Corsets and Skeletal Deformities: Anthropological Study
In September 2015, The Canadian Student Journal of Anthropology (Nexus) included an anthropological study of women’s skeletons from England and France in the 1800s, when corsets were at their height in fashion. In this research study, PhD candidate Rebecca Gibson aimed to find any correlation between skeletal morphology (shape and relative position of the bones) and lifespan.
She documented how the ribcages and spines of corset wearers were modified from a lifetime of corset wear, and she gives us a window into how these women may have lived in order for their bones to have been shaped to the extent that they were. Gibson states that despite the fact that nearly all women in England and France wore stays between 1700 – 1900, this was a fashion perpetuated by women, for women.
Women themselves used, championed, and criticized corseting, and men often interpreted and disseminated the literature regarding the practice. What this view lacked, and this study seeks to rectify, is two-fold. Firstly, impoverished women’s voices are missing, both from the modern studies and from the written accounts. Secondly, the extant evidence that corseting was inherently harmful comes completely from hyperbolic and unreliable doctors’ accounts and as such it cannot be verified using the literature alone. ~ Gibson, pg 48
What Gibson explains (in addition to Norah Waugh, Valerie Steele and several other authorities on historical corsetry) is that men wrote publicly and extensively about their distaste for the corset; often comparing the (then modern) small-waisted woman to the statue of Venus de Milo. Dr. O’Followell himself (if you remember my previous discussion of his 1908 X-rays of corseted women) made the argument that the Venus is universally and objectively considered beautiful, and through a game of logical hopscotch he concluded that anything not-Venusian (i.e. a nude small-waisted Victorian woman), therefore cannot be beautiful.
Gibson found however that 50 years prior to O’Followell’s study, in his 1868 book Freaks of Fashion: The Corset and the Crinoline, William Berry Lord wrote that “No fallacy can be greater than to apply the rules of ancient art to modern costume.”
Lucy’s note: The apparent volleying of subtle sass between writers during this era pleases me.
If you wish to skip over Gibson’s anthropological study itself, the conclusion is that she showed plastic deformation of the ribcage into a more circular shape as compared to the broad, ovoid flaring of a “control” modern ribcage, and also noted some downward bending and overlapping of the spinous processes in the thoracic spine. However, these deformations were not seen to correlate with a shorter lifespan of the subjects, and on the contrary the subjects reached or exceeded their life expectancy at birth.
Layperson’s explanation: The skeletons of 19th century corseted women were studied to see how their ribcages were flexibly bent into a more tapered shape from the corset. From the photos, you can see literal ‘bends’ in the ribs where the pressure from the corset formed the ribs into the shape of a circle. Also, the spinous processes seemed to be affected too: spinous processes are the small “spikes” humans have on their vertebrae; they look like spikes down a lizard’s back, but in humans these are small and one can occasionally see or feel them as the ‘bumps’ along one’s back. In the skeletons that showed rib shaping from a corset, these same skeletons also had “spikes” in the upper back that bent downward and overlapped like snaggleteeth. Despite this finding, the age at death for these subjects were average or older than the national life expectancy at the time, even correcting for infant/childhood mortality. Therefore, even though corsets have been shown to deform the skeletons of these subjects (and the reasons why will be discussed later), it didn’t affect how long they lived.
Below you’ll find my summary of the study, Rebecca Gibson’s answers to my questions concerning the study, and my thoughts on how this affects what we know about modern body modification through corsetry.
For this study, the following stats were given:
Studying corsets of the time:
44 corsets are kept in the Victoria & Albert Museum in England which qualified for the time period assigned for this study (200 year span, from 1700-1900). Of the 44 corsets, 32 of them were allowed to be examined – she measured dimensions of the corsets and noted any wear, tear and repair to the corsets. 24 of them were made from whalebone and 8 of steel. Their textiles ranged from cotton (most common), silk, linen, wool, and leather (least common).
Closed waist size of the corsets ranged from 34 – 80cm (13.4 inches to 31.5 inches).
