Technically speaking, the anatomical structure the consultant was looking at was the abdomen, which is schematically divided by doctors into a three-by-three grid. From top right to bottom left the squares are named: right upper quadrant, epigastric, left upper quadrant, right flank, umbilicus, left flank, right iliac fossa, hypogastric, left iliac fossa. The organs are clustered in each square – the liver and the gallbladder reside in RUQ, for example. When a patient has RIF pain, you know to think of appendicitis.
But what doctors in England haven’t quite solved yet is how I should ask you to show me this space. The medical word, “abdomen”, is not used by many people. But “stomach” is factually wrong. (Your stomach – LUQ – is the springy bag in which your food first lands to be churned before it continues on through your intestine; most “stomach ache” is felt nowhere near the real stomach – what most people point to is their umbilicus, underneath which lies the small bowel.) “Belly” is American. “Tummy” is a nursery term, but English doctors use it in parallel with the anatomical terms. You learn to say “poo” for faeces, too. But if questions such as “Have you had your bowels open?” and “Have you passed any stool?” are met with blankness, there is not much alternative.
Belly is American? That didn't sit right with me, as if a doctor asked me to show her my belly, I'd find it very strange--though I might suspect that the doctor spoke a different dialect from mine. I use tummy (or the anatomically-incorrect stomach). To me, belly particularly signals a round tummy--hence (orig. AmE) beer belly. Babies have bellies, Buddha statues have bellies, I have a belly--but let's not go there. One also hears people saying, typically while pinching more than an inch, I'm getting a belly. In all these uses, it's not the same as tummy or non-technical stomach. It describes a paunch (which, incidentally, used to just mean 'abdomen', without the negative connotations), but with rounder connotations.
The doctor writing in the magazine is not alone in this assumption that belly is American. In fact, this amateur (and very defensive about it, while not trying very hard*) BrE/AmE word-lister assumes that tummy is exclusively BrE.
But, while I had my doubts about the BrE/AmE tummy/belly divide, I've often heard tummy-button in the UK (though mostly from antipodean yoga/Pilates instructors), and never in the US. So, I decided to check it out.
First, the history. Belly goes all the way back to Old English, where it originally meant a bag, but from at least as early as the 13th century, it's used to mean a human or animal stomach and from at least the 14th century, it's used for the abdomen. So, it certainly did not originate in AmE. Tummy (a baby-talk simplification of stomach), in contrast, is only seen in print from the 19th century.
Next, the usage. I looked up stomach, belly, and tummy in British and American corpora of writing and speech, and calculated the percentage of the total number of instances of any of those words that was represented by any one of those words. Here are my results:
From this we can tell (a) belly is used quite a bit in BrE as well as AmE, and tummy is more frequent in BrE than AmE. I don't think this can just be due to differences in formality across the corpora, since if the AmE corpus had more formal writing in it, we'd expect the stomach percentage to be higher.
Now, within belly in either corpus, many instances do not refer literally to human abdomens. There are lots of instances of idioms like in the belly of the beast or a fire in one's belly. There are also lots of belly-dancing. To see whether the AmE bellies might be more specifically fat tummies, I looked at paunch, to see if AmE didn't need it as much--but that's not the case. In both corpora, paunch occurs between 5 and 6 times per million words.
As for the hypothesis that belly is more 'round', I note that I and my UK friends do say I'm getting/I've got a bit of a tummy, but looking in the corpora, there are a couple of instances of get/getting/got a belly in each corpus but tummy only occurs in that context in the AmE corpus. So, in BrE, belly is used for the 'rounded abdomen' meaning, just as in AmE, and AmE uses tummy in that context too.
What about bellybutton and tummy-button? OED has the former dated to the 19th century, but the latter only in the mid-20th century. COCA has zero instances of tummy-button, tummybutton, or tummy button. BNC has just one. Belly(-)button seems to be the default colloquialism for 'navel' in either dialect. In a strange turn of orthography, the joined-up bellybutton is by far the most common spelling in the BNC, but two-word belly button is very strongly the favo(u)red spelling in COCA. This is in contrast to another observation that I've made here, that AmE joins up compound words in writing more readily than BrE does. In that post, I noted that the Shorter Oxford Dictionary recommends pot belly, while the American Heritage Dictionary likes potbelly.
I'm writing this in the Helsinki airport, so am limited to dialect resources that are on-line--and I'm not finding them to be helpful at the moment. While the evidence does show American English using belly much more than modern British English, I still have the feeling that there is some regional variation at work here, since it's not a word that I would use for a human abdomen outside the 'paunchy' and 'baby' experiences. But that's my western New York State perspective. How would you feel if your doctor asked to look at your belly? (Don't forget to tell us where you're from!)
*In discouraging corrections to his list, he says 'life is too short to worry'--about accuracy, presumably. Life is also too short to spend on writing word lists without caring to do it right, I'd say.