Separated by a Common Language

So last time, I wrote about disease versus infection following the phrase sexually transmitted, and I started thinking (again) about how we talk about medical things--technical or non-technical? In the book I'm writing (for you!), I've touched on it a little with respect to bodily functions:

Sitting in my doctor’s waiting room, I’m amused and a bit astonished to find posters about what to do if there is blood in your pee or poo.* The equivalent American public-service advertisements say urine and stool. In the medical context, America avoids being crude by sounding more scientific, and Britain uses baby-talk.
* Make an appointment with your doctor immediately!

The discussion hits on things like BrE waterworks ('urinary tract') and back passage ('rectum') and  classes given to foreign nurses in the NHS on British slang--aka British euphemism. (It's a bit of the book that looks at the stereotype of Americans as euphemists, so yes, there's a lot of attention in the other direction.)

The "Americans use overly technical terminology" aka "Americans like jargon" stereotype that I contribute to in the quote above is one worth taking apart as well. I've been encouraged in that stereotype when I hear friends talk about their chest infections where I tend to have bronchitis. But then they're also talking about having cystitis (my poor, unhealthy friends), which I hadn't heard of before moving here. An infected cyst? Ew... No, it's a bladder infection. In AmE I'd call it a UTI (urinary tract infection). (The NHS website tells us that cystitis is a common type of UTI, so the terms are slightly different--but that's the case with bronchitis and chest infection too.)

So far, we're tied: Itises: BrE 1, AmE 1. Infections: BrE 1, AmE 1.

So I thought I'd have a look at which things Americans and Brits call infections and which they use Medical Greek -itis names for.

 The tables below are the statistically "most American" (left) and "Most British" (right) nouns that come before infection in the GloWBE corpus. (If you click on the table, it should get bigger.)

"Most American" and "Most British" words preceding infection

What you can notice there (if you can read the small print) is that the "more British" preceding nouns include things that can get infected (wounds, chests, throats), whereas the "more American" ones tend to be the microbes that do the infecting. In AmE, I think I'd say someone has an infected wound rather than that they have a wound infection. And one kind of wound infection you can have is a staph infection (in the US list), which is a very familiar term from my AmE childhood (we were constantly being told that gym mats were very dangerous). I don't know that I've ever heard staph infection in the UK.

In the BrE column you can also see urine infection, another BrE way of saying urinary tract infection. This one names neither the pathogen nor the organ, and always strikes me as a bit odd. Urine might have germs in it, but can urine itself be infected?

BrE has more throat infections because Americans are more likely to say they have strep throat. In my experience, scarlet fever is heard more in BrE these days (which is not to say you never hear it in AmE). When my child was diagnosed with it (in the UK), I really felt like I'd been taken back to Victorian times. She wasn't all that sick. But when I looked it up and found that it's the strep germ, I thought: maybe you hear scarlet fever more often in UK because AmE has strep infection.

Some of the numbers up there, though, are art{e/i}ifacts of the corpus. AmE has 56 instances of HSV infection but all of them come from a single website (virologyj.com), so we shouldn't take too much from that. American, like British, English would typically call that herpes. HBV infection is found on a greater range of sites, but they are mostly medical journals and such. Laypeople would generally say Hepatitis B.

But that does seem to sum up the difference between the AmE table and the BrE table: a lot of the AmE infection cases are use of medical jargon in a medical context--staph infection was the only one I knew as a layperson. Whereas in BrE the body-part+infection cases are terms that non-medical people would use when talking about their maladies.

If we look at the infections that American and British English have in common, we can see  that Americans do talk about infections too, sometimes with body parts, even.


But what about -itises? Is it mostly Americans using the fancy words? No, but there again maybe some effects here of one source being over-sampled in the corpus. Here I'm showing what came up as 'most American' (left) and 'most British' (right), with a bit of the 'neither one nor the other' showing in white. This is going to be very hard to read on a phone (sorry!), but I'll write up the highlights below.


