Popular Science Monthly/Volume 44/March 1894/The Origin of Right-Handedness




THE question "Why are we right or left handed?" has exercised the speculative ingenuity of many men. It has come to the front anew in recent years in view of the advances made in the general physiology of the nervous system; and certainly we are now in a better position to set the problem intelligently and to hope for its solution. Hitherto the actual conditions of the rise of "dextrality"—as the general fact of uneven-handedness may be called—in young children have not been closely observed. It was to gain light, therefore, upon the facts themselves that the experiments described in the following pages were carried out.

My child H—— was placed in a comfortable sitting posture, the arms left bare and free in their movement, and allowed to reach for objects placed before her in positions exactly determined and recorded by a simple arrangement of sliding rods. The experiments took place at the same hour daily for a period extending from her fourth to her tenth month. These experiments were planned with very great care and with especial view to the testing of several hypotheses which, although superficial to those who have studied physiology, yet constantly recur in publications on this subject.[1] Among these theories certain may be mentioned which my experimental arrangements were aimed to test. It has frequently been held that a child's right-handedness arises from the nurse's or mother's constant method of carrying it; the child's hand which is left free being more exercised, and so becoming stronger. This theory is ambiguous as regards both mother and child. The mother, if right-handed, would carry the child on the left arm, in order to work with the right arm. This I find an invariable tendency with myself and with nurses and mothers whom I have observed. But this would leave the child's left arm free, and a right-handed mother would be found with a left-handed child. Again, if the mother or nurse is left-handed, the child would tend to be right-handed. Or if, as is the case in civilized countries, nurses replace the mothers, it would be necessary that most of the nurses be left-handed in order to make most of the children right-handed. Neither of these positions is true. Further, the child, as a matter of fact, holds on with both hands, however it is itself held. Another theory maintains that the development of right-handedness is due to differences in weight of the two lateral halves of the body; this tends to bring more strain on one side than the other, and so to give more exercise to that side. This evidently assumes that children are not right or left handed before they learn to stand. This my results given below show to be false. Again, we are told that infants get right-handed by being placed on one side too much for sleep: this can be shown to have little force also, when the precaution is taken to place the child alternately on its right and left sides for its sleeping periods.

In the case of the child H—— certain precautions were carefully enforced. She was never carried about in arms at all—never walked with when crying or sleepless (a ruinous and needless habit to cultivate in an infant); she was frequently turned over in her sleep; she was not allowed to balance herself on her feet until a later period than that covered by the experiments. Thus the conditions of the rise of the new right-handed era were made as simple and uniform as possible.

The experiments included, besides reaching for colors, a great many of reaching for other objects, at longer and shorter distances, and in unsymmetrical directions. The following table (I) gives some details of the results of the experiments in which simple objects were used, extending over a period of four months (fifth to ninth in her life). The number of experiments at each sitting varied from ten to forty; the position of the child being reversed after half of each series.

Table I.

Date. No. of
No. of ex-
1890. February 10th to March 15th. 30 744 173 166 405
March 14tb to April 14th 25 623 134 141 348
April 14th to May 14th 25 546 213 130 203
May 14th to June 19th 16 274 57 131 86
Total 96 2,187 577 568 1,042

It is evident from Table I that no trace of preference for either hand is discernible during this period; indeed, the neutrality is as complete as if it had been arranged beforehand, or had followed the throwing of dice.

I then conceived the idea that possibly a severer distance test might affect the result and show a marked preferential response by one hand over the other. I accordingly continued to use a neutral stimulus, but placed it from twelve to fifteen inches away from the child. This resulted in very hard straining on her part, with all the signs of physical effort (explosive breathing, sounds resulting from the setting of the larynx, rush of blood to the head, seen in flushing of the face, etc.). Table II gives the results; the number in each series was very small—i. e., one to twelve (in one instance only):

Table II.

Date. No. of
No. of
1890. May 26th to June 10th 32 80 74 5 1

The same cases, distributed according to distance, give us Table III.

Table III.

12 inches. 13 inches. 14 inches. 15 inches.
Right hand 29 10 33 2
Left hand 5 . . . . . .
Both hands 1 . . . . . .

