Side biases for holding and carrying infants: Reports from the past and possible lessons for today

Abstract

Most adults hold human infants on the left side, with the infant's head to the left of their own body midline. The discovery of this bias is credited to Lee Salk, who first reported it in 1960, but the same was reported at least 300 years earlier and many times again through the early decades of the twentieth century. Along with the left-side reports, however, others named the right as the preferred side. Authors on each side explained the preference and foresaw consequences for the infant: different ones in each case. This article describes the two kinds of reports, asks whether and how they might be reconciled, and discusses their possible lessons for theory and research today.

Keywords:

• Infants
• Cradling
• Side of holding and carrying
• Handedness
• Laterality
• History

Notes

¹The reports cited in this introduction are only examples. For further citations, along with more extensive accounts and assessments of the proposed explanations of the bias as presented below, several reviews are now available (Damerose & Vauclair, 2002; Donnot & Vauclair, 2005; Harris, 2007; Turnbull & Lucas, 2000).

²The tables show a total of 31 reports, but they come from only 30 different published accounts because one author, who reported right as well as left biases in different groups, appears in both tables.

³The cavities refer to the chambers of the heart, but their identity is uncertain. According to Lloyd (1973, p. 184, fn. 32), the possibility that best fits Aristotle's account is that offered by D'Arcy Thompson who, in his translation of Historia Animalium (Aristotle, 1910), supposed that the right-most, largest chamber referred to the right auricle and ventricle combined, which would account for the other chambers being “far smaller.”

⁴In some reports the infant's hand was said to be constrained, in others it was the arm, and in still others it was both. My descriptions follow the original reports. Although no report pointed this out, it is possible that the outside hand (or arm), instead of being completely free, was partly constrained by the holder's arm, making it free only relative to the inside hand, that is, the one pressed directly against the holder's body. Likewise, the inside hand might not be completely constrained by the holder's body, or it might not be constrained at all by the holder's body but rather by the infant putting that hand around the holder's neck for support. These variations are illustrated in drawings and photographs reproduced in this article.

⁵It is also consistent with age differences in swaddling. Through at least the mid-nineteenth century, most infants in Western countries were swaddled. The Greek physician Soranus (fl. early second century ce) set the standard: wrap the newborn in soft cloths from the head “to the very tip of the toes” (1956, Book II, p. 85), which restrained movement and kept the infant from scratching or otherwise injuring itself. The main goal was to guide its physical development so that it would not grow “Crooked, Awry, or Lame” (Mauriceau, 1710, p. 367). During the first weeks, the infant was kept swaddled except for cleaning and changing. After that, the cloths were removed in stages, first from the upper body, arms, and hands usually between the first and fourth months, then from the torso and legs by the eighth or ninth month or at latest by the end of the first year (Nelson, 1756, p. 105), the order corresponding to when each body part had presumably become firm enough that “there is no longer any fear” of its “being distorted” (Soranus, 1956, Book II, p. 114). The upper limbs were freed first because they were ready first. Then, “having his arms and hands at liberty, [the child] may use and stirre them” (Guillemeau, 1612, p. 22). For the hypothesis linking side-of-holding to infant handedness, side-of-hold would have been less of a factor for fully “wrapped” infants because both hands and both arms were constrained. For holding side to affect hand use asymmetrically, at least the upper limbs would have to have been freed, meaning that the infants were at least 1 to 4 months old, with 3 months as the more likely lower bound (see Harris, 2003, for further details).

⁶It is safe to assume that all the nurses were female. In the historical periods under review nursing was overwhelmingly women's work, and male nurses, in any case, were not in charge of infants and young children. A census in 1901 for England and Wales estimated 63,500 female nurses and 5700 male nurses, with 3000 of the men working in mental institutions and with most of the rest in their patients’ homes. The ratios had not changed much by 1931 (Abel-Smith, 1975, p. 52 and Appendix I).

⁷Infants in maternity hospitals most likely would have been children of the very poor because, in Europe and the United States until about the mid-nineteenth century, most maternity hospitals were charitable institutions for indigents. These were dangerous places—many infants and mothers died from contagious disease (childbed fever, or puerperal sepsis)—so women who could afford to were delivered at home by physicians or midwives. In the United States a significant shift to hospital births for all social classes began only in the 1920s (Thoms, 1960; Wertz & Wertz, 1989, esp. chs. 4 & 5).