I've given a comment in red if (a) the things are not diseases, but just coincidentally spelled with -itis, or (b) if it's a spelling issue. Though oesophagitis shows up in the British list, it's not because Americans don't use an -itis name for the problem, but because we spell it esophagitis. (Click here for my old post on oe/e spellings.) The British list is lengthened by a misspelling of arthritis and having two spellings for tonsil(l)itis.

After discounting those, the British list is still a lot longer than the American one, but I'm very much suspecting some bad corpus effects here. Tonsillitis, colitis, dermatitis, gastroenteritis, appendicitis, pancreatitis--I or members of my family have had all of these and that's just what they were called in the US. The numbers for these diseases are greater than expected in the British part of the corpus--but they're hardly absent in the American part. For example, note that there are 756 AmE occurrences of meningitis--which is here counting as "rather British", while only 16 AmE hits for phlebitis make it "very American". Some of these cases are going to seem "more British" or "more American" to the software just because the corpus happened to hit on some websites that talked about these things a lot. But I think what we can say from this exercise is that we have no particular evidence for British English avoiding -itis words, despite its greater use of body-part+infection.

Still there are a few itises worth mentioning for BrE/AmE interest. One is labyrinthitis, which I had an unfortunate encounter with this spring. When I described my symptoms (the room going upside-down and inside-out every time I turned my head left), lots of British friends said "Oh, that's labyrinthitis. I've had it. It's horrible!" But it was not a word that my American friends seemed to have at their fingertips--to them it was an inner-ear infection. (Why do Brits seem to get it more often, though?)


Conjunctivitis shows up on the British list, though it is a word that Americans use too. But Americans have another informal term for the problem: pink-eye. That will push the US conjunctivitis numbers down. (There are a few UK hits for pink-eye--with or without the hyphen, but a lot of US hits.)

In the white part of the table--where the numbers are similar for AmE and BrE -- are the two itises that are earlier in this post: cystitis, which I've experienced as more British, and bronchitis, which I've experienced as more American. Because the corpus is imperfect, I'm not going to totally discount my experience on these. But it would be interesting to hear if others (particularly transatlantic others who can compare) think I'm off my rocker...

I was surprised to see only one made-up disease in the list: boomeritis on the AmE side. (It was the name of a book--click on the word to learn more.) I would have bet that (AmE) senioritis would appear. (As it happens, there were only two US examples of it in the corpus--most are Canadian.) To quote Wikipedia:

Senioritis is a colloquial term mainly used in the United States and Canada to describe the decreased motivation toward studies displayed by students who are nearing the end of their high school, college, and graduate school careers.

For a minute there, I was worried that I expected senioritis to be there because I am OLD and UNCOOL. But I'm happy to report that in both the Corpus of Historical American English and in Google Books, the rate of senioritis use has only gone up in the decades since I was a high-school/college senior. Not happy for the teachers who have to teach these seniors, but happy that my vocabulary is not a complete dinosaur--yet.

If you're interested in other disease names, do have a look at the medicine/disease tag--thanks to (a) having a small child and (b) being a complete hypochondriac, quite a few have come up over the years--but there are still many more to cover in future.

I suffer. I do. At this point, the pollen people tell me it's alder trees. But it's always something.

Alder catkins, via Wikipedia

I complained about this on Facebook last night with the status "Hay fever? Already?" and this led a former (British) student, now working in New York to ask:

They don't really say that here do they? More just 'allergies' in general.

I grew up with hay fever in Upstate New York, and much of my family suffers, so I'm used to hearing the phrase in American English. But, of course, I had to look it up.

I found on the Corpus of Global Web-Based English more mentions of hay fever in Britain than America and more of allergies in North America than in Britain. But allergies wins overall in both countries. Of course, allergies can refer to more than just pollen allergies, so that's not totally surprising.

(The darker the blue in these tables the more a phrase is associated with a particular country in this corpus. The raw numbers can't be directly compared because the sizes of the sub-corpora for each country differ, but the US and GB sets are roughly the same size.)



