display | more...

A cranberry morpheme is a morpheme that only appears in one word, and whose meaning by itself is unclear or unknown to the everyday speaker.

They are named after the word 'cranberry' which is an example of the phenomenon: The morpheme 'berry' indicates that the word refers to a kind of small fruit (or whatever a berry is). However the morpheme 'cran' by itself is dead and doesn't mean anything, as opposed to, say, the blue in 'blueberry' or even the goose in 'gooseberry'.

The writeup given above (explaining that the "cran" in "cranberry" doesn't mean anything) is fairly typical of the explanation given for cranberry morpheme in half of the linguistics classes in the world, which just goes to show than in any group, almost 50% of them will be below average. Students and professors alike generally take this explanation as Gospel and don't bother to spend the 5 minutes it would take to consult the OED and discover that cran is directly related to crane (the bird), and that cranberries are really crane-berries - berries that grow in marshes habituated by cranes, who love to eat them, especially as they come to fruition just about the time that the cranes return exhausted from their annual trip south to Germany, where I'm told they deliver babies. The original form of the word was the 17C Low German "kran-beere", and related forms are still in use in countries that have marshes, cranes, and cranberries.

(Incidentally, compounding the confusion, cranberries in the US are not the same plants as the marsh cranberries from Europe. So in ignorance of the marsh/crane connection, a folk etymology has sprung up in the US to explain that cranberries were so named because their flowers looked like the head of a crane.)

This ignorance doesn't stop here. Not even knowing their own field well enough to realise that the non-free bound dead morpheme concept had already been named, the presumably above-average other half of the world's linguists decided to call the same phenomenon "huckleberry morphemes", explaining that there is no such thing as a Huckle. Well, my huckleberry friends, there is; huckleberry is derived from hortleberry which in turn came from whortleberry. Whortle comes from Old English "Wyrtil" meaning a shrub or small bush, which in turn comes from the same root as, well, "root", in the form of wort or wyrt.

Here's an interesting coincidence. An alternative name for the cranberry is the "marsh whart" - with "whart" being the very same as in whortleberry. Both are therefore berries which grow on low roots or shrubs. I'll leave it to another day to talk about the coincidence, or lack of it, in the terms related to words (worter, in German, cf root) such as roots or stems all being derived from plant metaphors.


Update
In the post below, gritchka explains that the significant determiner of a cranberry morpheme is not that the root is unknown but that prefix is currently not recognised. Putting aside the fact that this is a 'proof by personal ignorance' (not quite the same thing as argument from ignorance) - he may not know that cran is related to crane, but I did and I suspect many of our other readers did, especially our German and Scandanavian ones - I have a Norwegian friend who recognised the word immediately as the same as tranebær, with 't' being a consonant shift for 'k'; tranbär in Swedish - I'll confess that I didn't fully explain the history of the expression "cranberry morpheme". It was originally coined by Aronoff in 1976 and meant what I explained above - that the morpheme was meaningless and no root was known. In time some linguists realised that cran (and huckle) did have meanings and were a little embarassed by this name for the phenomenon (note, I don't dispute the phenomenon, I'm just questioning the choice of name for it) and slightly revised their definition to be merely that the morpheme now exists in only one word. I guess they could be excused for not knowing that 'cran' is still the Lallans Scots word for crane (the bird; it's also used for another bird which I think is actually a swift, from memory), and in (relatively) common use in such expressions as 'cowping the cran' (which I suppose is the Scottish equivalent of cow-tipping). And how about 'cranage'? (Ask a longshoreman if you don't know what it means) OK, it is derived from 'crane' but the morpheme is the same 'cran' as in cranberry which is merely the more recent spelling of craneberry, so "cranberry morpheme" is an inappropriate name even using the modified definition.

However, my point was that the term is explained in linguistics classes using the original definition, which I gave above. I don't know if Gritchka has ever attended a linguistics class, or was lucky enough to attend one taught by someone using the revisionist definition of cranberry morpheme, but a quick net search shows that the vast majority of linguistics classes (or at least those whose lesson plans are online) still teach the definition of cranberry morpheme which says that cran is a meaningless prefix. I can assure him for personal experience and discussions with linguistics professors that they truly believe that 'cran' is literally a meanless collection of phonemes that no-one could possibly make sense of!

I attach a very small sample of pages from linguistics classes below; feel free to check for yourself with Google where you will find that over 90% of the references are similar; only a very few will quote the less permissive rule about the morpheme occuring in only one word, and none that I have found admit that cran means crane.


1) From Ohio State linguistics class:
“cranberry morphemes”: morphemes that have no constant associated meaning
  • cranberry, huckleberry, boysenberry
  • permit, commit, submit
  • receive, perceive, conceive

2) From Lancaster University:
Cranberry morphemes

If a word contains ambiguous morphemes it can be difficult to work out its internal structure. Consider the word cranberry.

Berry is obviously a morpheme, but is difficult to know what to make of cran-. Words such as elderberry and waxberry can easily be analysed as Noun+Noun compounds: elderberries come from the elder tree, while wax berries come from wax-myrtle. However, there is no such thing as a *cran tree or bush, and the form *cran does not appear by itself anywhere in the English language, or even as a bound morpheme.

Because of this problematic case, cran- is treated as a bound (noun) root morpheme which is restricted to just one word (cranberry). Morphemes like this are referred to, after the most famous example, as cranberry morphemes.


3) From the Lexicon of Liguistics at Utrecht:
cranberry morpheme

MORPHOLOGY: a type of bound morpheme that cannot be assigned a meaning nor a grammatical function, but nonetheless serves to distinguish one word from the other. EXAMPLE: the English word cranberry seems morphologically complex, since it must be distinguished from words such as raspberry, blackberry, and gooseberry. Still, cran has no meaning and does not function as an independent word: cranberry is the only word in which cran appears. The existence of cranberry-morphemes plays a role in the discussion whether morphology is word based or morpheme based (e.g. Aronoff 1976).

