Instinct: A study in social psychology
Chapter 8: The Classification of Instincts
Luther Lee Bernard
Thorndike lists four methods of classifying instincts which are now in use. These are:
"By the functions which the tendencies perform.
By the responses which are their end-terms.
By the situations which are their first-terms.
By their origin or affinities in development." 
He maintains that the classifications by function are the commonest, and they have the advantage that it is easy to describe the instinct in an easily recognizable manner in this way. Most people think in this "functional" way, or in terms of results. The difficulty of this method is that it describes our attitude toward the activity process or our understanding of its value in society, rather than the process itself. Such a definition does not enable us to distinguish one instinct or set of instincts from another, so far as origin or structure is concerned, and hence does not enable us to control instincts in the character-forming process or in the process of social organization. The functional classification does not function very well in the problem of social control. The classifications of the psychoanalysts referred to in the preceding chapter, dividing the instincts into reproductive, nutritional and social or herd instincts are of this type. They describe the function performed or the social and individual value of the act rather than the process itself. This makes possible all unlimited confusion by which acts of the most diverse character are classed under the same instinct term merely because
(149) they serve the same function or have the same value in the life process or in the struggle for existence and survival.
Classification by responses or end-terms of the act is also a favorite one. Such instincts and reflexes as crying, smiling, coughing, sneezing, yawning, come under this category. The advantage of such a classification is obvious. It is easy to visualize such acts. They are usually clear cut and distinct; there is little chance for confusion so far as the overt aspect of the instinct is concerned. And there must be a fairly close connection between the neural organization or inner process and the overt act which gives the name to the instinct. But there is a real danger of confusion here. In the first place, a great variety of stimuli may produce the same overt result. Thorndike quotes Borgquist to the effect that there are at least forty-seven groups of causes for crying. Evidently, then, there must be at least an equal number of distinct psychophysical processes which we may call by the name of the same instinct, if crying is truly instinctive. Such a method of defining instinct ignores the inner or neural complement or aspect of the inherited process; and this is unquestionably the more important aspect, since it is the part— the connecting part — which is inherited. The overt part of the act is primarily determined by the neural organization. It is also in part determined by the gross external organization of the body and by the impact or pressures of the environment upon the responding mechanisms and the organism as a whole. But these external factors also are not included or taken account of in this method of classification and definition.
The same criticism in general may be brought with respect to the third method of classification, that by the situations which are the first terms. This method, like the first, is an external method. It does not define the psycho-phsyical mechanism involved. It takes account only of the situations which
(150) produce the result. These may vary greatly and yet, especially where habit modifications enter, produce essentially the same responses. Nor is it possible to determine accurately from these external situations what internal processes or mechanisms will be involved. There is no direct or definite correspondence between external situation and internal neural mechanism. This is because the external situation as a whole does not operate as a unit upon the neural processes. Different parts of the external situation, giving rise to the activity complex, may call into operation conflicting neural processes which prevent unitary simple responses altogether, or at least vary these responses so greatly that it is difficult to classify action accurately on the basis of the stimulus correlates. Thorndike himself points out that no one has made a classification of the instincts on this basis, although he thinks G. Stanley Hall had something like this in mind in some of his studies. Such a classification is probably physically impossible, for the reason that every situation has infinite chances for variation, according as the relationship between organism and situation changes. Every change of temperature, light, distance, humidity and possibly scores of other factors in the environment would modify the situation until its measurement would become impossible because of its infinite variety. Even if the situation could be measured or kept constant it would fail to meet the requirements fully, because the situation is external to the neural process or organization which is central to the instinct. An instinct is an activity process based on a neural mechanism. The mere mention of instincts of this type— such as temperature, light, sound, food, humidity instincts— discloses the indefiniteness of the category and the impossibility of making a definition
(151) in mechanical or quantitative terms. The only characterization possible, without entering into the most minute details, thereby differentiating an infinite number of instincts and transferring the description from the purely external to include also the internal or neural aspects, must remain merely qualitative. And science cannot be founded on qualitative distinctions alone.