But looking at Google Books gives a different story:

This shows hay fever as peaking earlier in the US (around the 1940s) and later in the UK (1970s), but not more common in BrE than in AmE. It also shows the rise of allergies--earlier in the US than in the UK. I feel like I use allergies a lot these days because I'm never really sure what I'm sneezing out. But I do seem to be sneezing for most of the year.

So, it looks like the US is leading a change to allergies over hay fever, but this little exercise does demonstrate that a lot depends on the make-up of the data you're using.

If it's not hay fever I have, then perhaps it is THE DREADED LURGY. 

The most obvious difference in American and British theat{er/re} is the spelling, but on top of that there are a number of meaning differences. And then the meanings interfere with the spellings again. Much fun, but this is why I can't write short blog posts. Here we go...

the spelling

Theater is one of those American spellings that is attributed to Noah Webster.* But like most of successful American spelling reforms, it wasn't made up by an American. It was a long-standing spelling in England, and the predominant spelling at the time when the English colonies in America were first being settled. The OED says:

The earliest recorded English forms, c1380, are theatre and teatre; from c1550 to 1700, or later, the prevalent spelling was theater (so in Dictionaries from Cawdrey to Kersey), but theatre in Holland, Milton, Fuller, Dryden, Addison, Pope; Bailey 1721 has both, ‘Theatre, Theater’: and between 1720 and 1750, theater was dropped in Britain, but has been retained or (?) revived in U.S.

The word started as theatrum in Latin, and in French it lost its -um. The French pronunciation makes sense with the -re spelling, but the modern English pronunciation does not. However we pronounce that syllable, in whichever English accents we have, it is the same syllable that is spel{led/t} -er in words like butter or later. It's thus no wonder that English writers preferred the -er for some time (and Americans have preferred it for most times), since it is the more Englishy spelling, if by 'Englishy spelling' we mean (as I do) 'spelling that reflects English pronunciation'.

The fashion (for these things are fashion) of using the French spelling has won out in Britain for this and many other words of its ilk: centre, calibre, litre, lustre, sombre, etc. But the fashion is not consistent. Cloister, coriander, and disaster (among others)  have -re spellings in French from -rum spellings in Latin, but -er spellings in all standard contemporary Englishes. And then there's metre and kilometre but perimeter and thermometer, etc. Note, though, that despite their common Latin/Greek etymology (metrum), they have different vowels in the me syllable in BrE. American pronunciation of kilometer as 'kill LAH mitter' drives some Brits I know batty, as it obscures the relation between the met{er/re} and the kilomet{er/re}. They prefer 'KILL-o-meetah'. (I just tried to get Better Half to say it. He said 'kill LAH mitter' and explained 'I'm disarmingly transatlantic'.)

This particular difference has a lot in common with the -or/-our difference: variant spelling in early modern English, then American English settling on the more phonetic spelling, and British English settling on the more French spelling. I've more to say about that, but that's going in the book.

(By the way, I'm trying to get into the habit of listing BrE/AmE variants alphabetically. I may not always succeed, but it's why the ones in this post are listed in those particular orders. I'm also trying to alternate which goes first in British/American, US/UK, BrE/AmE, etc. )

the meanings

place where you...	What Americans usually say	What the British usually say
watch a play	theater*	theatre
watch a film/movie	(movie) theater*	cinema
hear a (university) lecture	lecture hall, auditorium	lecture theatre
have surgery	operating room; OR	(operating) theatre

The spelling foetus has no etymological basis but is recorded from the 16th century and until recently was the standard British spelling in both technical and non-technical use. In technical usage fetus is now the standard spelling throughout the English-speaking world, but foetus is still found in British English outside technical contexts. 

At the foetus entry, it just says: "Variant spelling of fetus (chiefly in British non-technical use)."

How true is this, that it's the accepted technical spelling in the UK? In The Lancet and the BMJ, it's doctors writing for other doctors. What about the rest of the medical professions? What about when medical types communicate with patients?