15 of these corsets were apparently designed to be worn completely closed (see my discussion with Gibson later in this article). Of these, the mean average waist size is 56.33cm (22.1 inches).
Studying skeletons of the time:
Two skeletal collections were studied, one collection in London England (Centre for Human Bioarchaeology at the Museum of London) and the other collection in Paris France (Musée de l’Homme in the Museum National d’Histoire Naturelle). The French skeletons were preserved and articulated, while the English skeletons came from a burial site typically used for people of lower social or work status: country residents, prostitutes, prisoners and inmates.
Gibson first chose skeletons showing classic deformation of the ribcage from the pressure of the corset (which in O’Followell’s study, pressure from the corset is shown to be measured up to 80psi). The ribs were more circular compared to an anatomically “normal” human ribcage (the “control” ribcage).
Of course, Gibson took into account that ribs can be deformed by other factors such as rickets (Vitamin D deficiency causing soft bones into adulthood), congenital deformity (birth defects), ankylosing spondylitis (fusion of the vertebrae over time), or osteogenesis imperfecta (brittle bone disease). These issues had to be weighed against the characteristic deformation from a corset being worn from adolescence, when the bones were still malleable.
The age of each subject at the time of their death could be estimated by studying various factors, like:
- cranial suture fusion (the plates of one’s skull fuses tighter as one ages)
- osteoarthritis (breakdown of the joints, and of bones near the joints)
- ankylosing spondylitis (when the joints fuse together and become stiff/ immobile)
- mandibular bone resorption (when the jawbone erodes over time – this can affect tooth loss), and how worn down the teeth were overall from a lifetime of chewing.
(These factors are all independent of corset wear, and are still conditions that affect many people today.)
What deformations were visible from corset wear?
Gibson looked for deformation of each vertebra individually, and also nested the vertebrae together to see how they fit as a set. She also looked at how the rib shaping was affected and whether this was symmetric or asymmetric.
How does one know that a ribcage has been formed by a corset and is not a sign of rickets? Gibson points to a 1981 publication by Putschar, which explains that rickets flattens the curves of the ribcage, and most of the ‘bend’ occurs at the costal joints, especially at the sternum (“pigeon chest” is common) – and in extreme cases of rickets, the pressure from one’s own arms laying at the side of the body can cause the ribcage to cave in at each side.
In the French skeletons, instead of seeing “flattened curves”, Gibson noted that the ribcage was more rounded, such that when the dimensions were measured, the coronal (front-to-back) and sagittal (side-to-side) diameters were identical (or close to it). Also, the area of the rib with the greatest bending was closer to the back of the body, not at the sides where the arms would be. Gibson said that formation was seen as high as the 4th rib (imagine right up in the armpit) and the corset molded each successive rib consistently and uniformly. The floating ribs (11th and 12th) were sometimes affected even more.
The circumference measured around the 6th or 7th rib (which I consider to be close to the “underbust line”) was calculated to average to 57.9cm or about 22.8 inches around. Keep in mind that muscles and flesh over this area may contribute to a few more inches in a living person!
I found Figure 6 of the article to be of particular interest, where a single pair of ribs shows plastic deformation of the rib (actual bending near the back), a broken area that had healed later in life, and a post-mortem breakage (obviously not healed), showing how different all of them appear.
Lucy’s note: I remember reading in a book a few years ago that the way skeletons were preserved for scientific research in the 19th century was very different to how it’s done today, and the way the ribcage was boiled in solvent and hung to dry would have caused the costal joints to sag (especially at the sternum) and contribute to the deformed result. This is always my concern when studying preserved, articulated skeletons. However, these disarticulated ribs exhumed from an old English cemetery wouldn’t have had the same preservation process, so this study is the first time I’m observing solid bone showing antemortem plastic deformation in the ribcage, and not the cartilage breakdown post-mortem). Fascinating!
Figure 8 of the article shows the deformation of the spinous processes, how the “spikes” don’t stick directly outward but rather angle themselves downward and overlap with one another.
