Lose your shoes: Is barefoot better?
Posted by gregdowney on July 26, 2009
In 1984 at the Los Angeles Olympics, the women’s 3000-meter final was marred by controversy when American Mary Decker fell after making contact with Zola Budd, a runner from South Africa who represented Britain (due to the boycott of South African sport).
Although Budd had been setting the pace, she faded to seventh in the end and was booed by the partisan LA audience (Decker would later say that she was inexperienced at running in a pack and, as the trailing runner, was responsible for their contact). Maricica Puica of Romania won the event, and Britain’s Wendy Sly took the silver in a final that was seared into my memory by the televised replays of a stricken Mary Decker, hip injured from her fall, shattered and crying on the infield.
In all of the drama, one of the things that left the greatest impression on me as a high school student and sometime athlete was the simple fact that Zola Budd ran without shoes, an almost unimaginable idea to me at the time. Budd was one of a handful of famous barefoot runners, including Abebe Bikila, the Ethiopian marathoner who won his first Olympic gold in 1960 without shoes, Tegla Loroupe, the Kenyan women’s running legend and multiple world record holder, and Ken Bob Saxton, aka ‘Barefoot Ken Bob,’ a marathoner and guru to the shoeless.
I’ve been thinking about barefoot running for a while, oddly enough since I started writing about bare-knuckle punching in no-holds-barred fighting (or ‘mixed martial arts’ like the Ultimate Fighting Championship in its early days). Barefoot running, even more than bare-knuckle boxing, reveals the ways that very simple technologies, if used consistently enough, become part of the developmental niche of the human body, shaping the way that our bones, muscles, tissues, and nervous system develop.
Although this post is not strictly neuroanthropology, I thought I might share some of what I’m working on, in part because I’m interested to hear any feedback people have. In particular, this will focus on how hard it is to sort out what’s ‘natural’ when activity patterns, incredibly variable, are necessary ingredients in the development of biological systems. But also, as it will become clearer in the post, the ways that our nervous system adapt to different situations, such as having heavily padded feet or being barefoot when we run, illustrates well how even unconscious training is a form of phenotypic, non-genetic, adaptation.
Before I go any further, though, if you have anything to say in response to this, I would love to read it. This is my first attempt to put down some thoughts that will be in a chapter of an upcoming book…
I was sparked to finally put this down and post it by an item in Wired Science: ‘To Run Better, Start by Ditching Your Nikes,’ by Dylan Tweeny. (See below for a number of other recent articles online.) Tweeny writes:
Strong evidence shows that thickly cushioned running shoes have done nothing to prevent injury in the 30-odd years since Nike founder Bill Bowerman invented them, researchers say. Some smaller, earlier studies suggest that running in shoes may increase the risk of ankle sprains, plantar fasciitis and other injuries. Runners who wear cheap running shoes have fewer injuries than those wearing expensive trainers. Meanwhile, injuries plague 20 to 80 percent of regular runners every year.
The article shares quotes by a number of barefoot running advocates who argue strongly that running in minimalist shoes, or unshod, reduces the likelihood of injury: ‘After all,’ Tweeny writes in a discussion of the work of Daniel Lieberman, a professor of human evolutionary biology at Harvard University, ‘we evolved without shoes.’
In the passage, Tweeny refers to a study published in the British Journal of Sports Medicine (Clinghan et al. 2008) that found cheap running shoes correlated with better long-term health outcomes than more expensive footwear. Runners who used more expensive running shoes had a pretty shocking 123% higher rate of injury than those in less expensive shoes (see Robbins and Waked 1997). The Robbins and Waked (1997) study directly focused on the relation between deceptive shoe advertising and the force of barefoot subjects’ footfall when they came down on a surface designed to look like shoe padding. Led to believe that the surface was protecting them, people changed their running style in ways that increased impact.
The rate of injuries among runners, including the relatively consistent injury rate despite ‘improvements’ in shoe technology, make some observers suspicious that shoes might be causing, rather than protecting against, injury, even if the link is indirect through shifts in technique or even the population that can participate. Ross Tucker and Anthony Dugas of The Science of Sport point out that there are, in fact, many possible explanations for changes in injury rates – or changing reasons why rates remain constant – such as the demographic factor that many runners in the 1990s might be in significantly worse physical condition than runners in the 1970s as the hobby spread to less-fit individuals. But Tucker and Dugas, too, conclude that certain types of running shoes may not be good for all distance runners, a conclusion supported by a range of research (see, e.g., Richards, Magin and Callister 2009).
In a review of research on barefoot running and training, Michael Wharburton (2001) suggests that running and walking without shoes may decrease acute injury rates from accidents (sprains), diminish chronic injuries from repeated shock (among them, plantar fasciitis), and increase movement economy, because additional weight on the feet is harder to carry while running than weight elsewhere (see Divert et al. 2008). Wharburton asks in his conclusion why more runners don’t opt to run barefoot, suggesting it might be fear of puncture wounds, thermal problems, or even misperceptions about the dangers. He does allow that in inclement weather and with certain biomechanical problems, shoes would be essential to compensate for lower limb issues (see Burge 2001 for reservations about Wharburton’s advice, especially with a range of medical conditions that she details – highly recommended if you’re considering running barefoot but have some pre-existing foot problems or other health issues).
A number of groups advocate barefoot running for a host of reasons: health, injury prevention, greater sensation, enjoyment, and overall well-being (e.g., Driscoll 2004; Robbins and Gouw 1990). Especially prominent websites include
Barefoot Ken Bob, Barefoot Ted, and evangelist Barefoot Rick (who’s all about saving soles… I know, ‘ouch.’ Sorry, Rick.). A recent book, Born to Run: A Hidden Tribe, Superathletes, and the Greatest Race the World Has Never Seen, by Christopher McDougall specifically discusses the Tarahumara Indians, who run extraordinarily long races through rough country in sandals or barefoot. The interest in barefoot running and the possibility that some types of shoes may be increasing problems for devoted runners has produced a spate of articles (see the list at the end of this article for a few).