My first stop was the NHS Choices website, where the readers are would-be patients. A search for foetus brings up 27 hits, but fetus has 7. But, going the other way, foetal has 66 hits and fetal 82. What's going on?

I contacted the website to ask if they had a policy on this and they were extremely helpful (as the NHS always has been for me ♥♥♥). They put me in contact with their Head of Editorial Production, who sent me both a link to their style guide (which has fetus as an Americanism to be avoided) and his own document entitled 'Fetality', which he wrote when the Fetal Anomaly Screening Programme (so spelled) asked if the rest of the website could switch to fetal/fetus. In his paper he gives several arguments for retaining foetus/foetal, even on pages where it will conflict with the FASP program(me)'s spelling, but I think this first one is key:

NHS Choices is a ‘British English’ service and, as stated in its Editorial Style Guide, is bound to:

·       Write plain English

·       Avoid medical jargon and technical terms as far as possible

On the basis of those two points, if it is accepted that foetus is the general spelling and fetus the technical-medical, NHS Choices should use foetus.
(Bolton, Barry. 2014. Fetality. Internal document, NHS. Received with thanks from the author.)

Looking again at the o-less hits on the NHS Choices site, many of them seem to be in comments from site users—so the house style doesn't apply. Are they misspelling it, or do they know the 'technical' spelling? Why so many more fetals? Possibly because it's in the name of a lot of things, not just the FASP program(me), such as the 'Fetal Medicine Unit team at St George's Hospital', which is indeed how the hospital spells that unit's name.

It's an interesting mixture: the NHS website keeps the traditional British spelling in communication with patients in order to avoid technical language, but the hospitals and such seem quite happy to foist the technical spelling on patients in the names of units and program(me)s.

To investigate this a little further, I did a little survey in which I asked for UK medical personnel to tell me which spelling they would use in a work context: foetus or fetus, sulphur or sulfur and amoeba or ameba. F(o)etus was the only one that respondents disagreed about:

 
(The 'it depends' person gave that answer for every question and said they'd use the American spelling if they were writing to an American.)

I invited respondents to explain their preferences to me, but unfortunately only four did, and two of those used the space to tell me about words I hadn't asked about. The two relevant comments were:

I am an allied health professional who wouldn't use these words much in my work, but these were how I was taught to spell them at school. I've heard in the past that "foetus" is completely wrong, though I can't quite remember why and I write the word so infrequently that I wouldn't change my spelling of it anyway!

Homogenisation of the English language to accommodate American English is a pernicious assault on the richness and diversity of English usage. It shouldn't be tolerated!

Unfortunately, I didn't ask for demographic information beyond country of abode, so I can't see whether the people who prefer fetus are in professions in which they need the word more often than the ones who prefer foetus.

But my impression is that fetus/fetal seems to be something of a medical shibboleth in the UK now. Doctors use the e spelling and it sets them apart as 'in the know', and maybe they don't mind that the rest of the country goes about putting the o in it. All the better to tell who the truly educated are. I'd love to hear from people 'in the know' in the comments. Have I got that wrong?

And before I leave, a note about the other false etymological form that readers of The Lancet (well, at least one) have tried to change. Here's another letter to the editor:

SIR,-Spelling is a curious blend of phonetics, etymology, tradition, and nonsense ; we should take care not to let the last preponderate. Dr. Napier (Nov. 1) is to be congratulated on his attack on the absurd o which it is customary now to insert into fetus. I would like to raise support for a similar attack on the ph with which we generally mis-spell sulfur and the other words derived from it. Sulfur comes from a Latin word. Undeniably some Latin authors used the ph form, but there is good reason to think that this was a blunder, and most of the European languages that use the Latin root have not followed the erroneous spelling. The spelling sulfur was common in Britain from the 14th to 18th centuries, and this presumably explains its present day use in the U.S.A. It is in no sense an American innovation.  (Pirie, N.W. 15 Nov. 1952.The correct spelling of medical terms [Letter to the Editor]. The Lancet vol. 260, pp.987-8.)