A comparison of II and III with I shows a remarkable difference—i. e., during the month ending June 15th, the child showed no preference for either hand in reaching straight before her within easy reaching distance (ten inches); but she was right-handed to a marked degree during the same period as regards movements which required effort or strain, such as grasping for objects twelve to fifteen inches distant. The left hand was used in only five cases as against seventy-four cases of the use of the right hand; and further, all these five cases were twelve-inch distances, the left hand being used absolutely not at all in the forty-five cases at longer distances.

In order to test this point further, I varied the point of exposure of the stimulus to the right or left, aiming thus to attract the hand on one side or the other, and thus to determine whether the growth of such a preference was limited to experiences of convenience in reaching to adjacent local objects, etc. The result appears in Table IV:

Table IV.

12 inches. 13 inches. 14 inches. 15 inches. Hand used.
Deviations from median line— Right. Left.
2 to 6 inches to left 10 cases 15 cases 4 cases . . . . 35 . . . .
2 to 6 inches to right 2 " 3 " 1 " . . . .
Same conditions with color stimulus . . . . . . . . . . . . . . . . 15 2

This table shows that deviation to the left in front of the body only called out the right hand to greater exertion, while the left hand fell into still greater disuse. This seems to show that dextrality is not derived from the experience of the individual in using either hand predominantly for reaching within the readiest range of that hand.

Proceeding upon the clew thus obtained—i. e., that a stronger effort brought a preferential hand reaction—a clew which seems to suggest that the hand preference is a function of the relative strength of the influence of the eye stimulus, I introduced hand observations into a series of experiments on the same child's perception of the different colors which I was making at that time, thinking that the color stimulus which represented the strongest inducement to the child to reach, might have the same effect in determining the use of the right hand as the increased distance in the experiments already described. This inference is proved to be correct by the results given in Table V:

Table V.

Color stimulus,
10 to 15 inches
Hand Right. Left. Both. May 23d to June 19th.
Number of cases 86 2 ..

It should be added that in all cases in which both hands are said to have been used, each hand was called out with evident dependence of the other, both about the same time, and both carried energetically to the goal. In many other cases in which either right or left hand is given in the tables, the other hand also moves, but in a subordinate and aimless way. There was a very marked difference between the use of both hands in some cases, and of one hand followed by, or accompanied by, the other in other cases. It was very rare that the second hand did not thus follow or accompany the first; and this was extremely marked in the violent reaching for which the right hand was mainly used. This hand was almost invariably accompanied by an objectless and fruitless symmetrical movement of the other.

The results of the entire series of experiments on the use of the hands may be stated as follows, mainly in the words in which I reported them summarily some time ago:[2]

1. I found no trace of preference for either hand as long as there were no violent muscular exertions made (based on 2,187 systematic experiments in cases of free movement of hands near the body: i. e., right hand, 577 cases; left hand, 568 cases; a difference of nine cases; both hands, 1,043 cases; the difference of nine cases being too slight to have meaning).

2. Under the same conditions the tendency to use both hands together was about double the tendency to use either (seen from the number of cases of the use of both hands in the statistics given above), the period covered being from the child's sixth to her tenth month inclusive.

3. A distinct preference for the right hand in violent efforts in reaching became noticeable in the seventh and eighth months. Experiments during the eighth month on this cue gave, in 80 cases, right hand, 74 cases; left hand, 5 cases; both hands, 1 case. This was true in two very distinct classes of cases: first, reaching for neutral objects (newspaper, etc.) at more than the reaching distance; and, second, reaching for bright colors at any distance. Under the stimulus of bright colors, from 86 cases, 84 were right-hand cases and 2 left-hand. Right-handedness had accordingly developed under pressure of muscular effort in the sixth and seventh months.

4. Up to this time the child had not learned to stand or to creep; hence the development of one hand more than the other is not due to differences in weight between the two longitudinal halves of the body. As she had not learned to speak or to utter articulate sounds with much distinctness, we may say also that right or left handedness may develop while the motor speech center is not yet functioning.

