graft

To contact us Click HERE
JL in New York wrote recently with this observation:

Last week's Economist included an article ("Executive Pay: Money for Nothing?", in the Britain section) that begins:

Hard work builds character, and should be rewarded. But many Britons believe the link between graft and gain has broken down.

The word that struck me was "graft" -- in my AmE usage, it can only mean "corruption", not "hard work".  (Other than horticulturally.)

The link between graft (AmE) and gain has, sadly, not broken down, of course.

My first thought was that certainly AmE has the 'hard work' sense of graft, since the phrase hard graft is known there. But is it the case that AmE and BrE are divided by graft? 

The corruption sense of graft is listed in the OED as 'colloq. (orig. U.S.)'. Their first published citation for it is from an 1865 New York-based police gazette. West's Encyclopedia of American Law defines it as:

A colloquial term referring to the unlawful acquisition of public money through questionable and improper transactions with public officials.
Graft is the personal gain or advantage earned by an individual at the expense of others as a result of the exploitation of the singular status of, or an influential relationship with, another who has a position of public trust or confidence. The advantage or gain is accrued without any exchange of legitimate compensatory services.
Behavior that leads to graft includes Bribery and dishonest dealings in the performance of public or official acts. Graft usually implies the existence of theft, corruption, Fraud, and the lack of integrity that is expected in any transaction involving a public official.

This sense of graft may or may not come from the 'work' sense of graft; the OED lists them separately and doesn't have an etymology it trusts for the 'work' sense either. The 'work' sense is also listed as 'slang' and the first citation is in the phrase hard graft in 1853. An 1890 Glossary of Words of County Glouster lists it as meaning 'work', so perhaps it has dialectal origins there. Neither of these senses of the word, then, seems to be terribly old, but because they're colloquial and dialectal, they'll have unwritten histories going back further.

So, how well-known are the senses in AmE and BrE? A quick look at our (chiefly AmE) go-to corpora, the Corpus of Contemporary American English and the British National Corpus (via Mark Davies' interface) can give some indication.  First I looked at how much of the use of the noun graft in either corpus consisted of the phrase hard graft. For AmE it was only 6 of 640 (less than 1%), for BrE 28/145 (19%).

Taking a sample of 100 sentences containing a noun graft from each corpus, the use of particular senses breaks down as:

Sense	BNC (BrE)	COCA (AmE)
work	33	2
corruption	14	29
tissue	42	62
spade/shovel	7	0
proper name	0	6
??	4	3

So, the first thing to notice is that the 'work' meaning is indeed much more common in BrE. Both cases in the AmE sample were hard graft. Most of the 'work' uses in BrE were also modified by an adjective, but in addition to hard, there was honest, sheer, real, tireless etc. 

Second thing to notice: the 'corruption' sense is hardly unknown in BrE--but about half as frequent as in AmE. In both corpora, tissue grafts (on trees, skin, veins, bones, etc.) are the most common kind of graft.

Third, the 'spade/shovel' sense is particular to BrE. The OED defines it as 'a narrow crescent-shaped spade used by drainers', and its only citation is from a 1893 Worcestershire dialect glossary. One of the corpus examples mentioned it as a Norfolk term--these are not particularly close to each other, but who knows what was really happening dialectally 100 years ago or what changed in the 100 years till the BNC. (I mention shovel because of the American tendency to use the term instead of spade, discussed back here.)

And then there are more people or at least more famous people named Graft in the US than the UK (probably the former, it's a German name).

The ?? cases were those that I couldn't really tell the meaning of in the little bit of text I was given (e.g. in the BNC: His father quarrelled with the Colonels over some detail of graft). I didn't go to the effort of looking at the larger contexts, which might have helped. But what this 3-4% of ambiguous cases tells us is that even though graft has lots of meanings, they don't cause too much difficulty in understanding the language. The people who originally heard/read those seven ambiguous cases in full context probably had no problem with it at all.

So, my initial reaction 'Americans know about 'work' graft' might only (or particularly) be true of Americans like me who hang around a lot of British people and are able to separate the word from the phrase hard graft. And it just goes to show, you shouldn't trust your memories of words and meanings you've "always" known, as those kinds of memories just aren't very good. Can anyone tell me: is there a name for that kind of false memory/familiarity? It's the opposite of the Recency Illusion, but I've not found a particular name for the 'I've always said it that way' illusion.

Wait, wait! A little message to Arnold Zwicky, and I have the answer: the Antiquity Illusion. I feel like there should be a corollary of it for the effect when one moves from dialect to dialect--i.e. the 'it is old, but not old for you' illusion. The 'native-speaker illusion', perhaps.