The fourth method of classification of instincts— by "affinities in the development of the race," or by origin and descent— is the one by which Thorndike sets most store. He says, "Such a classification would be a `scientific' or `natural' one because it would arrange man's instincts and capacities for purposes of study in an order corresponding to their genesis in the real world, and so incite students to note the elements in which heredity carries along man's equipment and the possibilities for its future evolution." We must agree with Thorndike that such a classification is conceivable. It has been attempted by Colvin in the list of instincts presented in the previous chapter, with what degree of success no two people are likely to agree. Such a classification would also have its value, for such a purpose as Thorndike mentions. The other classifications also have their values. Each classification enables us to make a grouping, and consequently a selection, of action patterns according to some category which we may desire to manipulate or control. In one case it is the use to which we may put the action patterns, that is, the socially desirable ends which we may wish to achieve. In another case the attention is thrown upon the responses which may be decidedly significant in the social control process. In a third case it is the type of situation, or the immediate environment, which may condition the forms of activity; and no one wilt deny the importance of environmental analysis any more than he will question the significance of ends or social values.
( 152) But no one of these points of emphasis tells us anything definite and final about the nature of the instinct itself. It does not enable us to recognize, to distinguish and to measure, the instinct. It leaves us where we may still confuse one instinct with another because we have classified the instinct in each case in terms of its service or its external connections or its origins and relationships. Such classifications are valuable, but they are only partial or supplementary. They are not final and they cannot stand alone.
We must add to these categories listed by Thorndike a fifth type of classification, one based on inner structure and organization. This is of course an extremely difficult classification to make, because the process of localization of instincts has only begun. The ideal method in making such a classification would consist of at least two parts: (r) the tracing of the neural pathway of each instinctive activity process, and (a) the classification of these various neuro-psychical inherited processes according to the organic connections which they serve,— whether purely internal, as in the case of digestive and assimilative processes, circulation, etc.; or of a mixed inner and outer adjustment character, as in respiration, perspiration, etc.; or, finally where the stimulus and response are both primarily external, as in the case of withdrawing from a heated object or from a cutting or tearing surface or the stimulus of light, heat, etc. Only by means of such a classification will we be able adequately to distinguish one instinct from another and thus to classify the instincts according to their individual and collective identity, as well as from the standpoint of their use, origins and external associations, Already we know enough about the nerves system to locate and classify many of the instinct mechanisms according to the second subdivision of this fifth category. We
( 153) can locate those connected with movements of the eye, tongue, heart-beat, respiration and a host of others of similar functional types. But we have not yet learned to trace the actual neural processes or fibers which connect sense organs with muscular or glandular mechanisms. This will come in time, although slowly, largely as the result of experimentation upon lower animals. The knowledge basic to this fifth type of classification will be invaluable, not only because it will enable us to distinguish between inherited and acquired processes, but it will also make it possible to judge from the positions in the nervous system of the instinctive processes the relative degree of educability under the control of the cerebral cortex. It may conceivably, some time in the future, enable the scientist to exercise some sort of taxinomic control over the educative process through the agency of surgery or vascular, muscular or glandular, control processes, or even by the use of drugs and other chemical and physical applications and appliances. But such results certainly are not now feasible, except possibly in the crudest and most general sort of way. Such a classification, based on the inner organization or structure of the instincts, is not possible without taking into account the end organs and the muscular and glandular response mechanisms. But an efficient classification cannot be based on these external factors alone. For the best working results in applied behavior, whether in education, psychotherapy, or any other field, including that of general social practice, all five classifications must be used together to supplement each other. But the classification on the basis of inner structure and organization must ever be basic and central to the other four. The fact that it arose last is not significant of its relative importance, but only of the difficulty encountered in the process of definition and segregation or differentiation. Definitions dependent upon internal analysis are most difficult to make.