As Ross and Jonathan have written of their own series of posts on running shoes, the topic is extremely controversial, provoking heated discussion, enthusiastic discussion, and strong opinions, no doubt because ‘shoes, more than any other topic, touches runners where it counts – their feet! And, unfortunately, their wallets, for it’s still the largest expense a runner incurs for the sport.’
They suggest that the trend in shoe design is toward very neutral (not motion controlling), cushioned shoes that are lighter than previous generations of footwear. In addition, virtually every shoe company has produced a ‘barefoot’ shoe design, minimalist footwear designed to mimic the dynamics of barefoot running. The Vibram Five Fingers, a glove-like light shoe, for example, was named by Time Magazine one of the Inventions of the Year in 2007. Vibram is even recruiting research subjects for Prof. Lieberman’s research on barefoot running dynamics.
I should point out that I have no personal interest in any shoe company, or in criticizing any shoe company. I run with shoes (when I run), but I do like to run barefoot on the beach whenever I can. And my border collie, Louie, is a fanatic about barefoot running…
Shoes, padding and running technique
The padding in running shoes changes the way that we run, even though we may be completely unconscious that our gait has compensated for the change in the biomechanical properties of the feet produced by footgear (see Divert et al. 2005; but c.f. De Wit et al. 2000).
Robbins and Gouw (1991) argue that, with padded shoes, ‘a perceptual illusion is created whereby perceived impact is lower than actual impact, which results in inadequate impact-moderating behavior and consequent injury.’ That is, the perception of impact that is diminished by modern ‘protection’ causes runners to neglect basic biomechanical adaptations to decrease stress on the legs, such as shortening the stride, changing the point of footfall, or increasing bend in the knees slightly.
Joseph Froncioni, an orthopedic surgeon, describes at length the way that shoes change the dynamics of running. Although the assertion that barefoot runners come down on the ball of the foot is controversial (some proponents and scholars argue that barefoot runners come down on the middle-outside of the foot; see Ross Tucker’s post on this debate), quite a bit of his description stands up:
During barefoot running, the ball of the foot strikes the ground first and immediately starts sending signals to the spinal cord and brain about the magnitude of impact and shear, getting most of its clues about this from the skin contact with the surface irregularities of the ground. Take away this contact by adding a cushioned substance and you immediately fool the system into underestimating the impact. Add a raised heel and the shod runner is forced to land on it. Strap the cushioning on tightly with the aid of a sophisticated lacing system and you block out shear as well, throwing the shock-absorption system even further into the dark…. The cushioned midsole of the modern running shoe robs the system of important sensory information necessary for ankle, knee and hip response to impact. The arch support (or orthotic) in modern running shoes not only prevents the arch suspension system from absorbing energy by preventing flattening but eventually leads to intrinsic muscle atrophy and complete loss of active muscular control of the arch leaving only the inelastic plantar fascia as a checkrein to flattening. The barefoot runner’s ‘foot position awareness sense’ which relies heavily on sensory input from the sole of the foot minimizes his risk of sustaining an ankle sprain on uneven ground. The shod runner is at marked increased risk of ankle sprains because his ‘foot position awareness sense’ is handicapped by the paucity of sensations coming from his soles.
Froncioni highlights here three distinctive problems with shoes in the dynamics of running: the first, a decrease in sensory information available through the foot; second, a shift in the position of the foot from a changed motion including an earlier heal strike and longer stride; and, third, an erosion of the impact-absorbing dynamics of the lower body, especially of the arch of the foot arising from both mechanical properties of the shoe and the previous two problems. Some of these detrimental effects are immediate, but others are gradual and cumulative, conditioning the body in patterns of behaviour and reaction that amount to a kind of adverse training that can result in chronic injury.
After a lengthy discussion in the comments on the Science of Sports blog posting on barefoot and shod running, Ross Tucker concludes that, in his opinion, the primary reason shoes cause injury is not the placement of the foot when it strikes the ground but the fact that heavily padded, stiff-soled shoes diminish sensation in the feet from the ground (similar to what Robbins and Gouw 1991 conclude, though they do so on the basis of less data). Without sufficient sensation, the foot and leg do not compensate as well for the mechanics of running; the feedback cycle is stifled and the dynamic suffers.
Research on foot impact by Robbins and Waked (1997) suggests that balance and impact are closely related, that a person coming down on a soft surface (like a gymnast landing on a thick pad or runner on a spongy shoe) intentionally, though non-consciously, comes down harder in order to find a stable surface. The spongier the landing material, theoretically, the harder the impact because the body seeks to compress the material to find some sort of stable footing.
According to Froncioni, shoes don’t simply disrupt the sensory feedback-control cycle through proprioception or the sense of impact through the legs, but also because wearing shoes changes the way that runners actively pursue sensory information through vision and use their bodies. That is, when we run in heavily cushioned shoes, we look differently and hurl our body against unknown surfaces.
The barefoot runner is constantly alert scanning the ground before him for irregularities and dangers that might cause him injury. The barefoot runner is a cautious runner and actively changes his landing strategy to prevent injury. He treads lightly. The shod runner is bombarded by convincing advertising stating or implying that the shoe he is wearing will protect him well over any terrain and he becomes a careless runner. He is heavy footed.
The loss of sensation in the feet is analogous to the effects of a degenerative disease, ironically enough. That is, by mimicking the long-term effects of neuro-degenerative conditions, shoes may bring on other forms of degeneration in the lower limbs. As Froncioni writes:
Finally, certain diseases in humans can cause a gradual destruction of the sensory nerve endings in the foot (and elsewhere) resulting in a significant increase in lower extremity injuries. Diabetes and tertiary syphilis are two. Extremities so affected are termed ‘neuropathic’. The shod runner, because of his sensory deprivation and high risk of injury may be termed as having ‘pseudo-neuropathic’ feet, a term coined by Robbins.
