Content uploaded by Ian Christopher Mcmanus
Author content
All content in this area was uploaded by Ian Christopher Mcmanus on Dec 28, 2020
Content may be subject to copyright.
3
The history and geography of human handedness
I. C. McManus
Summary
About 90% of people are right-handed and 10% are left-
handed. Handedness is associated with functional lateraliza-
tion for cerebral dominance, and may also be associated with
various types of psychopathology. Broadly speaking, the vast
majority of humans seem to have been right-handed since the
emergence of the genus Homo, some three to four million
years ago. Likewise, in all societies studied, there is a large
excess of right-handers. However, there have been few studies
exploring the detailed history and geography of handedness,
not least because adequate pre-twentieth-century historical
data are difficult to find, and very large sample sizes with
consistent measurement methods are required for geograph-
ical studies. This chapter overviews the various sets of data that
provide insight into handedness’s history and geography.
It is probable that about 8% to 10% of the population has
been left-handed for at least the past 200 000 years or so.
Detailed data only began to become available for those born
in the nineteenth century, and there is growing evidence that
the rate of left-handedness fell precipitously during the
Victorian period, reaching a nadir of about 3% in about 1895
or so, and then rising quite quickly until an asymptote is
reached for those born after about 1945 to 1950, with 11% to
12% of men and 9% to 10% of women typically being
left-handed in Western countries. The sex ratio seems to
remain constant, not only during historical changes but also
with geographical differences, and is presumably the result of
a biological rather than a cultural process.
Geographical differences in handedness are clearly appa-
rent both between continents (as in Singh & Bryden’s, 1994,
comparison of Canada and India) and within continents: rates
in Europe seeming to be highest in Britain, Holland, and
Belgium, and falling away towards the east and south, and
within countries, seen well in Stier’s (1911) study of the
German Army, in Leask and Beaton’s (2007) study of the
United Kingdom, and between the various states of the USA, in
the very large Gilbert and Wysocki (1992) database.
Ethnic differences in handedness are related to geographi-
cal differences, with left-handedness generally being more
common in White, Asian and Hispanic populations –a differ-
ence seen both in the UK, and historically in the United States,
where the difference between ethnic groups has grown smaller
during the twentieth century, but was still present even for
those born in the 1970s. Migration studies in the UK show that
the lower rate of left-handedness in those from the Indian
sub-continent is similar in those born in the UK and those
born outside the UK, implying that genes rather than environ-
ment are the primary source of the difference.
Different rates of left-handedness can reflect either environ-
mental or genetic differences between societies, and rates
alone cannot distinguish the two processes. However, a math-
ematical model shows that effects of different social pressure
or gene frequencies can be distinguished if family data on
handedness are available. That model suggests not only that
geographical differences but also historical differences primar-
ily reflect changes in gene frequency rather than direct social
pressure.
Introduction
The important discoveries of Dax and Broca in the
nineteenth century showed that human brains are
functionally asymmetric, most people processing lan-
guage in their left hemisphere (Finger, 1994; Finger &
Roe, 1999). However, it soon also became clear that a
minority of people process language with their right
hemisphere (Harris, 1991; Harris, 1993a), so that lan-
guage processing can be seen as what geneticists call a
Language Lateralization and Psychosis, ed. Iris E. C. Sommer and René S. Kahn. Published by Cambridge University Press.
© Cambridge University Press 2009.
polymorphism, there being two qualitatively different
types, akin to human blood groups. Since at least the
beginnings of recorded history, and probably long
before, people have also noted that while most people
are right-handed, a minority of individuals are the
opposite way around, being left-handed. Handedness
and language dominance also show a moderate corre-
lation, although the pattern is somewhat counter-intui-
tive, about 5% to 6% of right-handers showing right
hemisphere language dominance, compared with
about 30% to 35% of left-handers.
Language dominance is not easy to assess reliably in
large populations, with techniques such as functional
Magnetic Resonance Imaging (fMRI) (Pujol et al., 1999)
or transcranial Doppler (Knecht et al., 2000) requiring
complex technology that is expensive and not particu-
larly portable, while dichotic listening and tachisto-
scopic hemi-field studies are not particularly reliable
within individuals. As a result, handedness, which is
easily assessed by questionnaire or direct observation,
has been studied both as an important lateralization in
its own right, and also as a surrogate for language
dominance. Handedness is thought by most research-
ers to be genetic in origin, although there are differ-
ences in the precise models (McManus & Bryden,
1992), and, perhaps crucially, most models also assume
that the genes determining handedness also influence
language dominance, making the study of handedness
directly relevant to the study of language dominance. If
left-handedness is under genetic control, as several
theories suggest, then it is likely, as with other genet-
ically determined biological characteristics, such as
blood groups, that there will be geographical variation
(or clines), because of some combination of genetic
drift, founder effects, and selection, be it natural or
artificial.
A simplifying assumption for many earlier studies of
handedness, and here the present author is no excep-
tion (McManus, 2004), has been to regard either the
rate of left-handedness itself, or the frequency of the
underlying genes, as constant historically and geo-
graphically. However, neither proposition seems likely
a priori, not least because almost all human polymor-
phisms vary geographically (see, e.g., Cavalli-Sforza,
Menozzi & Piazza, 1994), and the frequency of some
polymorphisms, such as that of sickle-cell anemia, also
varies historically in relation to changing selection
pressures (Cavalli-Sforza & Bodmer, 1971). It therefore
seems probable that left-handedness, and perhaps the
genes underlying it, will also vary both geographically
and historically. If historical and geographical variation
has been little studied by researchers, it is mainly
because of the difficulty of obtaining adequate,
large-scale databases. Attempts at meta-analysis of
multiple small-scale studies have generally been
unsuccessful, mainly because methods of measure-
ment vary almost as much between studies as do rates
of handedness (Raymond & Pontier, 2004; Seddon &
McManus, unpublished manuscript, 1991).
Geographical and historical variation in handedness
also raises the possibility that language dominance will
also vary geographically and historically, as perhaps
will other traits related to handedness and language
dominance, and here one might think of dyslexia, stut-
tering, autism, schizophrenia, etc., in each of which
atypical cerebral lateralization has been implicated.
This chapter will concentrate on handedness, mainly
because there is extensive data concerning it, but
throughout the sub-text will be that similar conclusions
might apply more broadly to cerebral dominance and
its correlates.
Historical differences in the rate of
left-handedness
The previous two centuries
Historical data on left-handedness are surprisingly rare,
to the extent that a museum curator attempting to
curate an exhibition on handedness referred to left-
handers as being “a people without a history”(Sadler,
1997). Although estimating historical rates of left-
handedness might seem easy, until recent years there
has been very little systematic data. Modern work asking
whether the historical rate of left-handedness might have
changed systematically probably begins with that of
Brackenridge (1981). However, quite the most important
modern source on rates of left-handedness is the vast
study by Gilbert and Wysocki (1992), which although
never intended as a study of handedness has emerged
38 Section 1: Asymmetry, handedness and language lateralization
as a key resource. In 1986, National Geographic mag-
azine published a special issue on olfaction (Gibbons,
1986), which was accompanied by a “scratch and sniff”
card, which readers were encouraged to scratch, report
what, if anything, they could smell, and then, after
completing a brief demographic questionnaire, return
the card. Over 1.4 million people did so (Gilbert &
Wysocki, 1987; Gilbert & Wysocki, 1992; Wysocki &
Gilbert, 1989; Wysocki, Pierce & Gilbert, 1991). The
authors of the original study felt it was possible that
handedness and olfaction were linked (perhaps through
cerebral dominance), and therefore Gilbert and Wysocki
included two questions on handedness: one on writing
hand and the other on throwing hand. Subsequent
analyses have found no relationship between olfactory
acuity and handedness, and it seems reasonable there-
fore to regard the survey as unbiased in relation to
handedness (even if it is potentially biased in other
ways, such as in sex, age, ethnicity, and olfactory abil-
ity). Respondents of course also reflect the typical read-
ership of the magazine, which is likely to be more
educated and middle-class than the population as a
whole, but that is unlikely to be a source of bias in
relation to handedness, since other large-scale studies
have shown handedness to be unrelated to social class
or education (McManus, 1981; Perelle & Ehrman, 1994).
The Gilbert and Wysocki data show two key findings.
First, men are about 25% more likely to be left-handed
than women; there being about five left-handed men
for every four left-handed women, a finding that was
also found in a large-scale meta-analysis (McManus,
1991), and helps cross-validate the data. More interest-
ingly, there was also a strong relationship of handed-
ness to year of birth, only about 3% to 4% of those born
before about 1920 being left-handed, compared with
about 11% to 12% of those born after 1950, a three fold
difference. It should also be emphasized that the rela-
tive extent of the sex difference, expressed as an odds
ratio, remained constant for those born in the early or
late twentieth century.