5. In most cases involving the marked use of one hand in preference to the other, the second or backward hand followed slowly upon the lead of the first, in a way clearly showing symmetrical innovation of accompanying movements by the second hand. This confirms the inference as to such movements drawn from the phenomena of mirror-writing, etc., by Fechner and E. H. Weber.

Some interesting points arise in connection with the interpretation of these facts. If it be true that the order of rise of mental and physiological functions is constant, then for this question the results obtained in the case of one child, if accurate, would hold for others apart from any absolute time determination. We would expect, therefore, that these results would be confirmed by experiments on other children, and this is the only way their correctness can be tested.[3]

If, when tested, they should be found correct, they would be sufficient answer to several of the theories of right-handedness heretofore urged. The phenomenon can not be due to differences in balance of the two sides of the body, for it arises before the body begins to stand erect. It can not be due to experience in the use of either hand, since it arises when there is no such difference of experience, and since the hand preferred is used, as a matter of fact, for purposes for which in experience the other would be altogether more convenient.[4] The rise of the phenomenon must be sought, therefore, in more deep-going facts of physiology than such theories supply.

If, on the other hand, heredity be brought to the aid of these "experience" theories, it is possible to claim that, as structure is due to function, experience of function must have been first; and only thus could the modification in structure which is now sufficient to produce right-handedness in individual cases have been brought about. On the other hand, if we go lower in the animal scale than man, analogies for the kinds of experience which are urged as reasons for right-handedness are not present; animals do not carry their young, nor pat them to sleep, nor do animals shake hands! It must therefore be shown that animals are right or left handed, or that they differ in some marked respect in regard to function, in their nervous make-up, from man. Admitting the need of meeting these requirements; admitting again that we have little evidence that animals are dextral in their functions; admitting also the known results as to the control of the two halves of the muscular system by the opposite brain hemispheres respectively; admitting further that the motor speech function is performed by the hemisphere which controls the stronger side of the body, and is adjacent to the motor arm center in that hemisphere; and admitting, finally, that the speech function is one in which the animals have little share—all these admissions lead us at once to the view that there is a fundamental connection between the rise of speech and the rise of right-handedness.[5]

Looking broadly at the methods of nervous and muscular development, and accepting all the results of neurology we are able to gather, we may say that in the differentiation of functions in the animal series certain principles may be recognized: 1. The deep-seated vital functions represent least nervous differentiation, as is seen in the simple organs known as the lower nervous centers. 2. New unsymmetrical functions give a differential or twofold organic development, the great instance of which is found in the cerebral hemispheres. 3. New symmetrical or unilateral functions find their counterpart each in one of three kinds of nervous adaptation: (a) co-ordination of the hemispheres in a single function—i. e., functions which are crippled if either hemisphere is damaged; (b) co-ordination of particular functions in each hemisphere—i. e., functions which are not crippled unless both hemispheres are damaged) and (c) co-ordination of particular functions in one hemisphere only—i. e., functions which are crippled if one selected hemisphere is damaged. All these kinds of co-ordination exist.

It is easy to see that both speech and right-handed function belong under the last head of the last class—co-ordinations of particular functions in one hemisphere only—and that they belong in the same hemisphere. Why is this? What have they in common?

A very essential kind of hand movements are the so-called "expressive" movements, meaning those which serve to convey a meaning, or express a state of consciousness. Of course, speech is par excellence the function of expression. It is further only a part of the position upon which the psychological theory of expression is based, that all movements are in so far expressive, and that details of expression and its relative fullness are matters of co-ordination. Now, this co-ordination has attained its ripest and most complex form, apart from speech, in movements of the hand. Upon this view it is easy to hold that right-handedness is a form of expressive differentiation of movement, and that it preceded speech, which is a further and more complex form of differentiation and adaptation.