With this preliminary analysis of the bases or methods of classification of instincts to serve as a point of departure for criticism and comparison, we may now consider a number of typical classifications of instincts set forth by writers whose books are now current. In addition to the classification on the basis of genesis, cited above, Colvin offers two other classifications: (2) Egoistic, altruistic and mixed instincts; (3) Personal and impersonal instincts. The former of these two additional classifications is a very common one, to be met in some form in a great variety of books. It, also, belongs to the first of Thorndike's classification types. The second of the two classifications is not so commonly seen. It is difficult to classify it according to Thorndike's categories, but it probably belongs to the third. Another classification of the first of Thorndike's types is offered by Colvin and Bagley, as follows:I. Adaptive instincts
- Inquisitiveness or curiosity
II. Individualistic instincts
III. Sex and parental instincts
- Protection of the young
IV. Social instincts
V. Religious and Aesthetic
- Religious— self-subjugation
- Aesthetic— rhythm
It is not difficult to see how the terms in the foregoing classifications overlap, because the distinctions are not made in terms of structure but on the basis of the social values or functions which each so-called instinct serves. The " instincts" in this classification are not instincts at all, so far as any structural or organization definition is concerned. They are merely activity and abstract value complexes, in which no method or power of distinction between acquired and inherited elements inheres within the classification itself. The emphasis is upon the social value of the action, not upon its derivation and distinctness. The same sort of a mechanism, or identically the same mechanism, might easily be included under several or even all of these so-called instincts defined and classified according to function or value. Such an eventuality would in itself condemn the above grouping as violating the first condition of scientific classification, that of separateness of the terms. The same may be said in general ,of the following classification by Kirkpatrick, which Thorndike quotes
(156) and characterizes as being one of the best of the first type.
I. Individualistic or self-preservative instincts
II. Parental instincts
- Sex and courtship instincts
- Fighting for mates
- Nest building
III. Group or social instincts
- To arrange themselves in groups
- To coöperate for the common good in attack and defense
- Seeking companionship
- Desiring the approval of the group which one joins
IV. Adaptive instincts
- Tendency to spontaneous movement
- Tendency for nervous energy to take the same course that has just been taken
- Tendency to imitation
- Tendency to play
- Tendency to curiosity
- V. Regulative instincts
- The moral tendency to conform to law
- The religious tendency to regard a higher power
VI. Resultant and miscellaneous instincts and feelings
- The tendency to collect objects of various kinds and to enjoy their ownership
- The tendency to construct or destroy and the pleasure of being a power or a cause
- The tendency to express mental states to others of the species and to take pleasure in such expression
- The tendency to adornment, and the making of beautiful things, and the aesthetic pleasure of contemplating such objects.
Besides including a number of activity or adjustment values, such as imitation, play and curiosity, which are not now regarded by the conventional psychologists as instinctive, we find in this classification obvious signs of overlapping, for the same reasons that operated in the preceding classification. Fighting for mates, for example, would probably make use of the same neuro-muscular mechanisms as fighting for self-preservation, yet they are classified as separate instincts. Again, self-exhibition might easily use the same neuromuscular or neuro-glandular processes as ambition, and ambition as pride. Seeking companionship might employ the same organic processes as courtship, at least in part. If such is the case, obviously this classification is not very significant. It needs to be corrected by the use of a classification on the basis of inner mechanism and organization. This classification illustrates again the impossibility of distinguishing instincts on the basis of their use. Instincts cannot be defined in such terms; they can only be characterized and evaluated socially in this way. Also, all of the instances of instincts listed under classes V and VI are obviously largely learned its character. They ire habit complexes rather than instincts; but the functional basis of classification does not make it possible to detect such a confusion.
Two other classifications of instincts, carefully made, but of the same general character, may be cited. They serve not only to illustrate the confusion set forth above; but they also suffice to show how differently two or more distinct persons may see the meaning or significance of the same facts when they are employing speculative subjective, instead of quantitative objective, criteria of measurement. The former of these two classifications is by Howard C. Warren  and is as follows:
I. Nutritive instincts
- Metabolic expressions
- Wandering (hunting)
- Acquiring (hoarding)
II. Reproductive instincts
- Mating (sexual attraction, courtship)
- Filial (of infancy)
III. Defensive instincts
- Modesty (shyness)
- Clothing (covering)
- Constructing (home-making)
IV. Aggressive instincts
V. Social instincts
- Family (parental, filial)
- Tribal (gregarious)
To these so-called instincts or classes of instincts may be added certain reputed instinctive tendencies of man.