Figure 3.1 shows the Gilbert and Wysocki data in two
versions. The original paper (Gilbert & Wysocki, 1992)
contained only data from 1900 onwards (indicated by
the vertical dashed line), and the solid line shows a
constrained Weibull function, which has been fitted to
the data (see McManus et al., in press a). A reasonable
account of just these data might be that the rate of
left-handedness was low in the nineteenth century,
and then rose through the twentieth century, reaching
its current asymptote in about 1950. Interpreting the
finding is, however, not so straightforward, mainly
because the data are not proper historical series, but
instead are cross-sectional, so that cohort effects must
be inferred from individuals of different ages. The
group born in 1900 in the Gilbert and Wysocki data
were therefore aged 86 when the study was carried
out in 1986. One possibility, extensively discussed in
the handedness literature, is that left-handers die earl-
ier, which results in a lower rate of left-handedness in
older individuals (Coren & Halpern, 1991; Halpern &
Coren, 1988; Halpern & Coren, 1991). Subsequent anal-
yses of other data have convincingly shown that there is
little evidence for differential mortality of left-handers
(Ellis et al., 1998; Halpern & Coren, 1993; Harris, 1993b;
Harris, 1993c; Marks & Williamson, 1991; Wolf,
D’Agostino & Cobb, 1991), although there is one study
that compellingly suggests a higher mortality of young
left-handed males in World War I, perhaps due to hav-
ing to use right-handed equipment (Aggleton et al.,
1994). An alternative explanation of the lower rate of
left-handedness suggests that the elderly are more
likely, because of social pressure, either to have been
forced to shift from writing with the left hand to writing
with the right hand, or they prefer to call themselves
right-handed, because of a taboo against left-
handedness (Hugdahl et al., 1993; Hugdahl, 1996).
Both this and the differential mortality explanation
become unlikely when one looks at the entire Gilbert
and Wysocki database, which included unpublished
data on individuals born between 1887 and 1899 (see
McManus et al., in press a). These data on these very
oldest respondents are shown in Fig. 3.1, and the heavy
dotted line shows the fit of a mixture of two constrained
Weibull functions. Now it is clear that the very oldest
respondents have a higher rate of left-handedness than
those who are somewhat younger, an effect which is
significant (McManus et al., in press a), and is utterly at
odds with explanations due either to differential mor-
tality or greater social pressure to be right-handed. The
best account of the Gilbert and Wysocki data is that it
Chapter 3: The history and geography of human handedness 39
directly reflects the actual rate of left-handedness in the
population.
The additional Gilbert and Wysocki data implies that
the rate of left-handedness might have been falling in
those born in the last decade or so of the nineteenth
century, and subsequently rose again in the twentieth
century. Understanding the history of left-handedness
in the nineteenth century therefore becomes impor-
tant, although it is far from easy, adequate data sources
being few and far between and not easy to interpret.
The earliest scientific estimate of the rate of left-
handedness is that of Ogle (1871), who asked 2000
consecutive patients at St George’s Hospital whether
they were right- or left-handed, 85 (4.25%) responded
saying they were left-handed. Since these patients were
adults in 1871, their mean year of birth was probably
about 1835. Other somewhat later studies providing
estimates of left-handedness rates for those born
before 1900 include Lombroso (1884), Mayhew (see
Crichton-Browne, 1907), Crichton-Browne (1907), Stier
(1911), and Schäfer (1911), and in addition Parson
(1924) and Burt (1937) provide early twentieth-century
estimates, which help to validate the broad picture
shown by Gilbert and Wysocki. Two other sources
have also been analyzed recently. In 1953, the BBC
broadcasted an early television science programme
called Right Hand, Left Hand, to which over 6000
people returned postcards describing their handedness
and basic demographics (McManus et al., in press b).
Although biased, with left-handers being substantially
over-represented among the respondents, it is never-
theless possible to estimate the true rate of left-
handedness, which is of particular interest for the
respondents born before 1900. Finally, ratesof left-hand-
edness have also been estimated from the early docu-
mentary films made by Mitchell and Kenyon between
1900 and 1906, the oldest participants of which were
born before about 1850 (McManus & Hartigan, 2007).
13%
Rate of left-handedness
Year
12%
11%
10%
9%
8%
7%
6%
5%
4%
3%
2%
1%
1860 1880 1900 1920 1940 1960 1980 2000
Figure 3.1 The overall rate of left-handedness in the data of Gilbert and Wysocki, 1992 (with permission), for those born from
1900 onwards (solid line); the fitted line is a constrained Weibull (for further details see McManus et al., in press a). The data before
1900 (birth year 1887–99) are unpublished data from the Gilbert and Wysocki study, and are fitted by the dashed line, which is a
mixture of two constrained Weibull functions.
40 Section 1: Asymmetry, handedness and language lateralization
The films showed large numbers of people waving at
the camera, mostly with the right arm, and given mod-
ern data on the relationship between waving and hand
preference, one can estimate the rate of left-handed-
ness. Perhaps most striking is that left arm-waving is
substantially more common among the older individu-
als, precisely the opposite pattern to that seen in the
Gilbert and Wysocki data where left-handedness is
most common in the younger individuals. The data
from the Mitchell and Kenyon films are almost impos-
sible to explain in terms of differential mortality or
social pressure.
Figure 3.2 is a complex figure, taken from McManus
et al. (in press a), which summarizes all of the historical
data from the nineteenth and twentieth centuries. The
solid black line consists of a mixture of two constrained
Weibull functions, fitted using a maximum likelihood
method, the two pale gray lines showing its compo-
nents. The best statistical description of the recent his-
tory of handedness is that the rate was about 10% at the
end of the eighteenth century, the rate then fell
throughout the nineteenth century, until it reached its
nadir in about 1890–5, and then rose during the twen-
tieth century, reaching its asymptote in about 1950,
after which rates seem to have been unchanged.
The historical reasons for the nineteenth- and
twentieth-century changes are unclear at present, but
the nineteenth century changes may reflect an increas-
ing visibility and stigmatization of left-handers, resulting
from the Industrial Revolution, with large numbers of
individuals using complex machinery in mills and facto-
ries, coupled with increasing rates of education and
literacy (Stephens, 1990; Stone, 1969; West, 1978). In
an agricultural society, left-handers are relatively invis-
ible (except perhaps, as Thomas Carlyle noted, when a
group of men is scything a field, see Pye-Smith, 1871).
However, both complex machines and education
would not only have made left-handers more visible,
but left-handers may also have appeared less capable
and more clumsy, as left-handed adults worked on
20%
10%
Rate of left-handedness
Year
9%
8%
7%
6%
5%
4%
3%
2%
1%
1760 1780 1800 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000
Figure 3.2 Summary of data from multiple studies on the rate of left-handedness during the nineteenth and twentieth
centuries, from McManus et al. (McManus et al., in press a). The solid black line is the fitted mixture of two constrained Weibull
functions, which are shown separately as the two solid gray lines. Note that the ordinate is on a logarithmic scale.
Chapter 3: The history and geography of human handedness 41
machines that were almost certainly designed with
right-handers in mind, and left-handed children were
taught to write with steel dip pens that needed to be
dragged across the paper from left to right by
right-handers, and were not capable of being pushed
across by the left hand without digging into the paper
and making blots and stains. Whatever the mechanism,
it seems undoubted that there was a general stigmati-
zation and discrimination against left-handers at
the end of the nineteenth century, which Bertrand
(2001, pp. 88 and 91) refers to as “La haute époque
de l’intolérance”, such that there was “La gaucherie
persécutée”.
The distant past
The history of handedness before 1800 consists almost
entirely of a few isolated points, which often are illumi-
nated only briefly through indirect evidence that has to
be treated with great care. Claims that, for instance,
left-handedness was much more common in medieval
than modern Britain (Steele & Mays, 1995), must be
treated with caution, because they are based on bone
asymmetries, which even in modern samples are inac-
curate indicators of handedness, Steele (2000) pointing
out how, “perplexingly …left-handed subjects are
equally likely to have a stronger grip in either hand”
(p. 205). Likewise, although it is often hoped that cul-
tural artifacts may provide insight into rates of handed-
ness, interpretation is often difficult. For instance,
although the twist of spun cotton or other fibers (“Z”-
or “S”-twist) might at first seem to indicate handedness,
the relationship of spinning direction to handedness
seems to be weak (Minar, 2001), different fibers such
as cotton and flax naturally twist in opposite directions
(Batigne & Bellinger, 1953), there is evidence of com-
munities of practice in different directions (Minar,
2001), and technological development can override
pre-existing manual asymmetries (Crowfoot, Pritchard
& Staniland, 2001).
No attempt will be made to be inclusive, although the
broad picture that emerges, which is shown synopti-
cally in Fig. 3.3, is fairly straightforward. Note in partic-
ular that the time axis for Fig. 3.3 is logarithmic, in
terms of years before the present. The right-hand end
of the figure shows the last two centuries, with a mod-
ern rate of left-handedness of about 11% (section a).
The rate was similar, at perhaps 8% to 10% at the end
of the eighteenth century, but then fell to 3% or so
during the nineteenth century, rising again in the first
half of the twentieth century (sections b and c).
For the past 5000 years the best historical data are the
elegant study by Coren and Porac (1977), which looked
at five millennia of artistic representations of unima-
nual activity (such as playing board games, throwing
spears, writing, etc.). Overall about 8% of paintings,
drawing, and sculptures show the left hand being
used, with little variation over the entire period of
recorded history (section d in Figure 3.3). Specific writ-
ten references to left-handedness are rare, with the
intriguing exception of a use for left-handed workers
in Roman stone mines, where a left-hander and a
right-hander worked cooperatively on removing blocks
of stone in the very confined spaces of a mine (see
Steele & Uomini, 2005, p. 229 for an account of the
various work of Röder, Bedon, and Monthel).