Gibson acknowledges that it’s possible that a combination of malnutrition diseases (like rickets) allowed the skeleton to maintain a certain level of plasticity into adulthood. Vitamin D deficiency could affect all classes of people, not only those who couldn’t afford sufficient food; higher class women stayed out of the sun and covered large parts of their bodies up until the early 20th century. Also consider that adolescent girls started wearing corsets much younger in life than we would today; commencing when their bones were still relatively soft to begin with.
What do we know about the women who lived in this era?
Gibson says that “Corseting, for a woman during this 200-year period, was the default position.” It was a utilitarian garment which was worn by all demographics and classes in England and France. But just as people today don’t necessarily keep up with modern fashion, don’t necessarily wear clothes that fit properly, and don’t necessarily purchase their clothes brand new, a Victorian women from a lower income family may have worn an older, imperfectly fitting, but otherwise lovely silk corset passed down to them or purchased 2nd hand. Therefore we can’t presume that only the high class wore one style of corset and the impoverished only wore another style.
Comparing waist sizes, then and now:
The corsets that had to be laced completely closed (and constructed between 1700 and 1900) were an average of 56.33cm (22.1 inches) in circumference. A 1951 survey in the UK showed an average natural waist of 70cm (27.5 inches), and in 2001 the average natural waist in the UK was 86cm (33.8 inches) – so we are seeing a gradual expansion of the waist over time. This isn’t to say that a woman’s natural waist was 22 inches in 1800, but tightlacing was not an everyday occurrence.
Did these skeletal deformities affect life span?
In France between 1745-1905, life expectancy at birth was said to be 25-49 years. The average age at death for the subjects studied from the French collection varied: some were between 30 – 45 years, three of the subjects may have been up to or exceeding 50 years, and one of them could have been older than 60 years at the time of death. Gibson said that the age estimation was very conservative, encompassing a 10-15 year span, just in case these women underwent early signs of aging (as we know today: illness, stress, genes, etc. can lead to premature aging markers).
In England between 1706-1901, life expectancy at birth between was between 35-50 years, and age of death averaged between 48-60 (which may be a more accurate representation because it excludes infant/child mortality). Of the skeletons studied in England, 4 of them died between 35-45 years of age, and 14 of them were older than 46 years.
Therefore the women who showed skeleton deformity from a corset lived just as long, and the author argues that many of them lived longer than the life expectancy at birth.
Was tightlacing a common practice?
Gibson states that the comical artwork (where ladies held onto bedposts while their maids strained as hard as they could) was satirical, and just that.
“The corsets [that survived to be displayed in V&A] themselves tell a different story: a story of being often and well worn, but handled with care and used for years, something that would be less than likely were they pulled tighter than their frames allowed.” ~ Gibson, pg 57
Lucy’s note: I remember going to the Symington collections while in England in 2014, and observing some of the antique corset labels and adverts which boasted “guaranteed to last 1 year!” – which says something about how quickly a daily-worn corset could wear out, or how frequently they may have been replaced (when one could afford it). Of course, how many of us today use items past their warranty, and how often do we see antique corsets which are mended, taken in, had gussets added, repaired, etc.? Many. After all, even well-cared-for and gently worn clothing breaks down over time when used daily. It was said that Empress Sisi of Austria wore through her corsets in a matter of months and had them replaced several times a year to keep up with her tightlacing. Of course, Empress Sisi was not an average woman of the time.
But from a social point of view, when the occasional woman did tightlace, it was women who chose to tightlace, while men collectively shook their heads at the practice.
A Conversation with Rebecca Gibson (and my Chiro)
Over the weekend I emailed Rebecca for clarification on certain parts of her study, and also contacted my regular chiropractor for his take on the findings. Below is a summary of our points of discussion:
Lucy: In addition to measuring the ribcage diameter and calculating the circumference, were you also able to measure the total height of the subjects and compare overall smallness of frame compared to ribcage breadth? This information would help to confirm/ refute the common belief that people were shorter and smaller of frame overall during the Victorian era.
Rebecca: I was not able to measure the height, as skeletal height is more of an estimation (though a good, educated one) and it was unimportant to my overall study. This may be something I eventually address in my dissertation, although that may depend on how much time I have to devote to the subject of height or whether or not I can find an angle for it in the overall research plan.