- Dextrality (right-handedness)
- Aesthetic expression
The best that can be said of the so-called instincts under the general headings of this classification is that they also are composite categories, including many different specific instincts and reflexes as well as habits.
Woodworth offers a classification which cuts across Warren's and only partly coincides with it. It runs as follows: 
I. Responses to organic needs
- Drinking instinct
- Hunger instincts
- Seeking the breast
- Rejecting the breast
- Spitting out bad-tasting food
- Breathing and air-getting instincts
- Responses to heat and cold
- Flushing of skin
- Paling of skin
- Shrinking from heat
- Shrinking from cold
- Shrinking from injury
- Flexion reflex
- Rubbing the skin
- Clearing the throat
- Changing from an uncomfortable position
- Warding of a blow
- Huddling into the smallest possible space
- Getting under cover
- Clinging to another person
II. Responses to other persons
- Herd instinct or gregarious instinct
- Uneasiness when alone
- Seeking company
- Remaining in company
- Mating instinct
- Decoration of the person
- Demonstrating one's prowess
- Admiring attention to one of opposite sex
- Parental or mothering instinct
- Feeding the young
- Warming the young
- Defending the young
III. Play instincts
- Playful activity
- Kicking by baby
- Throwing arms about
- Holding up the head
- Sitting up
- Kicking with an alternate motion of the two legs
- Climbing (?)
- Cheerful babbling
- Manipulation (by baby)
- Turning things about
- Pulling things
- Pushing things
- Dropping things
- Throwing things
- Pounding things
- Exploration or curiosity
- Examination of objects by the hand
- Examination of objects by the mouth
- Listening to a suden noise
- Following a moving light with the eyes
- Fixing the eyes upon a bright object
- Exploring object visually by looking successively at different parts of it
- Sniffing an odor
- Asking questions
- Attention (?)
- Reasoning (?)
- Tendencies running counter to exploration and manipulation
- Overcoming obstruction
- Resisting domination by other persons
- Seeking for power over things
- Seeking to dominate other people
- Giving up in the face of obstacles
- Docility of the child
- Yielding to the domination of other persons
This classification has been constructed from the text and is believed to represent Woodworth's views. The critical reader will observe that there are many activities included
(163) which appear to be acquired and many social activity values rather than unitary psycho-physical action patterns. For example, submission and self-assertion, manipulation, fighting, and the like, are not unit action processes but general evaluative or descriptive terms covering a great many inherited and acquired action and thought processes which are adjudged by society generally to have the meaning values indicated by the terms attached to them. This fact is evidenced by the circumstance that most of these terms carry subclassifications. This is one of the most complete classifications of instincts in the literature of psychology and behavior. But it is not as full as a classification based on Thorndike's Original Nature of Man would be. I have not felt sufficiently sure of all of Thorndike's interpretations and classifications to venture upon such an interpretative classification of his terminology at this time.
Woodworth's classification is obviously of a mixed sort, judged by the categories of classification set forth in the earlier part of this chapter, but the dominant criteria of classification used are the functional or end and the response methods. He pays little or no attention to the matter of internal structure and organization, with the result that obviously the same structural processes, including in some cases the same stimuli, and in other cases the same responses, are classified under different instinct categories. This would certainly be true of self-assertion and fighting, of manipulation and curiosity, of exploration and locomotion, of gregariousness and mating, and of other cases also; although the author in question lists these as separate instincts. The fact is that the author, as is true of most writers on the instincts, rarely if ever gets down to the unit organization of activity as a basis for the
( 164) definition and classification of the instincts. He remains on the basis of complex social and personal values instead. Yet instincts are biological facts or organizations; they are structural and concrete. They cannot be mere words, abstract valuation processes, used as control symbols for the synthesis and distribution of activities of diverse sorts under one control category. Such abstract processes are acquired; are the product of civilization, not the result of hereditary selection. In this respect Woodworth errs in the same way in which the other psychologists quoted err. They do lip service to biology, in insisting upon an instinct theory of the interpretation of conduct or behavior; but they are short of biological insight, in that they are unable to distinguish between the concrete inherited activity processes on the one hand and the abstract acquired valuations on the other. They are far from getting down to the units of activity or even the native syntheses or organizations of inherited unit activities.