Data on handedness from the prehistoric period and
pre-literate societies are necessarily indirect, take many
forms, and can be difficult to interpret; see Steele and
Uomini (2005) for an overview. Frustratingly, some
data, such as one of the two arrows carried by the “Ice
Man”, Ötzi, which had been fletched in the
left-handed manner (Spindler, 1994), undoubtedly
indicate the presence of left-handers, but do not allow
an accurate estimate of the rate. However, the study of
Spenneman (1984), looking at stone and bone tools
from the Neolithic period of about 4000 BP (before
present), found a rate of left-handedness of between
6% (of 597 tools at Twann in Switzerland) and 19% (of
51 tools at Bodman in Germany). The data of Cahen
et al. (1979), from the Upper Paleolithic period of about
9000 BP found one likely left-handed toolmaker among
22 (5%), with left-handed knapping and counter-
clockwise rotation marks. The study by Faurie and
Raymond (2004) of silhouetted hand prints on the
walls of Upper Paleolithic caves from about 30 000 to
10 000 BP also allows a proper estimate of the rate.
About 77% of prints showed a left hand, a figure that
the authors showed was almost identical to that pro-
vided by a modern group of 179 students carrying out
42 Section 1: Asymmetry, handedness and language lateralization
the same task, 14 of whom were left-handed, implying a
similar Upper Paleolithic rate of left-handedness to
that of the modern period (point e in Fig. 3.3).
A study of a much earlier population by Fox and
Frayer (1997), looked at tooth striations on Neanderthal
teeth from about 130 000 BP (see also de Castro,
Bromage & Jalvo, 1988). These striations probably
come either from techniques for eating meat, or from
the use of animal tendons or plant matter as primitive
“dental floss”to remove interdental detritus. In the 20
specimens, the direction was compatible with right-
handed use in 18 cases and left-handed use in 2
cases, giving an estimated rate, albeit not a particularly
accurate one, of about 10% (point f in Fig. 3.3). Earlier
than this, there is once again clear but isolated evidence
of the presence of a left-hander who was knapping
stone tools at the Boxgrove Site of about 500 000 BP
(Roberts & Parfitt, 1999). Phillipson (1997) also looked
at edge modification in 54 stone tools from the Lower
Paleolithic period of about 500 000 to 1 000 000 years
BP at Kariandusi in Kenya, and suggested that 6 (11%)
were compatible with left-hand use.
Undoubtedly the oldest data on human handedness
are those of Toth (1985) (see also Ambrose, 2001) who
looked at the flakes left by the stone tool making of
Homo habilis at the site of Koobi Fora in the African
Rift Valley, which is from about 1.5 million years BP.
There was a modest excess of flakes typical of those
produced by right-handers, which was entirely com-
patible with the rate of such flakes found in modern
knappers who are known to be right-handed. The
implication, albeit not a strong one, is that perhaps all
humans at that time were right-handed (indicated as
point g in Fig. 3.3). Without repeating the theoretical
arguments again here, elsewhere (McManus, 1999) I
have argued that handedness in humans is likely to
have evolved in two stages, in the first of which was
the evolution from an ancient C* gene to what I call the
Dgene, when the majority of humans became
right-handed, and a second, subsequent stage, with
0%
50%
40%
30%
20%
10%
1010 M1 M 100 K10 K1 K 100
Years BP
Rate of left-handedness
1950
1914
1800
7–8 000 000 BC
1–2 000 000 BC 10–30 000 BC
140 000 BC 3000 BC
Primitive
great ape
ancestor
Primitive
mammalian
ancestor
(a)
(b)
(c)
(d)(e)
(f)
(g)
(h)
(i)
Figure 3.3 A synoptic map of the broad changes in the rate of left-handedness over the past ten million years.
Note that the abscissa is logarithmic in terms of years before present.
Chapter 3: The history and geography of human handedness 43
the evolution of the modern Cgene, when a substantial
minority of humans became left-handed, the polymor-
phism of Dand Cgenes presumably being maintained
by heterozygote advantage or some other mechanism.
Modern humans evolved from a primitive great ape
ancestor, perhaps about 7 to 8 million years ago, and
that primitive great ape ancestor must itself have
derived from a primitive mammalian ancestor. The
handedness of modern great apes is controversial,
with some researchers believing that great apes do not
show population level handedness (i.e., 50% are
right-handed and 50% are left-handed (Annett &
Annett, 1991; Marchant & McGrew, 1991; Marchant &
McGrew, 1996). However, a meta-analysis by Hopkins
(2006) suggested that perhaps 60% of bonobos and
maybe 55% of chimpanzees and gorillas show
right-handedness, although there are concerns that
some of the difference from 50% may result from imi-
tation learning in captive animals.
Geographical differences in the rate
of left-handedness
People everywhere are mostly right-handed, as was
recognized as long ago as 1837 by the English physi-
cian, Sir Thomas Watson, who wrote:
The employment of the right hand in preference to the left is
universal throughout all nations and countries. I believe no
people or tribe of left-handed persons has ever been known to
exist. …Among the isolated tribes of North America which
have the most recently become known to the civilized world,
no exception to the general rule has been met with. Captain
Back has informed me that the wandering families of
Esquimaux, whom he encountered in his several expeditions
towards the North Pole, all threw their spears with the right
hand, and grasped their bows with the left. (Watson, 1836)
Watson’s strong theoretical position is still accep-
table today, as also is Back’s perception of the right-
handedness of the “Esquimaux”, Delacato (1963)
reporting that in photographs of 46 Canadian and
Greenland Inuit using “an arm for one purpose or
another”, 43 were using the right hand and only 3
(6.5%) were using the left hand. Likewise data from
New Guinea (Connolly & Bishop, 1992), Amazonia
(Bryden, Ardila & Ardila, 1993) and Tristan da Cunha
(McManus & Bryden, 1993) all support the universal
predominance of right-handedness, but their small
sample sizes usually preclude any other detailed com-
parison of rates and the drawing of any strong conclu-
sions on mechanism and process.
Although Watson was correct that right-handers pre-
dominate in all human societies, the related question of
whether rates of left-handedness vary between coun-
tries is much more open to contention. Despite there
being many papers in the literature with titles such as
“The rate of left-handedness in ….”, such studies usu-
ally say little about whether countries differ because
they typically use different methods to measure hand-
edness, making it unclear whether differences are due
to the method of measurement or a difference in the
true rate of left-handedness. Indeed Raymond and
Pontier (2004), after their long meta-analysis, could
still only entitle their paper, “Is there geographical var-
iation in human handedness?”The problem of finding
geographical differences is compounded by the fact
that sample sizes are typically small (and although
several hundred individuals may seem reasonable, it
is not). Detecting differences in small proportions of
individuals between populations requires surprisingly
large samples, as can be seen even with the seemingly
straightforward question of sex differences in the rate of
left-handedness. We now believe that there about five
left-handed males for every four left-handed females,
male to females ratios of 1.238, 1.211, 1.207, 1.343,
and 1.273 being found in the very large studies of
Gilbert and Wysocki (Ross et al. 1992), Halpern et al.
(1998), Peters et al. (2006), Carrothers (1947), and the
meta-analysis of Seddon and McManus (unpublished
manuscript, 1991). However, to have an 80% chance of
finding such a difference with a one-tailed test at the 5%
significance level requires about 2500 males and 2500
females, a number that is far larger than in most of
the studies that had looked at sex differences (and
therefore, for instance, the conclusion of Erlenmeyer-
Kimling et al. (2005), that 517 children of schizophrenic
parents did not show the standard sex difference in
rates of left-handedness, is very unsafe). Using a similar
calculation, when the rate of left-handedness is 10% in
one population, then to find a significant difference
44 Section 1: Asymmetry, handedness and language lateralization
with 80% power at the 5% level when the true rate in a
second population is 5%, 6%, 7%, 8%, or 9% requires
samples in each population of 350, 600, 1100, 2500, and
11 000, making it unlikely that most studies will reliably
be able even to find quite largish differences.
One of the clearest studies to look systematically for
differences in handedness between countries was that of
Singh and Bryden (1994), which used large samples of
students in Canada and India, two countries expected to
be very different in their rate of handedness, and it used
the identical questionnaire in both countries. The rate of
left-handedness was 9.8%in Canada compared with only
5.2% in India, a nearly twofold difference, with factor
structure being very similar (see also Singh et al. 2001).
A parallel study comparing Canada and Japan found an
even larger difference, the rate of left-handedness in
Japan being only 4.7% (Ida & Bryden, 1996). Another
study finding clear differences between countries in the
rate of left-handedness is the important study of Perelle
and Ehrman (1994), which benefited both from a large
sample size and a single consistent questionnaire trans-
lated for use in all the countries.
The very large sample sizes needed for proper geo-
graphical studies of handedness, which allow some
form of mapping, are often only available when the
data have been collected for some other purpose, with
handedness being tagged on as an additional question
(as for instance in the National Geographic study,
described earlier). A similar situation exists in the case
of a recent internet-based study of sexual behavior and
attitudes, which was carried out under the auspices of
the BBC (Reimers, 2007). The survey was live from
February 2005 to May 2005, during which time more
than half a million people provided some data and
255 116 individuals completed all six sections of the
study. One of the questions asked, “Which is your
natural writing hand?”(Peters et al. 2006). Overall
there were sufficient respondents from Europe to
allow a map to be drawn, although for the map
shown in Fig. 3.4 it has been necessary to group
together some countries as sample sizes were other-
wise too small. However, a trend surface analysis,
which is weighted by the sample size in each country,
has no such problems, and from that it is clear that
the highest rates of left-handedness in Europe are
in Britain, the Netherlands, and Belgium. To a first
approximation the rate of left-handedness then
declines as one moves away from those countries, be
it west to Ireland, south-west to France and then
the Iberian peninsula, north-east to Scandinavia, or
east to Germany, Poland, the Baltic, and Russia,
or south-east to the Balkans, Greece, Bulgaria, and
Romania. The reasons for such geographical differen-
ces are not clear, although Medland et al. (2004) have
suggested that countries with a more formal education
system have lower rates of left-handedness than those
with a more informal education system.