The preceding writeup is missing the distinction between synchronic and diachronic explanations. Linguists know the etymology of most or all these words: they know where the element 'cran' came from, but that's not the same as what it means now.

Almost all berry names in English have meaningless first elements: cran, huckle, whortle, bil, boysen, logan, tay, /ra:z/ in raspberry, /gUz/ in gooseberry, and even 'straw' in strawberry. None of those is an obvious element the way we have in blackcurrant or oakapple. Unless you look up the dictionary to find their etymologies (and probably all of them have known etymologies), then any guessing you do as to why it's called that is unreliable folk etymology.

What have strawberries got to do with straw? Were they traditionally grown in straw? I look up my dictionary now and find it's possibly from a resemblance of the stalks. Hmmm.

There are numerous meanings of 'bill' but none obviously give rise to an explanation of 'bill-berry'. The spelling bilberry suggests there's something else going on. I could look up the dictionary again...

There's a cran that's a measure of herrings but I can't see how that helps.

The /ra:z/ of raspberry and the /gUz/ of gooseberry don't mean anything at all, but the spelling gives us a clue. Though, admittedly, not a huge one. If raazberries are actually rasp-berries, it's because they're, um... rough like a rasp? And a goose-berry... hmm... But if they didn't retain archaic spellings, we wouldn't even have those clues to meaning.

Now I happen to be fairly sure that loganberries are named after someone called Logan, and boysenberries after one Boysen; and there's a place in Scotland called Tay so perhaps that's where the tayberry was created; but the point remains that, however well we know their origins, all the initial morphemes are meaningless in synchronic terms. This is what the concept of 'cranberry morpheme' is about. Any native speaker recognizes the 'berry' part, and all the parts in redcurrant and blackcurrant, and understands how they're put together. But the cranberry morphemes don't have any clear meaning of their own, unless you delve into their histories.


The previous noder has now replied, but the only point they might have successfully made is that Aronoff made a mistake when coining the word. Possibly this is true. Though...
The cran of cranberry is often called a cranberry morpheme. Originally, the term referred to a morph which occurs uniquely in only one word (such as cranberry, though this word is not quite so unique after the invention of cranapple juice). However, the crucial property of such an element is that it can’t be assigned a meaning of its own.

http://privatewww.essex.ac.uk/~spena/372/372_ch3.htm

The rest of the matter the noder quotes correctly illustrates the (current) use of the term, without suggesting that any error has been made, or that etymology is pertinent.
Cranberry. Also craneberry.

app. from some LG. source; cf. G. kranichbeere, kranbeere, LG. krônbere, kranebere, etc. (all meaning craneberry).

- from The Shorter Oxford English Dictionary,
1933, 3rd ed. 1944

Crane, a bird.
...
   cranberry, (Low G.) Modern; from Low G. kraanbeere (Berghaus), G. kranbeere, lit. craneberry ; cf. Dan. tranebær (from trane = krane, as above); Swed. tranbär.

- from W.W. Skeat,
Concise Etymological Dictionary of the English Language,
Oxford, 1887, new ed. 1911

'as above' refers to the same change k > t in Danish, Swedish, and Icelandic for crane the bird.

I think that disposes of the possibility that linguists weren't aware of the etymology of 'cranberry'. The OED and Skeat are the two standard works. Aronoff must have been dead embarrassed if he did overlook it; though I really have to doubt whether he was the one who misunderstood. But the etymology isn't relevant. Native speakers don't have etymology in their heads: it's not part of their competence. This isn't an argument from personal ignorance, because 'competence' is a technical term in linguistics, the structured knowledge that all speakers have of how to make well-formed utterances. The history of a word doesn't come into it at all.

A morpheme is a meaningful element that can be used to make words: so un and break and able in unbreakable. A bound morpheme is one that can't exist as a word on its own, like un- and -ness. There's a word ness meaning a cape or headland, but it's a different morpheme from -ness the suffix; their meanings are quite different. A productive morpheme is one that can be used to make new words: un- and -ness, for example. We know what they mean, so we know what a new word containing them would mean.

Words perceive, receive, conceive and permit, remit, commit are unusual in that they clearly form a pattern and are internally divisible, but the bound morphemes -ceive and -mit have no assignable meaning in English. The native speaker of English can't extract, hasn't got it in their competence to extract, a common element for forming new related words.

This has nothing to do with etymology. No-one can imagine that linguists don't know the familiar Latin words these come from. Their separate meaning is quite straightforward in Latin. But English speakers aren't Latin speakers. This isn't personal ignorance, it's the internal nature of the language faculty.

The morpheme cran- in cranberry is like -ceive but occurs in only one word. So too are logan, tay and the rest. It doesn't occur in crank or cranium, and it also doesn't occur in crane or cranage. The word crane isn't composed of cran plus some extra bit that changes its pronunciation. In a native speaker's competence, they aren't related at all.

Actually a gooseberry morpheme is even more interesting. The phonetic chunk /gUz/ not only doesn't occur in any other word, the syllable doesn't fit the sound pattern of English (except in the North of England, where buzz and does rhyme with it). There are no words like /gUz/, /gUv/, /gUð/ with the short vowel of 'look' and a voiced fricative.

The reason cran- and /gUz/ are noticeable as morphs is that we can separate off the known element /bri/ (or /b@ri/ or /beri/) as berry. There are other words like breakfast, shepherd, and cupboard where the spelling tells us their history as two morphs, but synchronically they are as indivisible as curtain or haggard.

Log in or register to write something here or to contact authors.