Perhaps it may be excusable to add one more general classification of abstract value terms, under the guise of instincts, to the list. Hocking has worked out his classification much more systematically and schematically than have the preceding writers, as will be observed from the diagram which follows.
|POSITIVE (Expansive)||NEGATIVE (Contractive)|
|Instinct to Physical Activity (?)
Rubbing Eyes, etc.
|Instinct to Inactivity(?)|
Preparation for Repose, Sleep, Death
|Food Getting||Food Aversion|
||Spitting Out||Averting Head|
||Protective (extension of parental?) Aversion to Blood|
||Aversion to Novelty|
||Rejection of Contact:||Shame|
||Aversion to Children (?)|
|Attachment to Parent|
NOTE. Instincts of second order (Curiosity, Play, Pugnacity, Fear) written across page. Units of behavior in Italics. Indentation indicates degree of generality, not genetic priority.
This classification does not differ essentially from the others as to types of criteria used. It is mixed in character. It shows the same overlapping. It fails to get down to the actual concrete basic inherited processes; it is external instead of internal. It goes even further in the wrong direction than the others, because it is so constructed as to place most emphasis upon the social organization and evaluative aspects. As pointed out in the preceding chapter, the general or "central" instincts of curiosity, play, pugnacity and fear made use of by Hocking, are not instincts at all, but are synthetic acquired valuations of adjustment processes and needs which have grown up as a result of the experience or education of the person exercising them. They function largely under cerebral cortical control and are the product of the workings of the conscious or partly conscious mind, which is obviously ,not inherited but is the product of experience and education. The attempt to distinguish positive and negative, aggressive and defensive, instincts also belongs to the external and nonspecific treatment of instinct. It is evaluative and synthetic instead of analytical and definitive. It does not enable one to distinguish instinct from habit, or to locate and manipulate the actual activity processes in the service of individual or social control.
Watson is very modest in his claims regarding his classification of the instincts. He appreciates more fully than most of the psychologists the limitations to biological heredity. Yet, his classification is, for concreteness and its purely hereditary content, perhaps the most dependable of all those cited here. He says, "It is impossible, with the survey of animal activity which has been made, to give anything like a complete inventory of ' the various classes of instincts, or to give a classification which will be accepted by any large number of investigators. Nevertheless, we should expect to
(167) find, and indeed we do find, certain characteristic instincts in every species of vertebrates. If the animal has survived at all, it is necessary for it to have certain instincts which relate to food, shelter, reproduction, defense, and attack, etc. The table below may be found helpful in holding together the mass of material which one finds when examining the work of students of behavior, and naturalists."  Watson's classification follows:
|I. Structural characteristics, action systems, etc.||Bodily characteristics by means of. which species are determined; methods of locomotion, such as creeping, crawling, walking, etc.|
|II. Obtaining food.||Taking food from beak of parent; sucking, pecking, scratching, diving; selection of food (when not determined by habit, e. g., herbivorous, carnivorous, etc.); the use of salt instead of fresh water; washing the food; methods of drinking, etc.|
|III. Shelter.||Stretching the wing to escape the sun; sunning, huddling, ruffling the feathers when cold, etc.; burrowing, taking up of abodes in hollow logs and trees; boring into timbers; hibernation, etc. Overlaps to some extent the group on sex and food.|
|IV. Rest, sleep, play, etc.||Night and day periods of activity, purely instinctive, since no structural peculiarities account for the differences; length of periods at nest; brooding or caring for the young; habits of sleep, i. e., bodily attitudes chosen; play, hibernation, etc.|
|V. Sex.||Mating; copulation; nest site; material of nest; methods of building nest; number of eggs laid or young cast; length of mating period or period in which partnership endures; care of the young etc.|
|VI. Defense and attack.||Methods of attack, as lying on back; springing upward (to head and throat); spurring, goring with the antlers or horns, etc. Here belong also the hunting, stalking, seizing, and rending of live prey; shamming death; inflation of body; ejecting secretions, etc.|
