In the first article in this series (Murphy Reference Murphy2016), I recounted Geoff Pullum's (Reference Pullum2014) dismissal of British-American linguistic differences as ‘mostly nouns’. From a theoretical linguist's position, nouns can seem simpler and less interesting than other parts of speech, since concrete noun senses are fairly self-contained. Compare a noun like cup to an adjective like big. You can picture a cup in and of itself, but in imagining big we need to think about things that could be big. And what we mean by big changes depending on which thing we are talking about. Since the meaning of cup does not have to interact with other words in order to get its meaning, investigating concrete nouns is a low priority for many linguistic semanticists. It can be ‘difficult to distinguish where the discussion of a noun's sense stops and where discussion of its extension (the things it refers to) begins’ (Murphy Reference Murphy2010: 149), and so that aspect of meaning is often left to philosophers and psychologists: What does love mean, really? or How do you know which things to call green?

If American-British vocabulary differences are ‘mostly nouns’, then should we be less interested in them as linguistic problems? In the earlier article (Murphy Reference Murphy2016), I argued that being ‘mostly nouns’ is no reason to lose interest in these differences. The nouniness of the differences follows from the fact that nouns make up the bulk of English vocabulary. In this article, I am going further and saying that the differences between nouns in British and American English are so pervasive and yet so subtle that they can teach us a lot about semantics and culture.

Famously, American and British sometimes use the same word forms for different meanings. This is most likely to cause communication problems if those words are in the same semantic field. While dummy has various meanings, it is easy for Americans to cope with the fact that dummy is the British equivalent of American pacifier because the other senses of dummy are unlikely to be relevant when someone says The baby's dropped her dummy. An American hearing this probably would not assume that the baby had dropped a mannequin or an idiot. They would ask, or try to figure out from context, what dummy refers to. But if we have a word with the same form and different meanings in the same area of vocabulary, it gets more confusing. The semantic field of clothing offers great illustrations of this. British pumps are flat shoes (they might be called ballet flats in America), but American pumps have high heels (they are court shoes in British English). Transatlantically, a jumper is two different pieces of clothing. The British meaning has the American equivalent sweater; the American meaning has the British equivalent pinafore dress. Vest, tank top, overalls, dungarees, pants, purse, suspenders all have related but distinct meanings in American and British English. (search Murphy Reference Murphy2006- for more on all of these). Because of their semantic closeness, they can be the source of inadvertent humour. My American mother went into fits of giggles whenever my English husband declared his desire to put on a jumper, and my American friend raised eyebrows on a British train when she declared that she had a stain on her pants (US = ‘trousers’, UK = ‘underpants’). Still, these differences are relatively straightforward to learn, since the meaning of the word in one dialect maps onto a straightforward equivalent in the other language.

But harder to notice, and therefore harder to avoid miscommunication with, are words that overlap in meaning so much that it is easy to fail to understand that different meanings are at play. Dictionaries often overlook these differences, and the miscommunications that result from them are not necessarily recognised as dialectal differences. I spent years thinking that British waiters were incapable of remembering a sandwich order till I realized the problem was linguistic. The way I was using the word sandwich did not make sense to them. I tell the whole sad tale in Murphy Reference Murphy2014. These differences are differences in the prototype structure of the meanings, and they can demonstrate different ways that meaning might be structured in the mind and, across a collective of minds, in a dialect.

The prototype view of meaning holds that meanings are organized around some idealized view of a ‘typical’ member of the category they designate (a prototype). So, for example, whether something is called a cup depends on the degree to which it matches the ideal of cup-ness and whether it echoes ‘ideal cup-ness’ better than it echoes the ‘glass-ness’ or the ‘bowl-ness’ ideals. This is in contrast to a ‘classical’ view of meaning where something would be called a cup if it had all the properties that all cups have, and if those properties together are sufficient to distinguish cups from glasses, mugs or bowls. The classical approach does not work because there may be no properties (a) that all cups have and (b) that are collectively sufficient to distinguish cups from other vessels. The prototype understanding of meaning holds that the things called cup are more or less close to the prototype, and so the word cup can refer to things that are extremely cup-like and those that have some of the properties of an ideal cup, but not all of them (like, say, a medicine cup).