Somewhat surprisingly, it is sometimes easier to find
evidence for geographical trends within countries rather
than between them, in part because in national surveys
the same surveymethods are used in the same language
for subjects. One of the biggest, and still one of the best,
such studies is that of Ewald Stier (1911), who in 1909
surveyed the soldiers of the German Army. As expected
the overall rate was much lower than in modern Europe,
at about 3.9%, but the real interest comes in the details of
his study, as in the map shown in Fig. 3.5, which shows
how the rate of left-handedness was lowest for those
from Eastern Prussia, and highest for those from south-
ern Germany, around Stuttgart, where there were over
twice as many left-handers as in the East. Comparing
Fig. 3.5 with Fig. 3.4 suggests that many of the same
trends can still be found today, with higher rates in
Germany than in Poland and the Baltic States, and
higher rates still towards the Swiss border. Other studies
finding differences within countries are rare, but men-
tion should be made of the study of Olivier (1978) in
France, left-handedness being most frequent in the
north, and of lowest frequency in Brittany and the
Massif central, in Italy of Viggiano et al. (2001), where
left-handedness was more frequent in the north of the
country than the south (see also Salmaso & Longoni,
1983), and in Britain, where Leask and Beaton (2007)
showed that within mainland Britain, left-handedness is
less common in Scotland and Wales than in England (a
trend that perhaps is hinted at in Fig. 3.4, where Ireland
has a lower rate of left-handedness than the United
Kingdom).
Another example of a large national survey finding
geographical differences is the Gilbert and Wysocki
Chapter 3: The history and geography of human handedness 45
(1992) study, where the zip code for each US respond-
ent was recorded in the database, but no further anal-
yses were ever carried out on those data. However the
Gilbert and Wysocki data reveal some fascinating
trends, which are both geographical and historical
(McManus & Wysocki, in press). Zip codes for each
respondent can readily be translated into latitudes
and longitudes, and handedness can then be mapped.
Figure 3.6 shows the percentage of left-handers in
White Americans born in 1950 and afterwards in each
of the contiguous states of mainland United States.
Even at this level of spatial resolution it can be seen
that the highest rates of left-handedness are in the
north-east, in Maine, Vermont, Massachusetts, and
Connecticut, whereas the lowest rates are in the
mid-West, in Wyoming and North Dakota. More
detailed mapping suggests that left-handers are also
more frequent in the north-east of the USA, as well as
in Florida, and around the west coast cities of San
Francisco, Portland, and Seattle. The causes of these
differences are complex, but of particular interest is that
as one looks at those born a generation and then two
generations earlier, the geographical patterns shift,
with left-handers then being more common in the
agricultural areas of the United States, such as the
mid-west and the south. The implication is that there
may be differential migration of left-handers.
Finally, it should be mentioned that there must always
be a worry about whether there are response biases in
surveys, particularly those carried out using magazine
readers or internet browsers. To respond to the BBC
internet survey a respondent must have a computer,
must understand English well, and must be aware of
the survey, all of which may make biases possible.
Having said that, similar trends are apparent to those
in Stier’s (1911) study, which used a conscripted sample,
and was entirely in the subjects’native language, thereby
providing a validation in principle of the method.
8.60
10.42
10.51
9.63
9.83
9.1
12.24
11.61
11.65
11.15
11.00
13.10
9.19
8.29
9.03
12.18
7.81
6.09
13.23
10.19
1–2500
700–999
400–699
>2500
N
0.89 > SE > 0.69
1.09 > SE > 0.94
1.41 > SE > 0.92
0.44 (Eire) 0.11 (UK)
SE
Figure 3.4 Rates of left-handedness in different European countries, based on data in the BBC internet survey (McManus &
Peters, in press). Where sample sizes of contiguous countries are relatively low the countries are merged together, indicated by the
gray boxes overlapping borders (e.g., Spain and Portugal were grouped together). The rate of left-handedness is shown as a
percentage. The sample size and the approximate standard errors are shown by different sizes of numbers, the key being provided
at the top left.
46 Section 1: Asymmetry, handedness and language lateralization
Ethnicity and handedness
Analysis of handedness by ethnicity has been left until
last, since in the modern world, ethnicity, which in
some sense expresses the distant geographical origin
of individuals, perhaps many generations previously,
inevitably incorporates a historical component accord-
ing to when an individual’s family or ancestors
migrated from one geographical region to another.
Few studies have assessed ethnic differences in hand-
edness, and the two sets of data presented here, one
from the UK and the other from the USA, have both
been prepared specially for this chapter.
Singh and Bryden (1994) showed that the rate of
left-handedness was lower in the Indian sub-continent
than in the West. A classic epidemiological method for
distinguishing the effects of genes and culture is to
observe migrants between two countries which differ
in some characteristic. If migrants become like the
society to which they have migrated then socio-
cultural factors are probably responsible for the
difference, whereas if the difference remains in
the migrants then genes are probably responsible. The
method can be used to look at handedness in appli-
cants for medical education in the UK, considering
only those who are either White or from the Indian
sub-continent (Table 3.1; for further details of these stud-
ies see McManus et al., 1995; McManus, Richards &
Maitlis, 1989). The odds ratio for the difference between
White and Asian (Indian sub-continent) applicants is
Munster
VII
X
Hannover
Coblenz
VIII
Cassel
Kortsruhe
Prolz
Bayr.Ak
Stuttgarh
Strassburg
Altona
Magdeburg
Nurnberg
Munchen
bis
25,0
25,I
bis
30,0
30,I
bis
35,0
35,I
bis
40,0
40,I
bis
45,0
45,I
bis
50,0
50,I
bis
55,0
uber
55,0
seipeig Dresden
II. Bayr.AK
I. Bayr.AK
III. Bayr.AK
Berlin
Stettin
Danzig
Konigsberg
Posen
Bresiau
III
II XVII
I
V
VI
Frank
furk
Metz
XI
Wurzburg
IX
IV
XIX
XII
XVIII
XVI
XIII
XIV
XV
II
Figure 3.5 The rate of left-handedness in German soldiers in 1909 (plotted as left-handers per thousand) in relation to the
area in which they were recruited (Stier, 1911).
Chapter 3: The history and geography of human handedness 47
1.513×, which is broadly similar to that observed in
Singh and Bryden’s (1994) comparison of Canada and
India. Most importantly, though, there is no difference
between the Asian applicants born in the UK and those
born in the Indian sub-continent (and presumably
reared outside of the UK for at least their early
childhood), which suggests that socio-cultural factors
are relatively unimportant in the origin of ethnic differ-
ences in handedness, and implies instead that genes
may be more important in determining differences.
Ethnicity can also be looked at in the very large
Gilbert and Wysocki (1992) study. Although 97% of
9.7%
9.8%
10.5%
10.5%
10.7%
10.8%
10.3%
10.5%
10.8%
10.8%
10.8%
11.0%
11.0%
11.0%
11.1%
11.1%
11.2%
11.3%
11.4%
11.5%
11.5%
11.5%
11.6%
11.6%
11.6%
11.6%
11.7%
11.7%
11.8%
11.9%
11.9%
12.0%
12.0%
12.1%
12.1%
12.2%
12.3%
12.3%
12.6%
12.7%
12.8%
12.8%
13.3%
13.4%
13.5%
13.7%
13.2%
12.3%
11.4%
Figure 3.6 Rate of left-handedness of White respondents, born from 1950 onwards, in each of the contiguous states of the USA,
based on the data of Gilbert and Wysocki (1992); for further details see McManus and Wysocki (in press). Sample sizes vary from
1197 for the District of Columbia to 52 081 for California, with a mean of 8427, median of 5267, and inter-quartile range of
2879–11 939. The standard error for a state of median size is about 0.4%.
Table 3.1 Handedness of 4902 applicants to UK medical schools for admission in 1986 and 1991, comparing
self-classified Indian sub-continent applicants with White applicants, with non-White applicants divided into those
born in the UK and those not born in the UK. Logistic regression showed an overall effect of being male (OR = 1.387,
p< 0.001), and a highly significant effect of being White (OR = 1.513, p< 0.001), but no significant effect of being born in
the UK (OR = 1.218, p= 0.117). Restricting the analysis to those of Asian origin, there was still a significant effect of
being male (OR = 1.558, p= 0.017), but no effect of being born in the UK (OR = 1.273, p= 0.182). Analyses comparing
the 1986 and 1991 cohorts (not shown here) showed no significant differences.
Ethnic origin Males Females Total
White 13.0% (302/2331) 9.6% (248/2581) 11.2% (550/4902)
Indian sub-continent 9.2% (92/995) 6.1% (47/769) 7.9% (139/1764)
Born in the UK 10.7% (57/534) 6.0% (26/430) 8.6% (83/964)
Not born in the UK 7.6% (35/461) 6.2% (21/339) 7.0% (56/800)
48 Section 1: Asymmetry, handedness and language lateralization
the US respondents in the Gilbert and Wysocki survey
were White, the vast sample size meant there were still
sufficient non-White respondents in the USA to allow
an analysis by ethnicity and year of birth. Considering
only those born from 1910 onwards, numbers being
very small before that, there were 8387 respondents
describing themselves as Black, 10 080 as Asian (pre-
sumably mostly from the Far East), 2513 as American
Indian, and 12 049 as Hispanic, numbers that are larger
than even most of the largest other studies of handedness.
