magnetic fields
b. if these people were also super-spartans, normal relations to both causes and effects would be different, but pains are clearly present
D. he then considers verbal reports of pain
1. here’s a problem: if to be in pain means to be disposed to utter a certain kind of verbal report, how does one tell that a verbal report is of a certain kind? (103) (a problem Hempel had overlooked)
2. one might argue that verbal pain reports are the kinds of verbal pain reports that have been linked to unconditioned pain responses
3. but Putnam has argued that there is no reason to believe that there are unconditioned pain responses present in all species capable of feeling pain
4. and then what about an imaginary world in which super-spartans would not even say they were in pain? (104)
V. Conclusion
A. unless there is some hidden inconsistency in Putnam’s imaginary worlds, logical behaviorism is mistaken
B. both claims of even the weaker version are false:
1. cannot translate pain talk into behavior talk
2. statements about pains taken by themselves do not entail any statements about behavior