This and previous two drop quotes from Athletic Footwear and Running Injuries by Joseph Froncioni.
Conditions such as diabetes can throw off the fine orchestration of muscles in the feet that absorb and transfer force, as decreased sensitivity and response cause delays of dynamic reactions in the foot muscles (see Abbound 2002: 171, and for a review). As we’ve already discussed here on Neuroanthropology.net, some researchers who study loss of stability in older people point to diminished sensitivity in the feet as a potential contributing cause of falling. Not surprisingly, one of the prescriptions for people with this condition is to wear thin-soled shoes or, if the condition is worse, ‘high-tops’ so that sensation on the ankles can substitute for sensation on the soles of the feet.
Shoes as developmental niche for feet
People who habitually wear shoes wind up shaping their feet developmentally in distinctive ways. From the point of view of our feet – if I can be so anthropomorphizing – the shoe becomes the ‘environment’ in which feet are grown. Factors like temperature, abrasion, constriction, and the like become the environment with which the foot must contend adapt to, and rely upon. Shoes are a kind of developmental niche for feet, and like any ecological niche, exert their own influence on the anatomical unfolding of the foot’s anatomy. Of course, other factors in addition to shoes make up the foot’s ‘environment’, such as the very act and amount of walking we do, the surfaces we walk on, the sorts of forces exerted upon the bones in the feet by factors like our body size, built environment, athletic activities… and all of these can be affected by shoes, too.
In other words, from the point of view of the feet, a whole constellation of things make up the developmental environment, some of which are truly ‘outside’ us – like cold or wet or surfaces – but some of which are very much under human control, including activity patterns and habitual footwear. To the foot, the leg is part of the environment, and how the leg is used becomes one of the environmental factors feeding into how the feet develop. If we wear a pair of shoes that changes how our legs work (such as high heels or thickly-soled running shoes), these shoes affect the feet directly, but they also impact the feet indirectly through what they do to the leg and the dynamics of our gait and our patterns of activity.
In the simplest sense, shoes are designed to address what the shoe designers perceive as inadequacies in the human foot, whether these inadequacies are mechanical or aesthetic. Adam Sternbergh (2008) explained:
For decades, the guiding principle of shoe design has been to compensate for the perceived deficiencies of the human foot. Since it hurts to strike your heel on the ground, nearly all shoes provide a structure to lift the heel. And because walking on hard surfaces can be painful, we wrap our feet in padding. Many people suffer from flat feet or fallen arches, so we wear shoes with built-in arch supports, to help hold our arches up.
Of course, other design elements enter the mix along the way: the desire to be colour coordinated, the elongation of the leg provided by high heels, the undeniable cool of the tassel, the practicality of Velcro quick-release closures on kids shoes. But the basic ‘functional’ design elements of shoes are relatively consistent since the advent of modern, protective footwear (that is, providing more than simply insulation against cold by wrapping fabric or skin around the foot).
The basic effect of shoes on feet is relatively consistent as well. First, the sole of the shod foot does not develop the hardness that the unshod develop. Anyone who has ever lived in a variable climate (like I did growing up in St. Louis) probably has the experience of their feet fluctuating seasonally in toughness, going from soft and tender when constantly protected during the winter, swaddled in thick socks and insulating shoes, to toughened when barefoot or wearing sandals in the summer. When I worked as a lifeguard, by mid-July I could walk across the sun-heated asphalt parking lot at midday without my shoes. At the start of the summer, pampered winter feet were sensitive to every pebble or crack in the pavement.
In a study of shoe-wearing and habitually barefoot Chinese populations, Sim-Fook and Hodgson (1958: 1059) found:
The feet of the non-shoe-wearing populations showed thick soles with prominent skin creases apart from many minor lacerations due to traumata. The pachydermatous [!!] skin on the sole of the foot had an extraordinarily thick keratinized layer about 0.5 to one centimeter thick which permitted the individual to walk about without any discomfort. Although thick and tough, the skin was pliable and was marked by deep transverse folds which were similar to the lines of joint flexion found on the palm of the hand…
(Before I go any further, ‘pachydermatous’ is the coolest word EVER…)
Even though the groups studied spent quite a bit of time standing in water and unshod, Sim-Fook and Hodgson did not find many complaints about foot health, in part because their soles were so resilient and pliable, but also because the unshod did not have the constant low level friction on their feet provided by shoes. Ironically, this constant, low pressure against the foot can produce more severe chronic injury and malformation than the once-in-a-while and completely varied traumas of walking around with naked feet. Since the bones and tissue are, in a sense, being grown inside the shoes, they struggle to conform to some of the spaces and mechanical environments that we give them.
The second effect of shoes on foot development is that they influence the performance and architecture of the arch of the foot. As Dudley Morton (1964: 145) argued decades ago:
The natural foot is the naked, unclothed foot; and its arched conformation is not an element of weakness in design calling for artificial help, but of structural strength acquired through countless generations of unaided weightbearing. Occasionally we hear shoes referred to as a “natural support for the arch.” The suggestion should move our hearts in pity toward all primitive peoples were it not for the fact that they have no foot troubles, as well as no shoes. The phrase is one of many in which glibness overshadows accuracy, and unfortunately tends to promote erroneous ideas about the foot and its welfare.
The arch of the foot absorbs force when the feet impact the ground, stretching tendons in multiple directions, flattening and deflecting momentum. ‘Supporting’ the arch of the foot by placing it on a convex orthotic would make it virtually impossible for it to function as a shock absorber.