Figure 3.7 shows the rate of left-handedness in the five
ethnic groups in relation to year of birth. The Black and
American Indian groups show similar historical changes
to the White group, whereas both the Asian and
Hispanic groups show lower rates of left-handedness
overall, and also a lower rate of increase in the rate of
left-handedness than do the other groups. The lower
rate of left-handedness in the Asian groups is compat-
ible with other studies suggesting lower rates of
left-handedness in China, Japan, and the Indian
1910 1920 1930 1940 1950 1960 1970 1980
0
2
4
6
8
10
12
14
16
Hispanic
White
Black
American Indian
Asian
Decade of birth
Percentage of left-handers (±1 SE)
Figure 3.7 Left-handedness rates (± 1 SE) in US respondents from different ethnic groups in the Gilbert and Wysocki study,
averaged across sex. Groups are broken down by birth decade (1910–19, 1920–29, etc.) and plotted at the decadal mid-points,
with some groups moved slightly to the left or right to prevent standard error bars overlapping. The solid black line with open
circles is for White respondents, and standard errors are smaller than the size of the symbol. Ethnic groups are shown as Black
(○−−−○), American Indian ( Δ−−−Δ), Hispanic ( ▾−−−▾) , and Asian (▴−−−▴). Statistical analysis used hierarchical logistic
regression. At the first step, effects of year of birth and sex were entered, the non-linear age trend being taken into account by a
quintic polynomial. At the next step, ethnicity showed a highly significant effect (Wald chi-square = 374.7, 4 d.f., p< 0.001), with only
Asians and Hispanics showing significantly lower rates of handedness from the White reference group (p< 0.001 in each case).
Ethnicity by sex interactions were tested at the next step, but were not significant (Wald chi-square =3.508, 4 d.f., p= 0.477). Finally,
the interaction of linear trend of year of birth by ethnicity was tested, and was highly significant (Wald chi-square = 53.43, 4 d.f.,
p< 0.001), with only the Asian and Hispanic sub-groups showing a significantly lower slope than in the White reference group
(p< 0.001 in each case).
Chapter 3: The history and geography of human handedness 49
sub-continent (Iwasaki, 2000; Teng et al. 1976), and
the lower rate of left-handedness in the Hispanics is
similar to that found in the Iberian peninsula in the
European data (see Fig. 3.3), and the effects are also
similar to those of the large study of Halpern et al.
(1998), where among US medical school applicants,
left-handedness was reported in 13.1% of 92 523
Whites, 10.7% of 11 778 Blacks, 10.5% of 6171
Hispanics, 9.2% of 9055 Indian sub-continent appli-
cants, 6.3% of 3533 Vietnamese, 5.4% of 4087 Koreans,
and 5.3% of 7413 Chinese applicants. A striking feature
also of Fig. 3.7 is the excess left-handedness in males
being similar in all ethnic groups, again suggesting
some stable and constant mechanism maintaining sex
differences (and hence probably not cultural or social
pressures against women, as has sometimes been
implied).
Explaining geographical and historical
differences in the rate of left-handedness
The analyses of this paper have so far been mainly
descriptive, but provide a clear demonstration that
rates of left-handedness vary between different coun-
tries. Singh and Bryden’s (1994) comparisons of India
and Canada provide compelling evidence of differences
between countries in two different continents, and
other studies have shown differences across the con-
tinent of Europe and across the states of the USA. There
are differences between ethnic groups within the USA,
Asians and Hispanics having lower rates of left-
handedness, and there are also large historical shifts
in the rate of left-handedness across the twentieth cen-
tury, those historical shifts being paralleled within the
separate ethnic groups that comprise the USA. The
challenge is to explain the origin of these differences:
differences that are present in both space and time.
Explaining geographical differences
Most explanations in biology distinguish nature and
nurture, which to a large extent can be conceptualized
as genes and environment. When populations differ in
their rates of left-handedness then the most important
question concerns whether the differences are genetic
or environmental in origin. Distinguishing such explan-
ations were key questions for Phil Bryden during the
final years of his life, particularly after he had collected
his data showing large and clear differences between
Canada and India. However, the obvious theoretical
problem is that a low rate of left-handedness in India
can result either from social pressure, which results in
left-handers being forced, overtly or covertly, to behave
as right-handers, or from differences in gene frequency
between Canada and India, and prevalence data alone
cannot distinguish between genetic and social causes.
The key insight, however, which Phil Bryden and I
developed together in what as it happened were the
last months of his life, is that the effects of genes and
social pressure can be distinguished if each is modeled
separately, and family data are available.
Euchiria, Hipressia, and Lowgenia
In a popular book on handedness (McManus, 2002), I
illustrated the separate effects of genes and culture by
describing three mythical countries, which I named
Euchiria, Hipressia, and Lowgenia. The model is
based around the McManus genetic model of handed-
ness (McManus, 1985; McManus & Bryden, 1992),
which needs to be briefly described, although it seems
likely that any broadly similar genetic model will show
similar effects.
1
The model suggests that at a single
genetic locus there are two alleles, named D(for dex-
tral) and C(for chance). One hundred percent of DD
homozygotes are right-handed, whereas CC homozy-
gotes have a 50:50 chance of being right- or left-
handed. The alleles are additive in the heterozygote,
so that 25% of DC individuals are left-handed, and the
remaining 75% are right-handed. The model explains
not only how handedness runs in families, but also why
as a result of the random effects of the Callele about
one in five monozygotic twin pairs is discordant for
handedness. Finally, by assuming that the same alleles
determine handedness and language dominance, the
1
An exception may be the Annett model, which as well as a
parameters to describe the frequency of the RS-gene, also has a
threshold parameter that can be adjusted separately in each
population, and which therefore is confounded with rate of
handedness.
50 Section 1: Asymmetry, handedness and language lateralization
model readily explains why about 5% of right-
handers and 35% of left-handers are right hemisphere
dominant for language.
Euchiria is a country in which only genes determine
handedness, and because the rate of left-handedness,
p(L), is set at exactly 10%, the calculations of the
McManus model are particularly easy, because the
frequency of the Callele, p(C) is double that of p(L),
and hence is 20%. The top of the first column of
Table 3.2 shows the frequency of the Callele, the
probability of each of the three genotypes being
left-handed, and the resulting rate of left-handedness,
which is 10%. For convenience, and because two
left-handed parents is a relatively rare combination,
families are divided into those for whom both parents
are right-handed (R×R) and those in whom at least
one parent is left-handed (L
par
). When at least one
parent is left-handed, the proportion of left-handers
in the offspring, p(L|L
par
), is 19.5%, compared with
only 7.8% when both parents are right-handed, p(L|
R×R). Calculating a conventional odds ratio, as 19.5 ×
(100 −7.8)/(7.8 × (100 −19.5), shows a child with at least
one left-handed parent is about 2.87 times more likely to
be left-handed itself.
Lowgenia is similar in many ways to Euchiria, except
that the frequency of the Callele is lower, being exactly
half that found in Euchiria, so that p(C) is 10%, and the
unsurprising consequence is that the rate of
left-handedness is also half that found in Euchiria, p(L)
being 5%. What is rather more counter-intuitive, at
least for those not used to genetic calculations, is that
the odds ratio for the effect of having at least one
left-handed parent is higher in Lowgenia than in
Euchiria, being 4.79× in Lowgenia, rather than the
2.87× found in Euchiria. A reduction in gene frequency
therefore increases the odds ratio of the child of a
left-handed parent being left-handed.
Hipressia is more complicated, because not only
genes but also social pressure affect the rate of
left-handedness, both of which need to be modeled.
Hipressians do not like left-handers and do their best
to make them indistinguishable from right-handers,
but human resilience being what it is, they are only
successful in half the cases. The gene frequency in
Hipressia, p(C), is the same as in Euchiria, but instead
of a half of CC individuals and a quarter of DC indi-
viduals becoming left-handed as they would in
Euchiria, social pressure against left-handers in
Hipressia means that only a quarter of CC individuals
and an eighth of DC individuals become left-handed
(or putting it more precisely, a quarter of CC individ-
uals and an eighth of DC individuals, who would have
become left-handed in Euchiria, instead become
right-handed in Hipressia because of social pressure,
making them what geneticists call “phenocopy right-
handers”). The unsurprising result, once again, is that
the overall rate of left-handedness in Hipressia is
exactly half that found in Euchiria, being 5%. That
rate of 5% is exactly the same as the rate found in
Lowgenia, showing how two entirely different causal
mechanisms result in the same overall rate of left-
handedness. However, and it is a key point, the pattern
of left-handedness in Hipressian families is very differ-
ent from that found in Lowgenia, the odds of a
Hipressian being left-handed when they have a
left-handed parent being 2.34× higher than if both
parents are right-handed, compared with 4.79× in
Lowgenia. The Hipressian odds ratio is therefore lower
than in Lowgenia (and indeed is also lower than in
Euchiria). The key theoretical conclusion is that gene
frequency differences and social pressure can be distin-
guished by looking at odds ratios in families.
Table 3.2 Familial patterns of handedness in the
mythical countries of Euchiria, Lowgenia, and Hipressia
(see text for details). Note that p(L|DD) indicates the
conditional probability of being left-handed, given that
an individual has the DD genotype, etc.). L
par
refers to
families in which at least one parent is left-handed.