The argument for sulfur seems not to have been heard—sulphur still rules Britannia absolutely.


Footnote
*It's a digraph, not a diphthong, but what do doctors know?

Two matters for this belated blog post:  Words of the Year nominations and the Untranslatables Month summary.

WotY Nominations
Long-term readers will know that we have (at least) two Words of the Year here at SbaCL, and nominations are open for both categories as of now:

1. Best AmE-to-BrE import
2. Best BrE-to-AmE import

The word doesn’t have to have been imported into the other dialect in 2012, but it should have come into its own in some way in the (popular culture of the) other dialect this year. I retain the editor's privilege of giving other random awards on a whim.

Please nominate your favo(u)rites and give arguments for their WotY-worthiness in the comments to this post. It might be helpful to see my reasoning on why past words were WotY worthy and other nominations weren't. Click on the WotY tag in order to visit times gone by.

Vote early and often! I plan to announce the winners in the week before Christmas.


Untranslatables II
Last year, as a birthday treat to myself, I declared October to be Untranslatables Month, which meant that I tweeted an expression that was unique to one dialect or another, in that its meaning was not captured by an expression in the other dialect. This year, I did it again, but made the job easier on myself by deciding not to tweet on weekends. Here's a summary of the 'untranslatables' I tweeted. In some cases, you can follow links to places where I (or someone) have discussed them in more detail.

• BrE lie-in (noun). The act of staying in bed later in the morning than usual. Sleeping not required, but lazing is. Example: 'The family was away, so I had a lie-in on Saturday as an early birthday treat.'  (AmE & BrE both have sleeping in for when one sleeps late.)

• AmE cater-corner, kitty-corner, catty-corner (regional variations), adj & adv, meaning 'diagonally opposite to'. Example: 'I live kitty-corner to the bordello'.

•  BrE builder's tea. Very strong (hot, of course), basic (i.e. not a special cultivar/flavo[u]r) tea with milk and lots of sugar. The 'lots of sugar' part is in most definitions for it, but some of my correspondents don't consider 'sweet' to be a necessary feature.

• AmE Nielsen rating. The television rating system that determines advertising rates, used figuratively as a measure of popularity. Example: 'When you give babies a choice of what to listen to, a kind of baby Nielsen rating, they choose to listen to mothers talking to infants' (from The Scientist in the Crib).

• BrE It's not cricket. 'It shouldn't happen because it's not fair/proper'. Occasionally heard in AmE too.

• AmE poster child. Figuratively, an emblematic case of something, esp. a cause. Originally a child on posters promoting a charity. This one has come into BrE--as untranslatables often do (because they're useful). In the US, it's especially associated w/the (US) Muscular Dystrophy Association, which is also responsible for the US's longest-running charity telethon. It's interesting how different diseases are 'big' in terms of fundraising in different countries...

Goodbye Untranslatables month!

Mark Liberman at Language Log has saved you from the rant that this weekend's post was to be. Oh, thank you, Mark! His post from earlier today does what needed to be done about journalist Matthew Engel's BBC piece "Why do some Americanisms irritate people?" (Yes, people.) The Language Log post starts by pointing out that only one of the first five 'Americanisms' cited by Engel is, in fact, American in origin. The only fault I can find with Liberman's piece is that it is not entitled "Why do BBC language features annoy linguists?"*

So, instead of a turgid rant about the BBC's continued knack for employing non-experts** to spout nonsense about language, I give you:

babies!!    kittens!!!   dental maiming!!!!