The arch support, which is present in all running footwear, would interfere with the downward deflection of the medial arch on loading. Furthermore, the use of orthodics, or other structures that are fitted to the mold of the soft tissues of the foot, could cause similar difficulty. Such designs occur when an engineer looks at the foot as an inflexible lever which is delicate and thus requires packaging. Various myths persist about foot behavior due to poor understanding of its biology. (Robbins and Hanna 1987)
Shoes also bind together the toes, making it very difficult for them to move, let alone engage in the grasping motions that habitually unshod people make when they walk (see Robbins and Gouw 1990; more on this below). To return to Morton (1964: 218), the bare toes move relative to each other to bear the weight of the body, and shoes affect their angle of spread: ‘The toes of non-shoe-wearing natives are separated when weight is borne on the feet; but any light, closely fitted foot covering will prevent their separation, owing to the lateral mobility of the toes and the small size of the muscles that abduct them.’ Sim-Fook and Hodgson (1958: 1060) also found ‘a tendency to spread’ in the forefoot, especially between the first and second toes (see also Funakoshi 2005).
Normally, the big toe (or hallux) diverges from the second toe at an angle of 5 to 10 degrees. But, in a condition referred to as hallux valgus, the big toe angles toward the small toes. When the condition is also accompanied by hypermobility, it is often congenital and referred to as ‘atavistic’ (although I suspect that this designation is not evolutionarily accurate). But the condition is often caused by wearing ill-fitting shoes, and it occurs 10 times more often in women as in men according to Richardson, Hansen, and Kilcoyne (2000; see also this source for astonishing X-rays of the effects of shoes on bone configuration… I was gobsmacked by a couple of the images). Morton believes that shoes have no noticeable effect on the functioning of toes, but we do know that habitually binding together the toes does affect the skeletal structure of the feet, and the evidence of pathology from shoes seems to me to be pretty compelling.
Patterns of bone growth and remodeling due to use (commonly referred to loosely as ‘Wolff’s law,’ see Ruff et al. 2006) suggest that a shift in toe use and the increased support for the bones of the feet provided by habitually worn shoes, will lead to differences in bone structure between habitually shod and unshod populations (see, for example, Sim-Fook and Hodgson 1958). Bound together laterally and ‘supported’ by an arched shoes, the foot cannot act as efficiently as a shock absorber; at the same time, less dynamic loading on the bones means that the bones will be less robust. Shoes, then, have a range of developmental effects, from low-level, constant pressure and abrasion to a form of protection which leads to greater fragility.
As a result, Zipfel and Berger (2007) recorded substantially higher rates of bone pathology in the feet of shod populations that they studied (European, Sotho and Zulu) than in pre-pastoralist South African populations who likely were habitually barefoot foragers. Although Erik Trinkaus’ work (see below) suggests that pathologies caused by shoes might be uneven distributed among the bones of the feet, Zipfel and Berger (ibid.: 209) found ‘the foot on the pre-pastoralist group is uniformly “healthier” than the modern groups.’
Ironically, even though Zipfel and Berger acknowledge that pre-pastoralist people show some signs of ‘wear and tear’ that might arise from much greater amounts of walking, constant travel and nomadic foraging, this heavy use pattern did not correlate with higher rates of a wide range of bone pathologies.
The results presented here suggest that the unshod lifestyle of the pre-pastoral group was associated with a lower frequency of osteological modification. The influence of modern lifestyle including the use of footwear, appears to have some significant negative effect on foot function, potentially resulting in an increase in pathological changes. (ibid.: 212)
I found it especially curious that the relative rates of pathology types and locations tended to be pretty similar across the different groups, but the overall frequency of pathological conditions varied, with shod populations’ rates of most disorders higher. This suggests that the wear pattern on feet is pretty similar, whether a population wears shoes or not; they get the same sorts of disorders, but less frequently without shoes.
The only way I can explain this is to assume that the shoes themselves don’t cause pathologies (otherwise, we’d notice some abnormally frequent disorders), but that shoes uniformly make the foot susceptible to disordered development. In other words, it’s not the shoes doing the damage, it’s that they throw off the foot’s ability to cope with normal movement, making the organ more fragile and susceptible to all pathologies (but note that this was only a study of bones, not soft tissue lesions).
The problem is not simply that we wear shoes, but that we often don’t wear the right shoes. Abboud (2002:176) reports that,
Since its inception in 1993, most patients seen at the Foot Pressure Analysis Clinic (FPAC) in Dundee, regardless of how minor or complex their problem was, were using ill-fitting footwear with discrepancies in shoe width and size when compared to their feet. In some cases, there was a difference of up to 3 UK sizes and 4 cm in width across the metatarsal head area, needless to say causing abnormal biomechanical force through the foot joints. The cumulative damage caused by footwear over the years goes inmost cases unnoticed and gets ignored despite clear signs of pain and dorsal callus formation, the latter can only develop as a result of friction with the inner shoe.
I probably don’t need to remind you that, as an anthropologist, I make little distinction between what people ‘should’ be wearing and what they actually are wearing. From the point of view of the feet, ill-fitting shoes are just as much a part of the developmental niche as perfectly chosen footwear.
Sternbergh explains the developmental influence of shoes simply: ‘This is the shoe paradox: We’ve come to believe that shoes, not bare feet, are natural and comfortable, when in fact wearing shoes simply creates the need for wearing shoes.’ Shoe designers are convinced that feet need to be protected against the ground, and the result is that our feet are so sheltered that they do become fragile.
The earliest shoesOtzi the Iceman, discovered in the Tyrolean Alps in 1991, was wearing shoes, but he was only 5000 years old. Even older remains suggest shoes had been around for a while: mummies in the Americas as old as 9000 years have shoes, footprints left by moccasins have been found in the Upper Paleolithic, cave paintings suggest footwear, and burials sometimes have beads on the feet and ankles that might have been sewn to leather shoes of some sort.