Euchiria Lowgenia Hipressia
p(C) 0.2 0.1 0.2
p(L|DD) 000
p(L|DC) 0.25 0.25 0.125
p(L|CC) 0.50 0.50 0.250
p(L) 0.100 0.050 0.050
p(L|R × R) 0.078 0.038 0.045
p(L|L
par
)0.195 0.160 0.099
Odds ratio 2.87× 4.79× 2.34×
Chapter 3: The history and geography of human handedness 51
The model that Phil Bryden and I had developed was
quickly tested, because Bryden not only had data on the
rate of left-handedness in Canada and India, but had
already carried out a preliminary analysis of how hand-
edness ran in families in the two countries (Brydenet al.
1995). In Canada, where the rate of left-handedness
was 9.8%, the odds ratio was 2.09×, whereas in India,
where the rate of left-handedness was 5.2%, the odds
ratio was 3.07×. The implication was clear: the majority
of the difference between Canada and India must be
due to differences in gene frequency rather than due to
differences in social pressure. Subsequently, Bryden and
I collaborated with Taha Amir in the United Arab
Emirates (UAE), and Yokahida Ida in Japan, and we
also put together larger Western samples (mainly
Canada and the UK), and Indian samples. Of 17 850,
14 924, 4485, and 656 offspring in the West, UAE, India,
and Japan, for whom p(L) was 11.5%, 7.5%, 5.8%, and
4.0% respectively, the odds ratios for the effect of hav-
ing a left-handed parent were 2.11×, 2.23×, 3.18× and
3.57× respectively, which is the pattern expected from
gene frequency differences. Geographical differences
in the modern world seem therefore to be primarily
genetic in origin, rather than due to differences in
social pressure (or what I will refer to subsequently as
“direct social pressure”).
Explaining historical differences
If geographical differences in rates of left-handedness
can be explained in terms of differences in gene fre-
quency, what about historical differences? “The past is
a foreign country”, as L. P. Hartley said at the beginning
of The Go-Between (albeit often being misquoted as
“the past is another country”). If so, then the same
methods that distinguish the causes of geographical
differences should also distinguish the causes of histor-
ical differences. Fortunately, a number of family studies
of handedness in Western countries have been pub-
lished over the past century, the earliest being that of
Ramaley (1913), who described data collected in a
group of undergraduate students (the probands), who
therefore would have been born in about 1888. In 1992,
Phil Bryden and I (McManus & Bryden, 1992) had
already reviewed 25 such studies, and had broken
them down into three groups, those for whom the
probands were born between 1880 and 1939, 1940
and 1954, and 1955 and 1979, the rates of
left-handedness in the offspring being 7.28%, 10.83%,
and 13.25% (whereas, the parents, being born a gener-
ation earlier, had rates of left-handedness of 4.44%,
6.11%, and 9.34%). The odds ratios for the effect of
having a left-handed parent were 3.29×, 2.08×, and
1.64× in the three groups respectively. Just as with the
geographical data, when the rate of left-handedness is
lower, so the odds ratio is higher, implying that the
historical differences also reflect differences in gene
frequency. That suggestion was also strongly supported
by a reanalysis of data from the huge study of the
German Army by Stier (1911), the conscripts for
whom would have been born in about 1890, and of
whom 3.87% were left-handed, a lower value than any
of the family studies we had analyzed. Stier reported
the number of left- and right-handers with left-handed
relatives, and by making some reasonable assump-
tions, one can estimate the odds ratio for the effect of
having a left-handed parent as being about 5.2×, a
higher value than any of the odds ratios in the other
familial studies. Once again, Stier’s data suggest that
historical differences reflect genetic differences rather
than effects of social pressure.
Social pressure can take many forms, and it is useful
to distinguish between direct and indirect social pres-
sure. Direct social pressure involves left-handed indi-
viduals being made to write with their right hand, as
seems to have happened in some Victorian schools (see,
e.g., Ireland, 1880), and has occurred in many other
forms around the world to prevent left-handers using
their left hands (see McManus, 2002). However, direct
social pressure of this sort only alters the phenotype,
not the genotype, and the individuals still carry the
genes that made them originally left-handed, and if
transmitted those genes would allow those individuals’
offspring to become left-handed. Indirect social pres-
sure is much more subtle, and does not directly alter
the phenotype of the left-hander, but instead acts to
make left-handers stigmatized, ostracized, and taboo,
so that they find it harder to have offspring. The result is
that their genes are less likely to be passed on, and
hence the frequency of the genes responsible for
52 Section 1: Asymmetry, handedness and language lateralization
left-handedness falls, and left-handedness becomes less
common in the next generation. To see how this might
happen one must consider the very different social world
of relatively small nineteenth-century communities,
where most people knew one another, transport was
less good, most people married people living less than
30 kilometers away, marriage was relatively early, as also
was first childbirth, so that families were large, often with
eight or ten children, child-bearing only ceasing at men-
opause. In such a world, any subtle denigration, mock-
ery, or stigmatization of the left-handed, perhaps for
clumsiness or awkwardness at writing or technical skills,
or indeed for mere difference itself, might result in mar-
riage and hence childbirth being delayed by five or ten
years, so that the number of offspring would be reduced.
The consequence would be a fall in the number of C
alleles and hence in the rateof left-handedness. Indirect
social pressure, although less brutal than direct social
pressure, could be of far greater consequence in its
eventual effects.
If the theory of indirect social pressure is correct,
then there is a clear prediction: left-handers at the
end of the nineteenth century should have had fewer
children than right-handers. Fortunately that predic-
tion can not only be readily tested, but the data has
already been presented in our review of the genetics of
handedness (McManus & Bryden, 1992). Family stud-
ies typically include all children, and hence if the num-
ber of parents is known as well as the number of
offspring, then the mean number of offspring can be
calculated. Table 3.3 shows that while at the end of the
twentieth century, right- and left-handed parents had
similar numbers of children, despite parents around
the turn of the century in general having more children
than modern parents, left-handers had relatively fewer
children, two left-handed parents having only 2.32 chil-
dren, compared with 2.69 children when one parent
was left-handed, and 3.10 children when both parents
were right-handed. Two right-handed parents there-
fore had 34% more children than two left-handed
parents. It is therefore at least possible that historical
shifts in the rate of left-handedness are driven by differ-
ences in fertility (and the ultimate test of any evolution-
ary theories concerns whether groups of individuals
differ in the numbers of offspring).
The consequences of historical and
geographical differences in left-handedness
Were the rate of left-handedness to vary, either histor-
ically or geographically, and particularly if that varia-
tion is due to differences in gene frequency, what
consequences does that have for neuropsychology
and neuropsychiatry? The answer depends in part on
the nature of the genetic system underlying handed-
ness and cerebral dominance, and for obvious reasons I
will consider the McManus model, which suggests that
25% of DC individuals and 50% of CC individuals are
left-handed. More generally (McManus, 1984; McManus,
1985) the model says that 25% of DC individuals and
50% of CC individuals, but no DD individuals, will
have atypically directed lateralization for any character
controlled by the gene. A crucial corollary is that the
chance processes for each character will be statistically
independent.
If there is a probability p
G
that any individual modu-
lar character will be atypically organized (such as,
left-handedness or right-sided language) in a particular
genotype, G, and if we consider two modular traits,
such as handedness and language, then (1 −p
G
)
2
will
have the typical phenotype (the one described in neu-
ropsychology textbooks, which for handedness and
language is right-handedness and left-sided language),
2.p
G
.(1 −p
G
) will have one atypical trait, and p
G2
will
have both traits anomalously organized (in this case,
left-handers with right-sided language). For DD,DC,
Table 3.3 The average number of offspring in relation
to parental handedness, in familial studies of
handedness carried out in different periods, classified
by the birth year of probands (from McManus &
Bryden, 1992).
Birth
year of
probands
Number
of studies
Number
of parental
pairs
Parental handedness
R×R R×L L×L
1880–1939 5 4180 3.10 2.69 2.32
1940–54 5 3800 3.17 3.05 3.00
1955–80 6 7323 2.49 2.60 2.57
Chapter 3: The history and geography of human handedness 53
and CC individuals, p
G
is 0, 0.25, and 0.5 respectively.
However, DD individuals are far more frequent in the
population than DC who are more frequent than CC
individuals. Combining all the numbers, then it is easy
to show that if the rate of left-handedness is 10%, then
7.8% of right-handers and 30.0% of left-handers will
have language in the right hemisphere, which corre-
sponds broadly with the data.
There may, however, be multiple modular traits con-
trolled by the Dand Calleles, with perhaps several
separate modular traits for aspects of spoken and writ-
ten language, several modular traits for aspects of
visuo spatial and facial processing, and so on. If there
are nmodular traits, then (1 −p
G
)
n
individuals will have
the textbook pattern with no anomalies, and 1 −(1 −p
G
)
n
will have at least one anomaly (such as a right-sided
component of language, or a left-sided component of
visuo spatial processing). The number of modules is
not at present known, but Table 3.4 calculates the
percentage of individuals with anomalous organization
in relation to the number of modules and the rate of
left-handedness in the population. The basic finding is
very simple: irrespective of the number of modular
traits controlled by the Callele, the proportion of anom-
alous traits rises approximately linearly with the rate of
left-handedness. If it is the case that dyslexia, stuttering,
autism, schizophrenia, or other conditions are related to
atypical cerebral lateralization, and hence to the pres-
ence of a Callele, then the rate of those conditions
should change geographically or historically in parallel
with the rate of left-handedness.