Today's topic was suggested by American-in-Scotland @dialect and inspired by her (first?) visit to a UK dentist. And, actually, it's rather a simple one. But just to make it more complicated, let me throw in a technical term I've just learn{ed/t}: deciduous teeth. Americans tend to call them baby teeth, and the more common term for them in BrE is milk teeth.  For those who like numbers, the Corpus of Contemporary American English has 100 baby tooth/baby teeth and 18 milk tooth/milk teeth.  The British National Corpus (which is much smaller) has 15 milk tooth/teeth and 3 baby teeth (two of which should actually be Babyteeth the name of an album by Therapy?) and no baby tooth. When I was a child in the US, I only knew milk teeth as a term for kittens' first teeth.

A milk/baby tooth isn't forever, of course, and before it goes it is a loose tooth, but in BrE one also hears wobbly tooth a lot. As far as my Grover is concerned, this is the only term for a loose tooth, since she was first exposed to the concept through the Charlie and Lola episode "I do not ever, never want my wobbly tooth to fall out".  Checking the corpora for loose/wobbly/wiggly tooth, we get 25/1/1 in COCA (AmE) and 3/1/0 in BNC (BrE).

The tooth fairy tradition is alive and well in both the UK and the US. Reading about how much money the tooth fairy tends to leave these days has left me depressed and fearful for a completely spoil(ed/t) generation.

And as a final public service before I go: Parents, if your (orig. AmE) teenager's dentist ever suggests removing a baby/milk tooth in order to "encourage" the permanent tooth to come forward, say "NO", or else your child may spend most of her most awkward years awkwardly trying to hide the big gap where a bicuspid should be. She will have no chance of being invited to (orig. AmE) the prom and you will endanger her respect for medical/dental/parental authority for evermore.
 


* Very occasionally, the Beeb does allow experts on (rather than just famous users of) language to grace its broadcasts.  For example, I was once on a program(me) about Scrabble. It was good fun, and I thought it great that they involved a Scrabble-playing linguist in the production.  But the best part? They spelled my name wrong.

** At one level, we're all experts on the language(s) we speak--in the sense that we use the language expertly. (This is for the most part subconscious knowledge--and science is only a very small way toward(s) understanding that knowledge.) There are a lot of accomplished users of language out there, and that's who the BBC likes to ask for opinions (God help us, not facts!) about language. I would like to point out that I am an accomplished user of time and space (taking up more of it every year!). Therefore, I would like to be considered for a central role in the BBC's next program(me) on physics.

I've been unwell (which is a very BrE way to put it, see this old guest post) a lot this winter, which seems to be the price one pays for procreating. They say that minor illnesses are good for developing children's immune systems, so I try not to resent the germs that infect poor little Grover. But I supposedly have a developed immune system. Shouldn't I be immune to some of these preschool bugs?  At least our norovirus kept us away from the preschool this week, when Erythema infectiosum has been going around. Or, as the note to parents said, slapped-cheek disease. Never heard of it? Neither had I. A little research showed that the more common nickname for it in AmE is fifth disease. That didn't really help either.  All in all, it sounds like a fairly pathetic entry into the childhood illnesses roster. (The child illustrating the infection's Wikipedia page looks like he's having a pretty good time with it!)

Before the stomach bug, it was a bad cold that had downed Grover and me. Both since my last blog post. (Better Half stays curiously well. Maybe I don't have a British-enough immune system.) Pity us!

In fact, you should pity any expat or immigrant with a minor ailment (or [BrE] the dreaded lurgy), because the one thing you want when you're feeling (chiefly BrE) grotty is the comforts of childhood--which are thin on the ground when one is separated from one's childhood by miles, oceans and passport controls, not to mention the decades. When I'm ill, I want two things, which, in my home culture, are known to have magical-medicinal properties: cold, flat ginger ale and chicken soup.