Lucy: How was it determined that 15 of the corsets were designed to be laced completely closed in the back, vs having a gap in the laces? Was it the way that they were constructed that did not allow for a gap?
Rebecca: The ‘laced closed’ determination was made based on the construction/design patterns of the corsets. Some of them were designed to be buttoned or hooked in the front and had no other closures, which would preclude any gaps while wearing them, while others were built into garments–as was the case with the ball gowns and the riding habit. As your blog shows, there are ways to adapt many corsets to the ebb and flow of changes in weight, pregnancy, and needs for access. These 15 were of designs that did not allow for such adaptability.
Lucy: I noticed that the diameter of the ribcage was taken around the level of rib 8, just below the scapula in the back (and around rib 7 in the front, just below the xyphoid process). This area measured on myself would be just under the breast area (what we would call the “underbust line” when measuring for corsetry) and compression this high up on the ribcage is considered uncomfortable/ unusual for modern corsets, although it may have been an area of higher compression for those living in the 1700s in particular when the more conical “pair of bodies” were more popular than the flaring, demibust stays in the mid to late 1800s. Did you notice any correlation between the date that the subject had died (and thus the fashion era that they lived in) vs the measurement of their ribcage at this area?
Rebecca: The French skeletons were all from the 1800s, despite my period of inquiry starting from the 1700s–there are limits to any collection, and that was one of them. The English skeletons were disassociated from their individual burial plots, meaning it’s impossible to tie any particular skeleton to any particular burial location which would include such things as headstones bearing demographic information like death dates, and so I can only give a date range for their collective deaths based on the time during which the cemetery was active. As I was looking at single ribs and rib pairs from disarticulated skeletons, it was not possible to visualize the entire ribcage, and even if I could have done so, I could not state for certain when that woman died. This is a longer way of saying that there’s no way to collect that particular data, though I do so wish there had been.
Lucy: In Figure 8, it showed that the spinous processes were deformed
and overlap in the thoracic spine, and you mentioned that these
vertebrae were studied both individually and as a set. For these photos, were the individual vertebrae manually stacked? I observed that there was no spacing between each vertebra to represent the width of the invertebral discs, which in a healthy adult is an average of about 7mm between each vertebra – and I was wondering if that spacing would affect the frequency of overlap observed between the processes. I also noticed that the stack of vertebrae is very straight, whereas it is normal and healthy to have a very mild kyphotic curve through the thoracic region, and this curve may further prevent the processes from overlapping. If the section of vertebrae were given to you already fused (via ankylosing spondylitis or another condition), this makes the research more definitive. Were you able to take photos of a “control spine” (a skeleton that had no ribcage markers of corset wear) and see how the spinous processes aligned when the vertebrae were stacked in a similar manner?
Rebecca: I did manually stack the vertebrae. There is some space between the vertebrae, though a smaller amount than one would normally see in a healthy adult. I think what we’re seeing in Figure 8 is also extreme spinal compression downward, which compressed the intervertebral discs to smaller sizes than their typical 7mm. The reason they appear the way they do in the photo has to do with the way I matched each vertebrae to its neighbor–I matched articulation points. As bones move over each other, they create facets of articulation. I matched these facets together to ensure that I was getting the proper vertebral order, as well as using the same technique to side and order the ribs. Therefore, what you see in Figure 8 is exactly how the bones themselves fit together,
articulation point to articulation point, regardless of what we consider normal or healthy. Corsets are not accommodating when you wear them as often or as long as many women would have during the time between 1700-1900, and it is entirely possible that the normal curve changed over time. Also important to note is that I’m in no way claiming that these were healthy women–at least not in the way we understand health today–merely that they lived longer lives that were less constrained than how we normally view them.