2
In particular, in the
West there may well have been a three- or fourfold
increase in the rate of those conditions since Victorian
times, and in other cultures the rate might well be rising
as left-handedness increases in frequency. That may
help to explain how conditions that we now think of
as common, were rare and difficult to describe and
characterize in the nineteenth century. However, and
it is relevant in the context of current speculations
about a rising rate of autism, the rate of cerebral dom-
inance related anomalies should be relatively constant
for those born in the West after about 1950. Whether or
not there are historical and geographical variations in
neuropsychiatric conditions remains to be seen; col-
lecting adequate evidence to assess the idea will not be
easy, but the hypotheses relating their rate to handed-
ness and cerebral dominance differences are testable,
and have interesting implications for interpreting dif-
ferences in neuropsychiatric disease prevalence.
ACKNOWLEDGMENTS
I am very grateful to Chuck Wysocki and Avery Gilbert
for providing me with raw data from their large study of
handedness, and to Michael Peters and Stian Reimers
for their collaboration in studying the data from the
BBC internet study.
Table 3.4 The effect of the rate of left-handedness on
the percentage of individuals with atypical cerebral
organization (e.g., crossed cerebral dominance,
“anomalous” dominance).
Number of modular traits
Rate of left-
handedness 1 23510
“Very
large”
2.5% 2.5% 4.3% 5.7% 7.5% 9.2% 9.8%
5% 5.0% 8.6% 11.3% 14.7% 18.0% 19.0%
7.5% 7.5% 12.8% 16.7% 21.6% 26.3% 27.8%
10% 10.0% 17.0% 22.0% 28.3% 34.2% 36.0%
12.5% 12.5% 21.1% 27.1% 34.7% 41.6% 43.8%
Note: when the number of modular traits is very large (and is
effectively infinite), then all DC and CC individuals will show
at least one anomalous trait. If the rate of left-handedness is
p(L), then the frequency of the Callele is 2.p(L), the frequency
of the Dallele is 1 −2.p(L), the frequency of DD individuals is
[1 −2.p(L)]
2
, and hence the combined frequency of DC and
CC individuals, which is the proportion of individuals with
anomalies, is 1 −[1 −2.p(L)]
2
.
2
It should also be said that small numbers of anomalies may well
be beneficial, while large numbers of anomalies are deleterious.
Elsewhere in my “theory of random cerebral variation”
(McManus, 2002) I have argued that DC individuals in particular
are more likely to have single anomalies that might result in
beneficial consequences, perhaps in the form of special talents
for particular tasks that involve unusual interactions between
modules.
54 Section 1: Asymmetry, handedness and language lateralization
REFERENCES
Aggleton, J. P., Bland, J. M., Kentridge, R. W. & Neave, N. J.
(1994). Handedness and longevity: archival study of crick-
eters. British Medical Journal,309, 1681–4.
Ambrose, S. H. (2001). Palaeolithic technology and human
evolution. Science,291, 1748–53.
Annett, M. & Annett, J. (1991). Handedness for eating in goril-
las. Cortex,27, 269–75.
Batigne, R. & Bellinger, L. (1953). The significance and tech-
nical analysis of ancient textiles as historical documents.
Proceedings of the American Philosophical Society,97,
670–80.
Bertrand, P.-M. (2001). Histoire des Gauchers. Paris: Imago.
Brackenridge, C. J. (1981). Secular variation in handedness
over ninety years. Neuropsychologia,19, 459–62.
Bryden, M. P., Ardila, A. & Ardila, O. (1993). Handedness in
native Amazonians. Neuropsychologia,31, 301–8.
Bryden, M. P., Singh, M. & Rogers, T.T. (1995). Heritability for
degree and direction of human hand preference. Society for
Neuroscience Abstracts,21, 200.
Burt, C. (1937). The Backward Child. London: University of
London Press.
Cahen, D., Keeley, L. H. & Van Noten, F. L. (1979). Stone tools,
toolkits, and human behavior in prehistory. Current
Anthropology,20, 661–83.
Carrothers, G. E. (1947). Left-handedness among school
pupils. American School Board Journal,114,17–19.
Cavalli-Sforza, L. L. & Bodmer, W. F. (1971). The Genetics of
Human Populations. San Francisco: W.H. Freeman.
Cavalli-Sforza, L. L., Menozzi, P. & Piazza, A. (1994). The
History and Geography of Human Genes. Princeton, NJ:
Princeton University Press.
Connolly, K. & Bishop, D. V. M. (1992). The measurement of
handedness: a cross-cultural comparison of samples from
England and Papua New Guinea. Neuropsychologia,30,13–26.
Coren, S. & Halpern, D. F. (1991). Left-handedness: a marker
for decreased survival fitness. Psychological Bulletin,109,
90–106.
Coren, S. & Porac, C. (1977). Fifty centuries of right-
handedness: the historical record. Science,198, 631–2.
Crichton-Browne, J. (1907). Dexterity and the bend sinister.
Proceedings of the Royal Institution of Great Britain,18,
623–52.
Crowfoot, E., Pritchard, F. & Staniland, K. (2001). Textiles and
Clothing c1150–c1450. Woodbridge: Boydell Press.
de Castro, J. M. B., Bromage, T. G. & Jalvo, Y. F. (1988). Buccal
striations on fossil human anterior teeth: evidence of
handedness in the middle and early Upper Pleistocene.
Journal of Human Evolution,17, 403–12.
Delacato, C. H. (1963). The Diagnosis and Treatment of Speech
and Reading Problems. Springfield, IL: C C Thomas.
Ellis, P. J., Marshall, E., Windridge, C., Jones, S. & Ellis, S.J. (1 998).
Left-handedness and premature death. Lancet,351, 1634.
Erlenmeyer-Kimling, L., Hans, S., Ingraham, L. et al. (2005).
Handedness in children of schizophrenic parents: data from
three high-risk studies. Behavior Genetics,35, 351–8.
Faurie, C. & Raymond, M. (2004). Handedness frequency over
more than 10,000 years. Proceedings of the Royal Society of
London, Series B,271, S43–S45.
Finger, S. (1994). Origins of Neuroscience: a History of
Explorations into Brain Function. New York: Oxford
University Press.
Finger, S. & Roe, D. (1999). Does Gustave Dax deserve to be
forgotten? The temporal lobe theory and other contributions
of an overlooked figure in the history of language and cere-
bral dominance. Brain and Language,69,16–30.
Fox, C. L. & Frayer, D. W. (1997). Non-dietary marks in the
anterior dentition of the Krapina Neanderthals. International
Journal of Osteoarchaeology,7,133–49.
Gibbons, B. (1986). The intimate sense of smell. National
Geographic,170, 324–61.
Gilbert, A. N. & Wysocki, C. J. (1987). The Smell Survey results.
National Geographic,172, 514–25.
Gilbert, A. N. & Wysocki, C. J. (1992). Hand preference and age
in the United States. Neuropsychologia,30, 601–8.
Halpern, D. F. & Coren, S. (1988). Do right-handers live longer?
Nature,333, 213.
Halpern, D. F. & Coren, S. (1991). Handedness and life span.
New England Journal of Medicine,324, 998.
Halpern, D. F. & Coren, S. (1993). Left-handedness and life
span: a reply to Harris. Psychological Bulletin,114, 235–41.
Halpern, D. F., Haviland, M. G. & Killian, C. D. (1998).
Handedness and sex differences in intelligence: evidence
from the Medical College Admission Test. Brain and
Cognition,38,87–101.
Harris, L. J. (1991). Cerebral control for speech in
right-handers and left-handers: an analysis of the views of
Paul Broca, his contemporaries, and his successors. Brain
and Language,40,1–50.
Harris, L. J. (1993a). Broca on cerebral control for speech in
right-handers and left-handers: a note on translation and
some further comments. Brain and Language,45, 108–20.
Harris, L. J. (1993b). Do left-handers die sooner than right-
handers? Commentary on Coren and Halpern’s (1991)
“Left-handedness: a marker for decreased survival fitness”.
Psychological Bulletin,114, 203–34.
Chapter 3: The history and geography of human handedness 55
Harris, L. J. (1993c). Reply to Halpern and Coren. Psychological
Bulletin,114, 242–7.
Hopkins, W. D. (2006). Comparative and familial analysis of
handedness in great apes. Psychological Bulletin,132,
538–59.
Hugdahl, K. (1996). Left-handedness and age: comparing writ-
ing/drawing and other manual activities. Laterality,1,
177–83.
Hugdahl, K., Satz, P., Mitrushina, M. & Miller, E. N. (1993).
Left-handedness and old age: do left-handers die earlier?
Neuropsychologia,31, 325–33.
Ida, Y. & Bryden, M. P. (1996). A comparison of hand prefer-
ence in Japan and Canada. Canadian Journal of Experimen-
tal Psychology,50, 234–9.
Ireland, W. W. (1880). Notes on left-handedness. Brain,3, 207–14.
Iwasaki, S. (2000). Age and generation trends in handedness:
an Eastern perspective. In M. K. Mandal, M. B. Bulma n-
Fleming & G. Tiwari, eds., Side Bias: A Neuropsychological
Perspective. Dordrecht: Kluwer, pp. 83–100.
Knecht, S., Deppe, M., Dräger, B. et al. (2000). Language later-
alization in healthy right-handers. Brain,123,74–81.
Leask, S. J. & Beaton, A. A. (2007). Handedness in Great Britain.
Laterality,12, 559–72.
Lombroso, C. (1884). Sul mancinismo e destrismo tattile nei
sani, nei pazzi, nei ciechi e nei sordomuti. Archivi di
Psichiatria, Neuropsichiatria, antropologia criminale e
medicina legale, vol. 5.