Archaeologist Erik Trinkaus has written a number of articles on the evidence for footwear in prehistoric populations, arguing that, in order to survive the cold of glacial periods, hominins would have necessarily figured out how to create insulating protection of some sort: a kind of prehistoric Ugg boot. But more modern-style, mechanically supportive shoes would have been a later development, evident in the bones of the feet because a semi-rigid sole will alter the distribution of force on the foot (see Trinkaus 2005: 1516). When walking barefoot, the toes flex, making the bones on the outside of the foot stronger through remodelling (as mentioned in the previous section); Trinkaus hypothesized that a shift in the robusticity of bones in the hallux (big toe) relative to the smaller toes (or the outside of the foot) would be a possible sign of habitual hard-soled shoe wearing.
Trinkaus compared bones from three different recent North American populations to test the hypothesis that shoes caused shifts in the relative strength of the toe bones (Pecos Pueblo Native American, Inuit, and Euro-Americans). Within these samples, predictions about the robustness of the phalanges in the feet based upon their shoe-wearing patterns turned out to be accurate; Pecos Pueblo Native Americans wearing soft-soled moccasins had the most robust lateral toes, Inuit in harder soled boots had more gracile bones, and Euro-Americans in hard-soled shoes had the most marked disparity. The more support offered by the footwear, the less robust the bones of the feet associated with the smaller toes (especially the pedal proximal phalanges in the middle of the foot).
Trinkaus has used beam model analysis, a technique that scans cross sections of bones across their axis to get some idea of their density and configuration. These donut-like images gives some sense of the stresses placed upon the bones because they remodel to compensate for these stresses, get stronger, in general, to withstand habitual strains.
A similar comparison might provide insight into the earliest rigid footwear because, as Trinkaus puts it, ‘relative robusticity of human lateral toes might provide insight into the frequency of use of footwear’ (2005: 1515). Because the organic materials likely used to make the first shoes would not endure in the archaeological record, Trinkaus’ method is as intriguing as it is ingenuous. In the archaeological remains Trinkaus examined, the evidence from the feet suggest that shoes became more and more prevalent from the Middle Paleolithic to the middle Upper Paleolithic; he suggests supportive footwear is likely around 30,000 years old in his earlier work (2005), but some of his later work with Shang (2008) may push that date back closer to 40,000 years.
I’m not going to go into all of Trinkaus’ analysis here. Blogger Afarensis has a number of posts on the issue of prehistoric footwear including here, here and here. Please read Afarensis, especially What You Can Learn From Bones: When Did We Start Wearing Shoes? for a more complete discussion of Trinkaus’ work.
By comparing the shoes to an ‘environment,’ I don’t mean to suggest that 40,000 years of being shod is a form of ‘unnatural selection’ that has shifted the genetic contributors to the anatomy of our feet. Rather, I just mean to suggest that, if shoes are affecting the anatomy of our feet, we have been transmitting certain kinds of crucial traits through the artificial environment that we’ve created. We place our children in little training shoes so that their feet are sculpted into a configuration that fits within, and virtually demands the support of shoes. So should we lose our shoes and go back to ‘natural’ feet, unwinding perhaps 40,000 years of non-genetic biophysical heredity?
Paleo-nostalgia and lifestyle adviceI often get students who come up to me after a lecture and want to know where I stand on some lifestyle movement that purports to be ‘getting back to’ some earlier human way of life. When I lecture on human dietary change, they come up to me to ask about the Paleolithic Diet or whether vegetarianism is more ‘natural’; when I talk about pregnancy, brain evolution, and altricial infants, they ask my opinion of different approaches to child rearing, or issues like breast feeding or co-sleeping.
I suspect that I usually disappoint my students, who can be pretty fervent about these ideas. Most paleo-nostalgia movements seem to me to be very selective – for example, the whole Paleolithic Diet movement seems to overlook a host of problems, such as changes in activity patterns, the difference between wild and domesticated meat animals, the high incidence of parasites and low life expectancy in prehistoric periods, and the likelihood that much of human protein was not coming from delicious medium-rare steak or grilled chicken breasts but rather invertebrates, shell fish, small vertebrates, offal and carrion (that’s right, maybe it should be the ‘Bugs, Clams, Lizards and Roadkill Diet’ – not quite the same marketing potential as ‘Eat All the Steak and Chicken You Can!’). I’ve discussed this in Paleofantasies of the perfect diet – Marlene Zuk in NYTimes.
So what about shoes and foot health? Is there anyone out there preaching the Paleolithic Podiatry program? Zinjanthropus shares my scepticism of podiatric paleo-nostalgia, asking why one period of our evolutionary history is privileged over others. Zinjanthropus writes:
Either way, I’m usually very cautious about shaping my lifestyle to fit the needs of a paleolithic savannah-scape. We’ve done a lot of evolving since then, after all! If I push my lifestyle back to the Paleolithic, then who’s to say that I’m not even BETTER evolved for the Pliocene?
If a hunter and gatherer diet, for example, is allegedly ‘healthier,’ why not push back to a diet of astringent fruit like our arboreal ancestors (as Richard Wranger points out, you’d be able to look forward to hours every day of chewing to get enough calories, for example).
Paleonostalgia suffers from a number of deep problems. As Zinjanthropus suggests, how to choose which period in time to use as a model. Hominins have evolved over millions of years through a whole range of environments; paleonostalgia tends to arbitrarily pick a point of time in the past, which is not necessarily more valid as a lifestyle model than any other. In addition, paleonostalgists tend to ignore the likelihood that human niches were varied – not as varied as later humans – but the ability to occupy diverse environmental niches has been a hallmark of our ancestors. Too much dietary and environmental specialization hasn’t really been a hallmark of our genus; arguably, the members of our genus and closely allied ones who have become too specialized and inflexible, have all gone extinct (I don’t want to argue this too strenuously, as many of the ones we tend to consider highly specialized a) lasted a hell of a long time, longer than Homo sapiens in some cases, and b) we’re increasingly uncertain that we can know for certain adaptive behaviours from anatomy, as the case of Paranthropus teeth suggests.).