The neurological basis upon which this hypothesis rests is adequate, and affords a presumption as to the psychological development as well. The facts I have now given, for the first time, go some way to support the view: 1. Right-handedness arose before speech in the child H——. 2. Imitation by the hand of movements seen arise before articulate imitations of sounds heard;[6] this in spite of the fact that hearing, in its development in the child, becomes perfect before sight. 3. Characteristic differences in children in respect to their general mobility of arm and hand, manual skill, and quickness of manipulation, extend also to speech. As compared with my other child, E——, the first-born, H——, is remarkably agile and motile generally in her temperament; and her speech development was relatively much earlier and more rapid.

It is further interesting to note that musical ability is associated with speech ability—a connection which would be expected when one takes due account of the expressive character and function of music. As far as theories of the rise of musical expression have gone, they unite in finding its beginnings in the rudimentary emotional expressions of the animals. The singing of birds is undoubtedly connected with their mating instincts. Pathological cases also show a marked connection between musical execution and speech, to the extent that, while musical defect almost invariably involves speech defects, the reverse is much less generally true—a fact which confirms the view that music is an earlier form, but still a form, of expressive reaction.

Late observations also show, as far as they are sufficient, that the center for music expression is also located normally in the left hemisphere for right-handed persons. Oppenheim reports a case[7] of total aphasia with total amusia (lack of musical ability from disease) in which the recovery of speech brought with it musical recovery also. Furthermore, another case of Oppenheim's shows motor aphasia with motor amusia only—i. e., the patient could still understand tunes, and, further, could imagine tunes "in his head" (what the French call intérieur), while he could not sing them. This shows a close connection in locality between motor speech and motor music function, while a slight separateness of the two centers in locality in the left hemisphere, explains cases of motor aphasia in which musical execution is preserved. Further, Frankl-Hochwart declares that no cases are recorded of amusia from lesion in the right hemisphere,[8] and Starr says (in a private letter) of a patient of his:[9] "My patient is right-handed, and music does follow speech in being unilaterally located; . . . it is well proved that the musical faculty is onesided in location." Despite these positive opinions, I think more critical cases with autopsy are necessary to make the position quite secure.

All this means simply that the general cause to which is due the fact of right-handedness is also the cause, through further differentiation and emphasis in the same local seat, of the development of musical ability and of speech. It now remains to ask: What was or is this cause, and when in the race-history series did it begin to operate? There are only two hypotheses of any force—either "experience" or "spontaneous variation" at some stage in biological development.

It is extremely improbable that dextrality should have arisen among the quadrupeds (or amanous bipeds), for experience was lacking of unilateral stimulation, and a spontaneous variation of this kind would have produced such inconvenience of locomotion and ultimately such asymmetry of form that it would have been weeded out.[10] As an extreme example, fancy a bird which is dextral in its flight.[11]

As soon as we come to bipeds with hands, however, these reasons do not hold. Their locomotion does not depend on manual symmetry, and any dextrality, however slight, would be of direct advantage in climbing, fighting, breaking sticks, and pulling fruit; since a disproportionate growth of one side would give that side greater strength than either side would possess in animals of symmetrical development in the same environment. A very strong one-armed man can keep at bay a weaker man with two arms, or destroy him, and this is emphasized in animals, where brute force is the only resource. It is difficult to find, however, in the habits of simians any ground for believing that there has been a form of unilateral stimulation which would act to effect a structural change in one hemisphere over and above the other. But, apart from this, there is every reason to expect, quite independently of function, that two organs of such comparative separateness and independence of function would not remain exactly balanced in function; in short, spontaneous variations giving advantageous dextrality would inevitably arise and persist as soon as the habits of life were not such that more important functions, such as locomotion, tended to suppress them and restore bilateral equilibrium.[12] There are, as far as I know, very few published observations of fact in regard to simian or animal dextrality.[13]

It is likely, therefore, that right-handedness in the child is due to differences in the two half-brains, reached at an early stage in life, that the promise of it is inherited, and that the influences of infancy have little effect upon it. Yet, of course, regular habits of disuse or of the cultivation of the other hand may, as the child grows up, diminish or destroy the disparity between the two. And this inherited brain-onesidedness also accounts for the association of right-handedness and speech—the speech function being a further development of the same unilateral potency for movement found first in right or left handedness.