|VII. Special forms of instinct.||Migration (possibly homing in general) as exhibited by birds, mammals, fish, reptiles. Possibly overlaps sex and food, but may be wholly independent of either. Mimicry.|
|VIII. Vocalization.||Calls, cries, sounds uttered in receiving food; during sex activity, etc. Shown in almost all vertebrate forms, but especially in the monkeys of the higher types and in birds.|
|IX. Unclassified and non-adaptive but complex and complete acts.||Strutting, dancing, inflation of cheek pouches; secretions of musk; elaborate nodding (as in the noddy tern).|
|X. Unclassified and non-adaptive, in this case random and abortive, sometimes appearing in one and sometimes in another combination, and sometimes in isolation.||(Discussed under reflexes, p. 110.)|
|XI. Individual peculiarities in response.||Reference is made here to very definite phenomena but ones hard to describe, such as persistence of the reverse in an attack on a problem (as shown in learning); oldness, individual tricks of hands, beak; ambidexterity or preponderating use of either right or left hand by any animal below man; probably hereditary but not known to be.|
Of course it is not the intention to argue that such evaluational and abstract classifications of activities, and even of instincts, as those quoted in this chapter are not valuable. They are of the greatest significance in social control and in-
(170) -dividual training. But they are useful primarily as evaluations of activity types in a social situation and are not to be considered as accurate classifications and descriptions of native or instinctive processes. They constitute a large and valuable part of the subject-matter of the sciences of sociology and of education. But they belong to these subjects rather than to psychology, as this science has been developed historically. In its classification of instincts psychology should give us some criterion (1) for distinguishing one instinct from another, (a) for distinguishing instinct from habit, and (3) for acquainting us with the structure and organization of the instinctive processes themselves. Otherwise we cannot properly get hold of the instincts in our thinking and utilize them in the evaluative and constructive processes which are an essential part and function of sociology and education.
I shall not attempt at this point to arrange a classification of the instincts on a truly scientific basis. I do not believe that sufficient investigation has been made of the original or inherited activity processes to make this possible, except in the most temporary and fragmentary way. Watson is of the opinion that it is not possible to get a logical classification of the instincts. He says, "This has been tried many times, but it is certainly impossible to get a classification at present based upon any other grounds than that of general convenience."  The special investigations of particular instincts are slowly aiding us in the collection of the necessary data. And such works as those by Thorndike, Watson and Woodworth here cited throw considerable light upon the question. Shand's Foundations of Character also adds something to our knowledge of the subject through his analysis of the complex instinct categories into their component elements. To formulate a classification sufficiently accurate in its details would require more time than has yet been given to this
( 171) aspect of the subject and certain types of data which are as yet insufficiently available. In addition to the study of all the special investigations of the subject it would be necessary to go through the literature of experimental and descriptive zoology and make a complete list of all of the native reflexes and other presumably inherited activity processes in the whole field of animal life for comparative purposes. Finally, it would be necessary to study human reaction processes and types in the light of this comparative material and in the light of the conflicting social controls, arising from the distinctively human environment, operating both ontogenetically and phylogenetically, which tend to modify or repress the inherited activity processes of the lower animals. What has been attempted here is much less ambitions. It is in the nature of a criticism of the methods and presuppositions underlying current classifications of the instincts, looking forward to the formulation of a scientific method of procedure. It will be of value, however, to present in this connection, in the following chapter, the results of a statistical study of the current usage of the term instinct.