Lucy’s chiropractor found it interesting that there would be downward spinal compression of the thoracic region, as the support of a well-made corset acts as a traction brace to reduce and prevent spinal compression in a modern wearer. Of course, we are not dealing with modern corsets in this study. He says that we would not be able to study the thoracic curve of the subjects even if we wanted to, as how the spine is held neutral would be reflected in the soft tissue and not the vertebrae – the curve is influenced by the discs, and surrounding muscles and ligaments. He also notes that throughout the thoracic spine, normal spinous processes are already angled slightly downward, although may not necessarily overlap. It would not be impossible for these processes to deform with regular pressure and a predisposition to softer bones, but whether that would definitely cause health issues or symptoms, there’s no way to confirm this in the subjects obviously.
Here is an article with more information on the anatomy of the thoracic spine which does claim that sometimes the thoracic processes do overlap in modern people, but I think it would be cool to see a photo of a “control” stack of T1-T12, and the affected T1-T12 of the corset-wearing subject just to confirm.
Lucy: Figure 8 also pointed to the deformation of the spinous processes to show that they were visible as high as T3 and T4. I found this intriguing as this is around shoulder height on the body, and a corset that would have wrapped around the torso so high on the ribcage would certainly cut into the armpits and likely cause the shoulders to be raised or hunched, or in the very least leave the scapula immobile – however in Victorian photography and artwork, it seems that the opposite posture was more fashionable, with the shoulders down and back, and the neck elongated (which is consistent with the style of the corset in the 19th century typically coming up to nipple height only, or possibly 2-3cm higher than this for modesty more than for compression). When you measured the height of the corsets (from the waistline to the top edge) in the V&A museum and compared them to the height of the ribcages of your subjects, were you able to estimate how high up an average corset would have stopped on their body?
Do you think that an indirect postural change may have affected the deformation in the higher thoracic vertebrae, instead of a plastic deformation from regular compression?
Rebecca: While in general the styles of the corsets would not support the T3 and T4 data, you would be assuming that women a) wore corsets made specifically for them, which were well made and properly fitted; b) bought new ones as soon as the old were outgrown, and c) followed the styles completely and competently. It’s important to note that we as 21st century people do not do most of these things now, with, for example, high heels–and as a result we end up with bunions. This was the burgeoning age of off-the-rack, mass produced clothing, as well as a time of extreme poverty which led many to buy second hand clothing which did not fit very well. Furthermore, two of the ‘symptoms’ of corseting mentioned by O’Followell, even though I take his work with a grain of salt, are cuts in the armpits and rounded shoulders. I should correct you–I did not compare the heights of the corsets to the heights of the rib cages. I may do that in my dissertation, however. The larger research project is still in formation. However, bad posture alone is not likely to affect a plastic deformation in bone.
Lucy’s chiropractor agrees that a postural change is not likely to create deformation of the spinous processes.
Lucy’s thoughts: I’m now wondering if the strong downward compression Rebecca mentioned – and consistent pressure on the upper spine – may have been attributed to high-backed stays with stiff shoulder straps worn during childhood, especially if the strap length were not adjusted during periods of growth. In the related Youtube video, others have suggested that the loss of curve and downward compression may have also been attributed to a lifetime of heavy lifting and carrying, the use of Victorian backboards as children, and poor nutrition leading to thinner-than-normal vertebral discs.
Lucy: In figures 6 and 7, where the yellow arrows show breakage of the ribcage, is there any way of knowing whether this was caused by corset wear over time or whether these were occupational injuries (as these were thought to be working class women based on their associated burial place for families of lower income)?
Lucy’s note: This question stems from a statement I once read in a book that the force required to break a healthy human rib exceeds the maximum tension of cotton fabric – that is to say, the seams or the fabric of a corset would tear before a rib would be broken – and prior to this threshold of tension, there would be obvious pain felt in the affected rib, which would theoretically discourage the wearer from wearing a garment that tightly in the first place.
Rebecca: There are many, many ways to break a rib, including accidents, domestic violence, and various conditions that would make one more prone to bone breakages in general. I do not think I can speculate about the cause of these breaks, although that is what I originally went to look for–I only turned to circular rib cages and spinal deformation after I realized what I was seeing. Good science tries to explain, rather than to speculate on, what the researcher sees in the world, so unless I can at some point figure out a way to match the women to their burial plots, and then find either life histories or living relatives who can give me information about potential injuries sustained before death, I am not going to move my research toward the breaks seen on figures 6 and 7.