One way of thinking about prototypes is that they are represented in the mind as a group of properties that the most typical members of the category have, for example for the category cup, the prototype includes properties like: ‘is a vessel’, ‘for holding liquids’, ‘made of pottery’, ‘hemispheroid-shaped’, ‘can be held in one hand’. Things that we call cup need not have all of these properties (I have a glass cup, you could have a giant cup), but they have to have a lot of them. The more of these properties an item has, the more comfortable we would be in calling it cup. Some of the properties might be more core, or necessary, to the meaning (e.g. being a vessel), while others might be more peripheral (e.g. being made of an opaque material).

It might not be surprising that, say, a prototypical English cup differs from a prototypical Japanese 茶わん ‘cup’, since they are different words that arose in different cultural milieus. But even in the same language, among the same words, prototype-based meanings can differ. Willett Kempton (reported in Taylor Reference Taylor2003) investigated Texan versus British understanding of the word boot, showing that the two groups varied in their ideas of what constituted a typical boot, with the Texan prototype extending further above the ankle than the British one. The Texan prototype fits better with cowboy boots, the British one with walking boots or army boots. Although the prototype is different, we end up referring to most of the same things as boots because riding boots, ankle boots, and (UK) Wellington/ (US) rubber boots (etc.) still have more in common with our slightly-different boot prototypes than with our shoe prototypes. But consider the use of football boot in British English, for what Americans would call a soccer cleat or soccer shoe. Historically, the football boot did come above the ankle, but it no longer does. In British, it still makes some sense to call these boots because they have a ‘family resemblance’ to the central members of the boot category. They are sturdy and have good traction, and lace up, like a walking boot or combat boot. In Texas, where the height on the leg is a much more important criterion for deciding what can be called boot, it seems bizarre to call below-the-ankle athletic footwear boots.

The prototype structure of meaning has various effects in language. One such effect is how nouns are modified. While bacon is a general (America, Britain and beyond) English word, we can tell that the default interpretation of the word is different in the US and UK because the most characteristic modifiers differ in the two places. The most common bacon modifiers in American English are turkey and Canadian, whereas in British English streaky is most common (GloWBE corpus, Davies Reference Davies2013). These modifiers tell us what is considered to deviate from the prototype, i.e. turkey bacon is not prototypical bacon because it's not pork. (It's also not widely available in Britain, and therefore shows up as characteristically American.) We don't have to say pork bacon because bacon made of pork is the thing we usually think of when we think of bacon - the plain description works for the prototypical meaning. Streaky and Canadian indicate that Britain and America both have a range of bacon types, but that different types are considered ‘typical’ bacon. In America, bacon from pork belly is the norm. We do not even have a special word for that type of bacon (we might say normal bacon or regular bacon), but the British specifically call it streaky bacon. In Britain, the more typical bacon is cut from the pork loin. Americans tend to call any loin-cut Canadian bacon (which is technically slightly different from the British cut, but it is a familiar term in the US). Ask for just bacon in the two countries and you will get different things, because the other country's preferred style of bacon is considered to be a deviation from the truest meaning of bacon. The other country's stuff is bacon-y enough to be called bacon, but not bacon-y enough for bacon alone to be a sufficient description.

The story of prototype variation is more complicated when we combine meat and bread, as for a sandwich, a burger or a hotdog. It helps to think of a prototype as a bunch of properties of varying importance. For the British, the identity of the breadstuff is more central to the meaning burger or hot dog than it is in American, where properties of the main filling lead the meaning. In American English, a hot dog is a frankfurter. It may be served in a roll, but it is the sausage that is the hot dog. If you put an Italian sausage (a kind of spicy sausage that is popular in the northeastern US) or a bratwurst or a kielbasa in a long roll (for these are all sausages that Americans would serve in a roll), it is not called a hot dog. In British English, one can order a hot dog and find various kinds of sausages in the bun. What is important is that it is a sausage served inside a roll.Footnote ¹

Similarly, in British English, the word burger can be applied to things that would not be called burgers in the US. What is important in British English is that it is a flat thing served in a round roll. What is important in American English is that the flat thing is a patty formed from a ground (=British minced) meat or a similar substance. In the UK, one can order chicken burgers that amount to a cooked chicken breast in a bun. In the US, these tend to be called grilled chicken sandwiches. If I read chicken burger in the US, I would expect it to be a patty made of ground chicken. Though we do not tend to find such chicken things on menus, we do find turkey burgers made of ground turkey meat. Similarly, British restaurants offer portobello mushroom burgers and halloumi (a type of cheese) burgers of completely not-minced ingredients—but in burger-style rolls. Americans would call these sandwiches rather than burgers. In fact, for Americans even hamburgers-in-rolls can be called sandwiches. Perhaps thinking of burgers as sandwiches affects other non-linguistic behaviours. Put a large burger in front of an American, and they may well cut it in half and pick it up, like they would for a more prototypical sandwich. Put it in front of an Englishperson, and they will generally attack it with knife and fork, rather than attempting to eat it in a sandwich style.