Similarly, discussions of evidence from foraging peoples is often just as selective and slanted. Although we hear about the running capabilities of foraging people (and I, too, firmly believe that they were much more active than technologically-dependent sedentary people), we don’t hear about their injuries, including disabling ones, or their chronic health problems, including things like parasites that enter the body through the feet.
Alfred Gell, for example (I’m pretty sure, but I can’t remember in which text), wrote about travelling quickly through the rainforest with barefoot colleagues; although they were swift and sure-footed, they also had to stop every once in a while when one of them had to dig a thorn out of his or her foot.
One problem with paleonostalgia for barefoot running is the fact that we do not run in a paleolithic environment. As Trimble writes in Popular Mechanics:
The problem modern-day runners face, according to Hugh Herr, Popular Mechanics 2005 Breakthrough Award winner and head of the biomechatronic group at MIT, isn’t presented by our bodies but by the evolution of running surfaces. Humans that ran to scavenge or hunt for their food weren’t pounding concrete.
Running shoes offer a trade-off:
In his research, Herr focused on two problems with both shod and barefoot running-pronation angle and impact force. While barefoot running is best for a natural, stress-free pronation angle, Herr says, it is not ideal for coping with roads and sidewalks that can lead to stress-impact injuries. Shoes, on the other hand, excel at diminishing the force of impact on hard ground. But they do so at the cost of the natural stride-all the padding added to the shoe exaggerates the foot’s rotation.
So just throw away your shoes, right, and let your feet be free? Well, even the proponents of barefoot running caution that the transition from being habitually shod to running around au naturale can take some time because ‘the change in biomechanics and loading of joints, muscles and tendons threatens injury if you’re not careful’ (Tucker and Dugas, Running Shoes).
If running barefoot is so ‘natural’ to humans, why do we have to take it slowly? Because our feet become well adapted, as best they can, to wearing shoes. For all of the discussion of evolution having shaped human bodies and our feet for running, the body that habitually walks and runs in shoes has very much adapted to that niche. (See, for example, Tucker on attempts to change running techniques.)
But an interesting example of just how adaptable the feet can be comes from Shulman’s (1949) study of Chinese and Indian populations, in particular some individuals who might be expected to have the most damaged feet (if shoes were necessary to save our feet):
One hundred and eighteen of those interviewed were rickshaw coolies. Because these men spend very long hours each day on cobblestone or other hard roads pulling their passengers at a run it was of particular interest to survey them. If anything, their feet were more perfect than the others. All of them, however, gave a history of much pain and swelling of the foot and ankle during the first few days of work as a rickshaw puller. But after either a rest of two days or a week’s more work on their feet, the pain and swelling passed away and never returned again. There is no occupation more strenuous for the feet than trotting a rickshaw on hard pavement for many hours each day yet these men do it without pain or pathology.
Weren’t our feet designed for running barefoot?
In fact, a number of recent articles suggest that some of the traits of the foot (and other parts of the body) indicate that an ability to run barefoot might have offered a selective advantage during human evolution (e.g., Bramble & Lieberman 2004; see also Wired Science, These toes were made for running). But I don’t think that the issue is simply a debate between the running shoe industry and the growing ‘natural’ barefoot running movement. Instead, the anatomy of the foot, its sensitivity in development to the presence of shoes, and the evolutionary development of shoes and bipedalism, all illustrate how hard it is to talk about the natural human body at all or what the human body is ‘designed’ to do.
Patterns of activity, the most minimal technology, and the way we restructure our living environments all shape our physiological development. In fact, the role of activity, motor experience, and sense perception is so crucial in the development of so much of the human body and nervous system that I suspect we cannot even imagine how a person ‘without’ these sorts of influences might develop. Because humans are inherently adaptable — through culture, learning, technology, and even physiological change – it makes sense that plasticity itself would be a trait likely selected for in humans (an idea I take from Mary Jane West-Eberhard [e.g., 2005]).
Faced with the evidence that something as simple as wearing shoes can affect our soft tissue physiology, skeletal structure, gait kinetics, and the like, we can ask whether being shod or unshod is our ‘natural’ state. In a number of the internet postings about barefoot running, I find assertions about what sorts of surfaces or types of locomotion the human foot was ‘designed’ to accomplish. I think it’s too easy to just say, ‘barefoot is natural; shoes are artificial; feet were designed to run.’
In fact, the human foot and lege were not ‘designed’ for running or walking, barefoot or otherwise. They were not ‘designed’ at all. Evolution doesn’t design anything. Legs and feet are built by natural selection out of an appendage that, a very very long time ago, was a fin. If you were going to ‘design’ a limb and foot for running, you could do a lot better than the human architecture. Our knees, for example, are really lousy; they’re basically a rejiggered hinge joint and could certainly have been engineered better by a benevolent Creator. And She could have given us a more elastic set-up of tendons, too, something like kangaroos have. Oh, man, if some genetic engineer could just work on that kanga-human hybrid (a ‘kanga-hu’?), Olympic steeplechase would be so cool; no more of that stepping on top of the jump and landing in the water – but I digress.
Most of our readers will, of course, be completely familiar with the problems of the ‘Natural Selection as Designer’ metaphor, but it’s one that still crops up again and again in discussions of the evolution of traits. Normally, we can get by with the ‘design’ metaphor without too much trouble, but in the case of something like the role of activity in shaping the emergence of a physiological trait.
You see, human feet aren’t just good for running. They’re good for walking, standing, swimming, lifting, kicking, and a host of other functions. Like most primates, our limb use is actually pretty versatile; the arboreal niche of our ancestor presented a wide variety of challenges – hanging, swinging, walking on top of branches, standing bipedally, standing on all four. In addition, our primate ancestors, like us, don’t just use their limbs for locomotion; they use their limbs to manipulate objects, process food, hold offspring, interact socially, protect themselves, and a host of other activities.