The ginger ale can be achieved. Saint Better Half only had to go to three shops before finding some.  Here, it goes by the BrE name American ginger ale, which I find amusing because (a) where I come from, we think of it as Canadian, (b) I can see no other kind of ginger ale for sale, so why do they need the adjective? One can only guess that it's to distinguish it from ginger beer, a much spicier drink, which is far more common in the UK than ginger ale (which in the UK is thought of as a mixer and not a drink in its own right). I can feel a tangent coming on. Whoops, here we go... Ginger ale consumption in the US is fairly region-specific. I come from the kind of place (the northeast) where it's a drink that you can buy cold in a single-serving bottle from a (orig. AmE) convenience store/(BrE) corner shop, but this isn't true throughout the US. And if there is a down-home 'American' ginger ale, then it's not the stuff that's used as a mixer. The Canadian mixer type is 'pale, dry' ginger ale (like this Schweppes or Canada Dry). But there is also 'golden' ginger ale, which is darker, heavier and gingerier (more like a traditional ginger beer). This is rarer in the US and even more regional. You'll know if you're in one of the regions for it if the names Vernor's or Blenheim mean anything to you (or a few others...see Wikipedia).  At any rate, it's the dry stuff that one wants if one's had a (more BrE than AmE) tummy bug. Because ginger is good for nausea, you know. It should have lots of ice, so that it gets watery and flat and rehydrates you without causing any more gastrointestinal upset.  But I live in England with a man for whom ice trays are one of those mysterious plastic things that come with a fridge yet have no clear connection to it, so I water mine down with water straight from the (BrE) tap/(AmE) faucet. Hey, I'm not well. I'm desperate.

Hm, over 600 words and I haven't even started to get to the point of this post. A record? Probably not.

The point is the soup.

See, we Americans know that chicken soup is the cure for the common cold. And, when you're recovering from a stomach virus, a nice chicken soup is a good second foray (after toast) back into the land of the digesting.  But, of course, you can't make it yourself. You're sick, after all. Stay in bed. And who wants to cook a whole chicken when no one feels much like eating? This is what the (orig. AmE) can-opener was invented for.  

It is perfectly possible to find 'chicken soup' in the UK. The problem is finding the kind that is good for a cold. Send your English (and vegetarian) husband out in the rain to buy a (AmE) can/(BrE) tin, and he will come home with five kinds of wrong before you send him out again whispering cock-a-leekie to himself.  The tins/cans of wrong will include various cream-based, coconut-based, curry-based concoctions--not what an ailing American soul needs.

The problem, I have come to understand, is prototypes.

So here comes the linguistics. Soup in either British or American English will include puréed and strained things like tomato soup, things with lots of cream in them, broths like the cock-a-leekie to the right, with pieces of meat and vegetable. All these things come within the boundaries of the category 'soup' in English. But categories have more than boundaries (and those boundaries are often 'fuzzy'. Yes, that's the technical term). Categories, as represented in our minds, also have peaks...or cent{er/re}s...or cent{er/re}s that are peaks. Pick a metaphor that works for you.  That cent(e)ry peak or peaky cent{er/re} is known as the prototype of the category, and a particular thing (like cock-a-leekie) is deemed to be part of a category (like SOUP) if it is close enough to/has enough in common with the prototype.  To quote a fine reference book on the matter:

According to one view, a prototype is a cluster of properties that represent what members of the category are like on average (e.g. for the category BIRD, the prototype would consist of properties such as ‘lays eggs’, ‘has a beak’, ‘has wings’, ‘has feathers’, ‘can fly’, ‘chirps’, ‘builds nests’ etc.).  Category members may share these properties to varying degrees—hence the properties are not necessary and sufficient as in the classical model, but instead family resemblances.  In the alternative approach, the mental representation of a concept takes the form of a specific, ideal category member (or members), which acts as the prototype (e.g. for BIRD, the prototype might be a representation of a specific robin or sparrow).

In other words, when deciding whether or not something belongs to the BIRD category, one measures its birdiness against some (possibly very abstract) notion of an ideal bird.  Now, it's reasonable to believe that there might be some room for dialectal variation in what the prototype of a particular category is. But we have to be careful here--it's not just a matter of what is more frequent locally that determines what the prototype is.  Chickens and ducks might be the most common birds down on the farm, yet the farmer will not treat them as if they are the prototype against which 'birdiness' should be judged--that hono(u)r stays with the birds that (BrE) tick/(AmE) check more of the 'bird' boxes like 'can fly' and 'chirps'.