⁸The baby carriage was invented in 1733 but not commonly used until the 1870s and 1880s (Bellis, 2004; Brown, 2004). It brought the nurse some relief from carrying the infant when outdoors, but she still had to lift and carry it back and forth from carriage to house. In one report cited in this article, the nurse was observed while performing these acts.

⁹Private nurses should be distinguished from wet-nurses: women who, through the late-eighteenth century, were employed by most middle- and upper-class families to suckle their infants. Ordinarily, infants were “farmed out” to women in the country. For wealthy families, the wet-nurse more often lived in the employer's home (Still, 1965; Sussman, 1982, especially ch. 2), thus putting her within the purview of the authors of these early reports. But since only three holding-side reports referred to nursing, one involving mothers, if wet-nurses figured in any of the remaining reports, they evidently were not observed while performing their primary duties.

¹⁰This was the English translation of a German work written nearly a century earlier, in 1563. The translation, however, marked its first appearance in any form, and then only as an addendum to another translated work, Wurtz's (1656) An Experimental Treatise of Surgerie, in Four Parts.

¹¹The 1878 date is to the American magazine Harper's Weekly, where the letters also appeared. I have not seen the Daily Telegraph letters.

¹²This was one of over 40 examples given by Hall and Hartwell in support of the then-common idea that growth was more vigorous on the right side of the body. Other examples included growth of the teeth, hair and beard, and nails. By contrast, many diseases and congenital defects were believed to be more common on the left side, including fever and resulting weakness, pneumonia, cancer, epilepsy, and deformities of the lung, kidney, eye, and ear. All such differences were seen as evidence of the superiority of the right side of the body and, correspondingly, of the left side of the brain. Most of the examples, including the one for breast milk, were undocumented in Hall and Hartwell's article, although some were among those reported and documented by the French neurologist Armand Trousseau (1868) in his Clinique Médicale (Volume 2).

¹³This image is from the website VRoma (2008). The editor, Barbara McManus, notes that the temple in the background suggests that the standing woman might be a priestess, who may have just given the child to the mother. The man is perhaps the father.

¹⁴Wren's notes are undated, but given the publication date of the book, they would have been written between 1646 and 1658, the year of his death.

¹⁵Times have changed, but even today direct observations as well as family portraits and photographs show that the woman more often holds the child, especially if it is an infant (Harris et al., 2007; Lockard et al., 1979; Manning, 1991; Rheingold & Keene, 1965; Roopnarine, Talukder, Jain, Joshi, & Srivastava, 1990).

¹⁶In putting on the toga it was first thrown over the left shoulder so that about 6 feet hung in front and about 12 feet behind. The longer portion was then carried under the right arm and then diagonally across the chest and over the left shoulder again. A fold of this portion hanging in front formed the “sinus” (notes by Butler, in Quintilian, 1936, p. 318).

¹⁷There is one other kind of image that I have not included: men holding infants they have killed or are about to kill. Needless to say, their holds were not for giving comfort. For example, in Raphael's fresco The Judgment of Solomon (1509–1511; 1518–1519), the infant is held or swung roughly by one limb. In that instance, of course, the infant was spared.

¹⁸Buchanan did not identify the translation. In the Farquharson translation (Aristotle, 1912a, 705b, 30), the passage reads: “all men carry burdens on the left shoulder; in this way they set free the side which initiates movement and enable the side which bears the weight to be moved.”

¹⁹In describing the right hand as “delicate” in the sense of “agile”, Burt was evidently referring to a different quality from what Kenealy (1920) meant when she called the left hand more “delicate”. Thus Burt, while identifying the left hand as the weight-bearer, also incorporated the free-the-right-hand explanation into his account.

²⁰In that case, the belief presumably would override actual physiological differences, since the authors noted that the milk supply has been “abundant on both sides”, making it unlikely that Tanka women have “a poor milk secretion from the left breast owing to a genetic predisposition” (Ing et al., 1977b, p. 657).