Wait, you say, but we don’t use our feet this way. We’re humans. Feet are for walking and running…
Well, here’s the thing. Feet aren’t just ‘designed for’ walking or running; they turn out to be useful for all sorts of things. In the Chinese populations that Sim-Fook and Hodgson (1958: 1061) studied, habitually unshod people used their big toes often ‘to hold fishing nets and fishing lines taut so that the hands were free.’ The result was that these individuals developed ‘a remarkable degree of prehensile strength’ in the big toe (ibid.: 1060-1061). They conclude their discussion of the ‘unshod foot’ with the summary: ‘The unshod foot had laxity of the joints and tissues producing, in its natural form, a flexible foot with a degree of metatarsus latus, metatarsus primus varus, and hypermobility.’
You or I or the next guy may not be using our feet for things like peeling fruit or dialing the phone, but that doesn’t mean it can’t be done. In fact, many individuals congenitally born without arms or unable to control their arms due to a condition like cerebral palsy develop extraordinary dexterity with their feet, not only using them to do everyday tasks, but even activities like painting or playing an instrument. Painter Chan Tung-mui, for example, paints watercolours with her feet because she cannot control her hands due to cerebral palsy.Other prominent people who do a lot with their feet include painter and dancer Simona Atzori, Barbara Guerra (seen here on Medical Incredible), Mark Goffeney (guitarist for the rock band, Big Toe ), Tony Meléndez (barefoot guitarist, seen in this video playing ‘Let It Be’), and the late Bonnie Consolo, featured in the Academy Award nominated film A Day in the Life of Bonnie Consolo (released 1975) (here you can find a video of Bonnie Consolo typing with her feet (see also the site of the Association of Mouth and Foot Painting Artists of the World). Pravda carried the story of a Ukranian man, Sergei Vasyura, born without arms, who learned to shave, ride a bicycle, swim, build cars, bait a fishhook, weld, and even repair alarm clocks with his feet.
In most humans, especially shoe-wearing humans, the hallux is adducted, that is, in line with the other toes; but some degree of abduction is present in many of us, especially if habitually unshod, and may even develop to a slightly greater degree with use. Of course, no one approaches the abduction angles of our primate cousins who dwell in trees and have fully-functioning prehensile feet, but this crucial detail of human anatomy, one that distinguishes us from others, may be more variable than we think.
Shulman (1949) makes an off-handed remark about this that I found incredibly interesting: ‘Almost everyone surveyed showed a marked spacing between the first and second toes such as that found on young babies.’ I don’t know about the developmental dynamics, but it wouldn’t surprise me too much if, absent the adducting influence of shoes for more than half of our lives, and an even greater proportion of the time in which our feet were weight bearing, the angle of the toes found in infants was closer to the habitually unshod.
Although we may think that the Chinese practice of foot-binding is a kind of aberration, Zipfel and Berger (2007: 205-206) suggest on the basis of previous research that many Asian populations reveal the degree to which conventional shoes bind feet: ‘Studies of Asian populations whose feet were habitually either unshod, in thong-type sandals or encased in non-constrictive coverings have shown increased forefoot widths when compared to those of shod populations.’
As I wrote in the paper I presented at Univesité Montpellier (Downey 2009), just as Clifford Geertz (1973:67-68) argued that an uncultured human being would be a ‘mindless and consequently unworkable monstrosity,’ a skill-less human would not be capable of the most basic, defining ‘human’ physical acts. The fact that skills like foot painting or feeding oneself with one’s feet are rare does not mean that our feet were not ‘designed’ to do them.
If we were looking for a ‘natural’ foot, one without any influence of activity, we should probably focus on infants or on those who are disabled. We should realize that our feet were not ‘designed’ to do one thing or another; caring for them, and shaping them in ways that we desire, requires more than just figuring out what our ‘nature’ might be.
UPDATE: In January 2010, this issue was in the news due to the release of a new study suggesting that knee, ankle and hip damage might be greater for shod than unshod runners. Some sources have made the leap to the likelihood of osteo-arthritis, although the original study was biomechanical in nature. For a popular version:
Running Shoes May Cause Damage to Knees, Hips and Ankles, New Study Suggests
The original article (and abstract) is available here for download as a PDF:
Kerrigan, D. Casey, MD, Jason R. Franz, MS, Geoffrey S. Keenan, MD, Jay Dicharry, MPT, Ugo Della Croce, PhD, Robert P. Wilder, MD. 2010. The Effect of Running Shoes on Lower Extremity Joint Torques. PM&R 1 (12): 1058-1063. DOI: 10.1016/j.pmrj.2009.09.011
Bare Feet by Zinjanthropus at A Primate of a Modern Aspect
Ross Tucker and Jonathan Dugas at The Science of Sport published a whole series on running shoes and running dynamics in 2008:
Part 1: Do shoes cause injury?
Part 2: Shoes, injuries and training
Part 3: Running barefoot – the intelligent biomachine
Part 4: The footstrike – how should your foot land?
Part 5: The market and evolution of the shoe industry
Dylan Tweeny. 2008. Your Shoes Are Killing Your Feet. Wired Science (23 April). http://www.wired.com/wiredscience/2008/04/your-shoes-are/
Amby Burfoot. 2004. Should You Be Running Barefoot? Runner’s World. Available at: http://www.runnersworld.com/article/0,7120,s6-240-319–6728-0,00.html
Adam Sternberg. 2008. You Walk Wrong. New York Magazine (28 April). Available at: http://nymag.com/health/features/46213/
Tyghe Trimble. 2009. The Running Shoe Debate: How Barefoot Runners are Shaping the Shoe Industry. Popular Mechanics (22 April). Available at: http://www.popularmechanics.com/outdoors/sports/4314401.html
Joseph Froncioni. 2006. Athletic footwear and running injuries. Quickswood weblog (22 August 2006, but Froncioni admits to writing it much earlier).
Barefoot Ken Bob’s website http://runningbarefoot.org/
Barefoot Ted’s website http://barefootted.com/
Barefoot Rick Roeber’s website http://barefootrunner.org/ (Is it just me, or is there a pattern here?)