The portobello-mushroom-in-a-bun is not called sandwich in Britain because, again, the bread details are more important in the UK. In both countries, prototypical sandwiches are made with sliced bread. But for the British this is a more core property of sandwiches than it is for Americans. Put ham and cheese on a roll, and it has to be called a ham-and-cheese roll in British English, rather than a ham-and-cheese sandwich. Put it on a baguette and it is called a ham-and-cheese baguette. Put it on a bagel and it is a ham-and-cheese bagel. In American English, we can call them all sandwiches, and it makes sense to say things like I'd like my ham-and-cheese sandwich on a bagel, please. When I say such things in the UK, I am mocked with You want a sandwich on top of a bagel? This use of on to specify bread type is particularly American.

Many food words seem to have differences in meaning that can be attributed to prototypes. Another favourite example of mine is soup. In both dialects, the word includes both smooth purees and broths with chunks of food. But the British prototype for soup is smoother than the American prototype. Most of the time this would have no effect on communication as tomato soups are puree-types in both countries, and mushroom soups may or may not have some chunks. But if you order a vegetable soup in the ‘other’ English, it may not be what you expect. In Britain, the default is a puree. In America, you would expect to receive a broth with chunks. I had never experienced a smooth, opaque chicken soup till I moved to the UK, but that norm is firm enough that the prototypical American chicken soup is often called chicken broth in Britain. Chicken broth in the US means what the British mean by chicken stock. All of these food examples are discussed in greater detail in Murphy Reference Murphy2006.

My final example of transatlantic attention to different aspects of meaning is twang. I went for some time feeling slightly annoyed when English people would say that I had an American twang. I wondered that they could not tell the difference between my American dialect (from the northeast of the country) and those American accents that I would consider to have a twang, such as a cowboy-sounding accent. I finally figured out where I had been going wrong in my understanding when I read about an actor having a ‘Scottish twang’. Scottish people sound very little like cowboys in my mind, so I was forced to wonder: what could Scottish twang mean in British English? It was then that it started to occur to me that twang in British English does not just mean a particular kind of accent as it can also mean some hint of any accent. Once I looked, I could find northern twangs, German twangs, Welsh twangs, and Cockney twangs in British English. I conducted an online poll asking whether have a twang means ‘having a definite regional accent’ or ‘having a slight regional accent’. American respondents went for ‘definite’ 85% of the time, but British respondents were split between ‘slight’ and ‘definite’. When asked which ‘definite’ accent it was, the American south and west were mentioned a lot, as was Australian.

Dictionaries note the two possible interpretations of the word twang, i.e. (1) an accent that is perceived as ‘nasal’ and (2) any regional accent, but they do not note that the second of these is far more common in the UK, where accents give much more social and regional information than in the US. ‘It is impossible for an Englishman to open his mouth without making some other Englishman hate or despise him’, wrote George Bernard Shaw (Reference Shaw1916). In Britain, the perceptibility and ‘non-standardness’ of the accent is the key point in the use of the word twang, whereas in American, the key point is whether the sound of twang is a good imitation of the accent.

These differences are not generally noticed as British and American English dictionaries. Most dictionaries include both senses of twang without indicating that you are much less likely to hear the ‘any accent’ sense in American. Where the differences are more subtle, sometimes the differences come out in the definitions, if you compare them across countries. The Oxford English Dictionary (online) defines a sandwich as being ‘composed of two thin slices of bread’, whereas the American Merriam Webster online has ‘two or more slices of bread or a split roll’. American dictionaries (American Heritage, Merriam-Webster) explicitly mention the likelihood of solid pieces of food in soup, while British ones (Collins, Oxford Dictionaries) do not. Comparing American and British dictionaries can give us some insight into the subtle differences, but it does raise the question: should the dictionaries let us know about the very subtle differences? This may depend on whether the dictionaries purport to be dictionaries of global English or of a particular variety. The more dictionaries are accessed via the internet, the harder it can be to tell where the dictionary is from.