Marchant, L. F. & McGrew, W. C. (1991). Laterality of function
in apes: a meta-analysis of methods. Journal of Human
Evolution,21, 425–38.
Marchant, L. F. & McGrew, W. C. (1996). Laterality of limb
function in wild chimpanzees of Gombe National Park:
comprehensive study of spontaneous activities. Journal of
Human Evolution,30, 427–43.
Marks, J. S. & Williamson, D. F. (1991). Left-handedness and
life expectancy. New England Journal of Medicine,325, 1042.
McManus, I. C. (1981). Handedness and birth stress.
Psychological Medicine,11, 485–96.
McManus, I. C. (1984). The genetics of handedness in relation
to language disorder. In F. C. Rose, ed., Advances in
Neurology, vol 42: Progress in Aphasiology, New York:
Raven Press, pp. 125–38.
McManus, I. C. (1985). Handedness, Language Dominance
and Aphasia: a Genetic Model. Psychological Medicine,
Monograph Supplement No. 8, Cambridge University Press.
McManus, I. C. (1991). The inheritance of left-handedness. In
G. R. Bock & J. Marsh, eds., Biological Asymmetry and
Handedness (Ciba Foundation Symposium 162),Chichester:
Wiley, pp. 251–81.
McManus, I. C. (1999). Handedness, cerebral latera lization
and the evolution of language. In M. C. Corballis & S. E. G.
Lea, eds., The Descent of Mind: Psychological Perspectives on
Hominid Evolution. Oxford: Oxford University Press,
pp. 194–217.
McManus, I. C. (2002). Right Hand, Left Hand: The Origins of
Asymmetry in Brains, Bodies, Atoms and Cultures. London,
UK/Cambridge, MA: Weidenfeld and Nicolson/Harvard
University Press.
McManus, I. C. (2004). Grappling with the hydra: review of
Handedness and Brain Asymmetry, by Marian Annett.
Cortex,40, 139–41.
McManus, I. C. & Bryden, M. P. (1992). The genetics of hand-
edness, cerebral dominance and lateralization. In I. Rapin &
S. J. Segalowitz, eds., Handbook of Neuropsycholo gy, Volume
6, Section 10: Child neuropsychology (Part 1), Amsterdam:
Elsevier, pp. 115–44.
McManus, I. C. & Bryden, M.P. (1993). Handedness on Tristan
da Cunha: the genetic consequences of social isolation.
International Journal of Psychology,28, 831–43.
McManus, I. C. & Hartigan, A. (2007). Declining left-
handedness in Victorian England seen in the films of
Mitchell and Kenyon. Current Biology,17, R793–4.
McManus, I.C., Moore, J., Freegard, M. & Rawles, R. (in pressa).
Science in the making: right hand,left hand: III: the incidence
of left-handedness. Laterality.
McManus, I. C. & Peters, M. (in press). Handedness in Europe:
analysis of data from the BBC internet study.
McManus, I.C., Rawles, R., Moore, J. & Freegard, M. (in press b).
Science in the making: Right Hand, Left Hand: I: a BBC tele-
vision programme broadcast in 1953. Laterality.
McManus, I. C., Richards, P. & Maitlis, S. L. (1989). Prospectiv e
study of the disadvantage of people from ethnic minority
groups applying to medical schools in the United Kingdom.
British Medical Journal,298, 723–6.
McManus, I. C., Richards, P., Winder, B. C., Sproston, K. A. &
Styles, V. (1995). Medical school applicants from ethnic
minorities: identifying if and when they are disadvantaged.
British Medical Journal,310, 496–500.
McManus, I. C. & Wysocki, C. J. (in press). Variation in handed-
ness across the United States.
Medland, S. E., Perelle, I. B., De Monte, V. & Ehrman, L. (2004).
Effects of culture, sex, and age on the distribution of handed-
ness: an evaluation of the sensitivity of three measures of
handedness. Laterality,9, 287–97.
Minar, C. J. (2001). Motor skills and the learning process:
the conservation of cordage final twist direction in commun-
ities of practice. Journal of Anthropological Research,57,
381–405.
56 Section 1: Asymmetry, handedness and language lateralization
Ogle, W. (1871). On dextral pre-eminence. Medical –
Chirurgical Transactions (Transactions of the Royal Medical
and Chirurgical Society of London),54, 279–301.
Olivier, G. (1978). Anthropometric data on left-handed.
Biométrie Humaine,13,13–22.
Parson, B. S. (1924). Lefthandedness: a New Interpretation.
New York: Macmillan.
Perelle, I. B. & Ehrman, L. (1994). An international study of
human handedness: the data. Behavior Genetics,24, 217–27.
Peters, M., Reimers, S. & Manning, J. T. (2006). Hand prefer-
ence for writing and associations with selected demographic
and behavioral variables in 255,100 subjects: the BBC inter-
net study. Brain and Cognition,62, 177–89.
Phillipson, L. (1997). Edge modification as an indicator of
function and handedness of Acheulian handaxes from
Kariandusi, Kenya. Lithic Technology,22, 171–83.
Pujol, J., Deus, J., Losilla, J. M. & Capdevila, A. (1999). Cerebral
lateralization of language in normal left-handed people
studied by functional MRI. Neurology,52, 1038–43.
Pye-Smith, P. H. (1871). On left-handedness. Guy’s Hospital
Reports,16, 141–6.
Ramaley, F. (1913). Inheritance of left-handedness. American
Naturalist,47, 730–9.
Raymond, M. & Pontier, D. (2004). Is there geographical var-
iation in human handedness? Laterality,9,35–51.
Reimers, S. (2007). The BBC Internet Study: general method-
ology. Archives of Sexual Behaviour,36, 147–61.
Roberts, M. B. & Parfitt, S. A. (1999). Boxgrove: A Middle
Pleistocene Hominid Site at Eartham Quarry, Boxgrove,
West Sussex. London: English Heritage.
Ross, G., Lipper, E. & Auld, P. A. M. (1992) . Hand preference,
prematurity and developmental outcome at school age.
Neuropsychologia,30, 483–94.
Sadler, N. (1997). A sinister way of life: a search for left-handed
material culture. In S. M. Pearce, ed., Experiencing Material
Culture in the Western World. London: Leicester University
Press, pp. 140–53.
Salmaso, D. & Longoni, A. M. (1983). Hand preference in an
Italian sample. Perceptual and Motor Skills,57, 1039–42.
Schäfer, M. (1911). Die Linkshänder in den Berliner
Gemeindeschulen. Berliner klinische Wochenschrift,48, 295.
Singh, M. & Bryden, M. P. (1994). The factor structure of hand-
edness in India. International Journal of Neuroscience,74,
33–43.
Singh, M., Manjary, M. & Dellatolas, G. (2001). Lateral prefer-
ences among Indian school children. Cortex,37, 231–41.
Spenneman, D. R. (1984). Handedness data on the European
Neolithic. Neuropsychologia,22, 613–15.
Spindler, K. (1994). The Man in the Ice. London: Weidenfeld
and Nicolson.
Steele, J. (2000). Handedness in past human populations: skel-
etal markers. Laterality,5, 193–220.
Steele, J. & Mays, S. (1995). Handedness and directional asym-
metry in the long bones of the human upper limb.
International Journal of Osteoarchaeology,5,39–49.
Steele, J. & Uomini, N. (2005). Humans, tools and handedness.
In V. Roux & B. Bril, eds., Stone Knapping: the Necessary
Conditions for a Uniquely Hominin Behaviour. Cambridge:
McDonald Institute, pp. 217–39.
Stephens, W. B. (1990). Literacy in Scotland, England and
Wales, 1500–1900. History of Education Quarterly,30,
545–71.
Stier, E. (1911). Untersuchungen über Linkshändigkeit und die
funktionellen Differenzen der Hirnhälften. Nebst einem
Anhang, “Über Linkshändigkeit in der deutschen Armee”.
Jena: Gustav Fischer.
Stone, L. (1969). Literacy and education in England 1640–1900.
Past and Present,42,69–139.
Teng, E. L., Lee, P. & Chang, P. C. (1976). Handedness in a
Chinese population: biological, social and pathological fac-
tors. Science,193, 1148–50.
Toth, N. (1985). Archaeological evidence for preferential right
handedness in the lower and middle Pleistocene and its
possible implications. Journal of Human Evolution,14,
607–14.
Viggiano, M. P., Borelli, P., Vannucci, M. & Rocchetti, G.
(2001). Hand preference in Italian students. Laterality,6,
283–6.
Watson, T. (1836). An account of some cases of transposition
observed in the human body. London Medical Gazette,18,
393–403.
West, E. G. (1978). Literacy and the Industrial Revolution.
Economic History Review,31, pp. 369–83.
Wolf, P. A., D’Agostino, R. B. & Cobb, J. (1991). Left-handed-
ness and life expectancy. New England Journal of Medicine,
325, 1042.
Wysocki, C. J. & Gilbert, A. N. (1989). National Geographic
Smell Survey: effects of age are heterogenous. In
C. Murph y, W. S. Cain & D. M. Hegsted, eds., Nutrition and
the Chemical Senses in Aging: Recent Advances and Current
Research Needs. New York: New York Academy of Sciences,
pp. 12–28.
Wysocki, C. J., Pierce, J.D. & Gilbert, A. N. (1991). Geographic,
cross-cultural, and individual variation in human olfaction.
In T. V. Getchell et al., eds., Smell and Taste in Health and
Disease, New York: Raven Press, pp. 287–314.
Chapter 3: The history and geography of human handedness 57