…El gringo sin los zapatos … Barefoot running blog
Barefoot vs. the Shoe blog, which hasn’t been updated in a while, but the truly obsessive might find interesting
And if anyone else wants to read it in Portuguese, there’s Correndo Descalço.
Photo of runners in the 1984 Olympics from the site, Barefoot Concepts.
Painted foot. Photo by Tom Schierlitz; makeup by John Maurad and Jenai Chin.
From You Walk Wrong, by Adam Sternbergh, the New York Magazine.
Abboud, R. J. 2002. Mini-Symposium: The Elective Foot: (i) Relevant foot biomechanics. Current Orthopaedics 16(3): 165-179. doi:10.1006/cuor.2002.0268 (abstract)
Bramble, Dennis M., and Daniel E. Lieberman. 2004. Endurance running and the evolution of Homo. Nature 432 (18): 345-352.
Burge, Caroline. 2001. Comment on Barefoot Running. Sportscience 5(3), sportsci.org/jour/0103/cb.htm
Clinghan, R., G. P. Arnold, T. S. Drew, and L. A. Cochrane. 2008. Do you get value for money when you buy an expensive pair of running shoes? British Journal of Sports Medicine 42(3): 189-193. doi: 10.1136/bjsm.2007.038844
De Wit, Brigit, Dirk De Clercq, Peter Aerts. 2000. Biomechanical analysis of the stance phase during barefoot and shod running. Journal of Biomechanics 33: 269-278
Divert, C., G. Mornieux, H. Baur, F. Mayer, and A. Belli. 2005. Mechanical Comparison of Barefoot and Shod Running. International Journal of Sports Medicine 26(7): 593-598.
Divert C., G. Mornieux, P. Freychat, L. Baly, F. Mayer, and A. Belli. 2008. Barefoot-shod running differences: shoe or mass effect? International Journal of Sports Medicine 29(6): 512-518.
Downey, Greg. 2007. Producing Pain: Techniques and Technologies in No-Holds-Barred Fighting. Social Studies of Science 37(2):201-226.
_____. 2009. ‘Interculturality, body & movement: On studying someone else’s skill.’ Keynote lecture. Conference: Le Corps em Mouvement 2, Francophone Association for Research on Physical and Sportive Activities, Université Montpellier 2 and Santésih Laboratory (Health, Education and Disability Situations), 4 June.
Driscoll, Dennis G. 2004 (2003). Barefoot Running: A Natural Step for the Endurance Athlete. Track Coach 168: 5373-5377. Available in several forms online, such as in manuscript form here.
Funakoshi, Kimitake. 2005. Secular changes in the angle of divergence of the first two metatarsals in the Japanese. American Journal of Physical Anthropology 75(3): 341-345.
Morton, Dudley J. 1964. The Human Foot: Its Evolution, Physiology, and Functional Disorders. New York and London: Hafner Publishing.
Richards, Craig E., Parker J. Magin, and Robin Callister. 2009. Is your prescription of distance running shoes evidence based? British Journal of Sports Medicine 43(3): 159-162. doi:10.1136/bjsm.2008.046680
Richardson, Michael L., Sigvard T. Hansen, and Ray F. Kilcoyne. 2000. Radiographic Evaluation of Hallux Valgus. From the University of Washington School of Medicine, Department of Radiology website. Accessible at http://www.rad.washington.edu/anatomy/halluxvalgus.html (accessed on: 27 June 2006).
Robbins, Steven E., and Gerard J. Gouw. 1990. Athletic footwear and chronic overloading. Sports Medicine 9(2): 76-85.
_____. 1991. Athletic footwear: unsafe due to perceptual illusions. Medicine and Science in Sports and Exercise 23(2): 217-224
Robbins, Steven E., and Adel M. Hanna. 1987. Running-related injury prevention through barefoot adaptations. Medicine and Science in Sports and Exercise 19(2): 148-156.
Robbins S, and E. Waked. 1997. Hazard of deceptive advertising of athletic footwear. British Journal of Sports Medicine 31: 299-303. doi:10.1136/bjsm.31.4.299. (Abstract and full text.)
Ruff, Christopher, Brigitte Holt, and Erik Trinkaus. 2006. Who’s Afraid of the Big Bad Wolff?: ‘‘Wolff’s Law’’ and Bone Functional Adaptation. American Journal of Physical Anthropology 129: 484-494. doi 10.1002/ajpa.20371
Shulman, Samuel B. 1949. Survey in China and India of Feet That Have Never Worn Shoes. The Journal of the National Association of Chiropodists 49: 26-30.
Sim-Fook, Lam, and A. R. Hodgson. 1958. A Comparison of Foot Forms among the Non-Shoe and Shoe-Wearing Chinese Population. Journal of Bone and Joint Surgery 40: 1058-1062.
Trinkaus, Erik. 2005. Anatomical evidence for the antiquity of human footwear use. Journal of Archaeological Science 32: 1515–1526. doi:10.1016/j.jas.2005.04.006
Trinkaus, Erik, and Hong Shang. 2008. Anatomical evidence for the antiquity of human footwear: Tianyuan and Sunghir. Journal of Archaeological Science 35 (7): 1928-1933. doi:10.1016/j.jas.2007.12.002
West-Eberhard, Mary Jane. 2005. Developmental plasticity and the origin of species differences. Proceedings of the National Academy of Sciences USA 102 (suppl. 1): 6543-6549. http://www.pnas.org cgi doi 10.1073 pnas.0501844102
Wharburton, Michael. 2001. Barefoot running. Sportscience 5(3). sportsci.org/jour/0103/mw.htm. (pdf available)
Zipfel, B., and L. R. Berger. 2007. Shod versus unshod: The emergence of forefoot pathology in modern humans? The Foot 17: 205–213. doi:10.1016/j.foot.2007.06.002