Lucy’s take: What does this mean for modern corset wearers?
I personally believe that the rib and vertebral deformities seen in this study would be much more difficult to achieve in today’s waist trainer, for several reasons:
- Take into consideration that the silhouettes of popular corsets today are quite different compared to antique Victorian corsets. Looking at photographs of Victorian women and studying antique corsets: traditional Victorian corsets often had a smooth, almost concave looking sloped silhouette which compressed the upper and mid ribcage, yet didn’t always have a sharp nipped-in dip at the waistline. Additionally, in the profile view, a rounded lower tummy was normal – whereas today we are preoccupied with having a flat front, reminiscent of Edwardian corsets. We also allow for more room for the upper ribcage in modern corsets; not only for aesthetic purposes (avoiding “muffin top” – the flesh spillover in the back by the scapulae) but also for comfort purposes and the ability to take deeper breaths. The aesthetic for a broader ribcage has changed over the 20th century as well: consider photos of Victorian women where their upper ribcage looks relatively slim, and then consider photos of admired women in the 1950s where the archetype was to have a sharply nipped in waist, almost literally a dimpled waist, but then a full upper ribcage:
Today, the convex, rounded or cupped ribcage is becoming more popular. Even though I personally find a conical rib corset to be quite beautiful, I have to agree that I find a cupped rib more comfortable and easier to wear over longer durations.
- The majority of corsets today are also underbust corsets, often paired with a separate bra. Bras today, as mentioned in the publication, obviously don’t cover as much of the ribcage, and they’re elastic and not as restrictive. So underbust corsets affect primarily the lower thoracic region and lumbar area, and perhaps not necessarily the upper thoracic area as much as the antique style corsets.
- Today we also have a better understanding of proper standing posture – recall that the first medical X rays were used in 1895, and prior to this, medical understanding of posture of a skeleton was taken from cadavers lying flat, which is different from standing posture. (Same with the position of organs.)
Today, we can create corsets that are more anatomically correct and posture corrective to maintain a healthy neutral spinal curve; and not flatten the slight thoracic kyphosis and slight lumbar lordosis.
But if compression from a corset can over time cause a change to the angle of the spinous processes, this is perhaps one of the best arguments to wear a corset with a small gap in the back laces to clear the spine, as opposed to wearing a corset completely closed (one I must begrudgingly admit, since I prefer the appearance of closed laces).
- Lastly, these days corset wearing starts much later in life. Gibson did say that children between 1700-1900 wore stays and I agree; although according to the curator at Symington they were typically not boned as rigidly, they did not lace up (and therefore could not be tightlaced), and they were said to more provide warmth and support to encourage growing upright (much like staking a sapling) as opposed to shaping the waist – likely a response to scoliosis, congenital spinal deformities, or nutrient related deformities; the cause of which were not widely understood until much later. When adolescence was reached, this was when an hourglass shape was made pronounced and rib shaping likely became a bigger player (see the silhouettes for the ‘Good Sense’ styles above, with seemingly increasing curve corresponding with each older age group).
Still, as Gibson explains: not every child, young girl, or woman was fitted with a state-of-the-art, correctly-fitting and age-appropriate corset. Young age and malnutrition-driven bone softness allowed the ribcage to be much more malleable and more prone to bending or breakage in a working class 18th and 19th century woman, as compared with the more rigid bones of a modern, healthy, and well-nourished person who chooses to commence waist training as an adult. Therefore it would be much more difficult to deform bones as a modern healthy adult to the extent that one sees in this study.
Huge thanks to Rebecca Gibson and my chiropractor for both taking the time to answer my ridiculously detailed questions!
What do you think about this study? Leave a comment below!
Le Corset (O’Followell, 1908)
X-Rays from “Le Corset” (1908) Explained (Lucy’s Corsetry, 2013)
Corsets and Bones [which I now have to update to include this study] (Lucy’s Corsetry, 2012)