As far as I know, not much work has been done on regional variation in prototypes. The only example I can think of is a small study by Willett Kempton (reported in John Taylor's Linguistic Categorization) on Texan versus British concepts of BOOT, showing that even though both groups considered the same range of things to be boots, there was variation in their ideas of what constituted a central member of the BOOT category, with the Texan prototype extending further above the ankle than the British one.

Though I've not done the psychological tests that would tell us for sure, I'm pretty sure that the American SOUP prototype is along the lines of this:

a warm broth with pieces of meat, vegetables, and/or starchy things (e.g. noodles, barley, rice, matzo balls) in it

And the English one is more along the lines of this:

a warm, savo(u)ry food made from vegetables and possibly meat that have been well-cooked and liquidi{s/z}ed

 These are not the definitions of soup, but the core exemplars of what belongs to the SOUP category, from which the 'soupiness' of other foods is measured. So, each culture has soups that don't conform to these ideals, but they nevertheless have enough in common with them (e.g. being liquid, considered food rather than drink, containing vegetables) to also be called soup.  The differences in the prototypes might have some effects on the boundaries of the category. So, for instance, since the English prototype has more emphasis on liquidi{s/z}ation, you'd expect the extension of the word soup to tolerate less in the way of (orig. AmE) chunky pieces than the AmE use of the word, which is stemming from a prototype that likes pieces and therefore will tolerate bigger ones (see point 3 below).

My experiential evidence for the differences in prototype are as follows:

1. American dictionaries (American Heritage, Merriam-Webster) explicitly mention the likelihood of solid pieces of food in soup, while British ones (Collins, Oxford) don't.
2. The soup of the day in English restaurants is very often a puree. In US restaurants, that's much more rare--the people want stuff in their soup.
3. Some of the things I have made and called 'soup' have been met with a puzzled "that's more of a stew, isn't it?" from the Englishpeople I've served it to.
4. Some of the most common soups in England are generally smooth: leek and potato, tomato (often 'tomato and basil', which to me is like eating pasta sauce with a spoon), carrot and coriander. Whereas American soups are often full of solid things: chicken noodle, beef and barley, vegetable (which brings us to...)
5. Order 'vegetable soup' in England and it will almost certainly be smooth. Order it in the US and it will almost certainly be a broth with diced vegetables. 

But this could be more rigorously tested, so I mention here that dialectal differences in prototypes might be an interesting area for a student dissertation project to cover.  (Are any of our second years reading this?)

Two more things to cover before I go. (I must be feeling better...I haven't collapsed in a heap yet.)

First, notice that I've been saying 'English' rather than 'British' when talking about the prototype differences. The two most famous Scottish soups, cock-a-leekie and Scotch broth, are broths with (more BrE) bits in them, so the prototype might be different up there.

Which brings us to broth. It's a word found in both AmE and BrE, but in AmE it basically means BrE (but also AmE) stock--that is, a liquid made by cooking things in water, then straining the things out. In BrE, it can be used to mean a stock with stuff in it (hence Scotch broth).  So, when I've expressed my longing for a more American-style soup to an Englishperson, I've been told "oh, you mean a broth". But AmE also has bouillon, which is again broth, but I'd call it bouillon if I were drinking it out of a mug (as I used to have to do in the days when I had to go on clear liquid diets a lot. I'm not the healthiest character), especially if I'd made it with a (AmE) bouillon cube (or powder), which in BrE would be a stock cube (or, more colloquially, an Oxo cube--the dominant brand).

I'm going to stop there and go to bed, trying not to think about how much easier my life would be if I could write this many words in grant proposals in an evening.  That way lies insomnia.

P.S. [Jan 2024]  Here's another American take on stock v broth, which doesn't work so well in BrE. From All Recipes: Soups and Stews magazine.