The Marquis of Salisbury has been nominated as president of the next meeting of the British Association, which will be held in Oxford, August 8th. In proposing him, Sir F. Bramwell mentioned, as among the claims of the marquis, that he had been Chancellor of the University of Oxford since 1880, that he would therefore represent both hosts and guests, that he was a distinguished statesman, a courteous gentleman, a member of the Council of the Royal Society, and a true man of science. Ipswich has been designated as the place for the meeting of 1895.
  1. Cf. Vierordt's remarks, Physiologie des Kindesalters, pp. 428, 429. For a detailed statement of theories on this topic, see chapter x of the very learned monograph on The Right Hand: Left-handedness, by my late lamented colleague and friend Sir Daniel Wilson.
  2. Science, xvi, October 31, 1890; discussed by James, Science, November 8, 1890, by Dr. J. T. O'Connor, ibid., xvi, 1890, p. 331, and by myself, ibid., xvi, November 28, 1890. The report is quoted in full in Nature, November 13, 1890, and in part in the Illustrated London News, January 17, 1891. See also Ebbinhaus's Zeitsch. für Psychologie, ii, 1891, p. 239; Wilson, The Eight Hand: Left-handedness, pp. 128-131; Revue Scientifique, 1891, ii, p. 493; discussed by Mazel, Revue Scientifique, 1892, i, p. 113. Both writers in the last named journal cite these experiments wrongly as Wilson's.
  3. Vierordt says concerning such experiments: "Adequate observations are wanting on the grasping movements of the infant's left and right arm a a—kind of observations which would be of the first importance for this inquiry" (Physiologie des Kindesalters, p. 428); and Wilson: "Only a prolonged series of observations, such as those by Prof. Baldwin already noted, made at the first stage of life, and based on the voluntary and the unprompted actions of the child, can supply the needful data" (Left-handedness, p. 209).
  4. An additional point, which I think is true, is that a right-handed child learns to shake hands properly using—the more inconvenient hand across his body—more easily than the left-handed child.
  5. This much has been before surmised (see Mazel, Revue Scientifique, 1892, i, p. 113). He makes no attempt, however, to account for the association, except by calling both functions expressive.
  6. It is interesting that of both hand and speech movements the latest to be lost in disease are those involved in the so-called "mimicry" of movement and in imitative speech.
  7. Charité Annalen, xiii, 1888, p. 286.
  8. This means that all cases noted have been right-handed. Deutsche Zeitsch. für Nervenheilkunde, 1891, i, p. 295 and foot-note.
  9. Referred to in The Psychological Review, January, 1894, p. 92.
  10. For this reason the human leg, as Brown-Séquard says, is not as right-sided as the arm.
  11. The only evidence I know of such a thing is that a cat swims in a circle; but then dogs and horses do not.
  12. It is on this point that I differ from Wilson, who claims that while some are naturally right or left handed, most people owe the peculiarity to education; the evidence, apart from my experiments, is well put by Mazel, loc. cit.
  13. I know only the assertion of Vierordt that parrots grasp and hold food with the left claw, that lions strike with the left paw, and his quotation from Livingstone i. e., "All animals are left-handed" (Vierordt, loc. cit., p. 428). Dr. W. Ogle reports observations on parrots and monkeys in Trans. Royal Med. and Chirur. Society, 1871. Dr. Ogle informs me in a private letter that the chimpanzee which recently died in the Zoölogical Garden in London was discovered by him to be left-handed. I have addressed a circular letter to some of the officials in zoological institutions here and abroad, and hope to gather some facts in this way. It is evident that on this theory of spontaneous variation any change which produced a permanent organic superiority of one hemisphere would be sufficient, and the view that the difference in the hemispheres is due to a better blood supply to the left hemisphere might thus have its justification. As a matter of fact, the arterial arrangements do seem to indicate a more direct blood supply to the left hemisphere (cf. the note of Dr. J. T. O'Connor, apropos of my experiments, in Science, xvi, 1890, p. 3a 1). It is an interesting inquiry whether this arterial arrangement is reversed in left-handed persons. Wilson cites two cases in which there was no such correspondence (loc. cit., p. 179).