A look under the bonnet in the October 2018 update

A look under the bonnet in the October 2018 update

It’s a routine feature of words in the English language that over time they begin to demonstrate a number of figurative or extended uses of the original meaning, which then themselves become established meanings of the word. Sometimes these uses supersede the original core sense (consider, for example, throw, which was used in Old English only in senses to do with twisting or turning something).

The noun bonnet is a good example of this phenomenon. Although the original sense, first attested at the start of the 15th century, is still familiar enough, denoting an item of headwear (especially a type of hat worn by women, although there are particular men’s hats, such as the traditional Scottish tam o’shanter, that were once, or still are, referred to as bonnets), many extended and metaphorical uses have also arisen during the word’s history.

As an aid to understanding the sequence in which these uses arose, the OED entry places them together in a single section or ‘branch’ headed ‘Extended uses’, following a branch containing the various headwear senses.

In most cases the motivation for the application of the word bonnet is immediately clear. Principally we are looking at things that resemble a bonnet in either form or function. In the former category we have sails, fortifications, flowers, and molluscs whose appearance is in some way thought to resemble that of a bonnet. Also in this category is one rather surprising-looking sense, referring to the reticulum or second stomach of a ruminant, but here the motivation is explained in a quotation from 1688:

“The second Stomach, by Aristotle called κεκρύϕαλος, and by the Latins, Reticulum, by reason that it has some Eminencies which do represent a little Net, which has made this Stomach to be called Bonnet, because that this Net resembles the lace Bonnet, in which Women heretofore inclosed their Hair.”

As the quotation implies, the original meaning in Latin of reticulum itself is ‘small net or meshwork bag’ and one can see a very similar appearance-based metaphor at work in the ruminant-stomach-related senses of both words.

The second category, encompassing things with a function likened in some way to that of a bonnet, gives us a multitude of things whose purpose is to cover the top of other things, from valves to chimneys to lamps to lighthouses. Chief among these senses, at least outside of North America, is the covering for the compartment containing a car’s engine at the front of the vehicle (or the covering for the luggage compartment, if you happen to own a rear-engine Volkswagen Beetle). In North America, of course, an extremely similar figurative extension of a different word, hood, performs the same job.

It is not until we get to the last sense in the branch, sense 13, that we are presented with anything puzzling. Between 1831 and the start of the 20th century bonnet was used with the meaning ‘A person employed as a decoy or accomplice, such as a pretended player in a gambling house or a fictitious bidder at an auction, who colludes with the proprietor or auctioneer to encourage other gamblers, buyers, etc. ’  Why? For what reason is someone who pretends to be someone else, as part of a shady scheme to part others from their cash, likened to a hat?

There are four main possibilities.

1. The white bonnet

Once one becomes aware that there is an obsolete Scottish term, white bonnet, dating back to 1760, that has the meaning ‘a fictitious bidder at an auction’, it is very tempting to see our bonnet as a shortened form of this, a product of our natural preference for brevity in speech. But there are a few problems with this. Firstly, white bonnet is a very determinedly Scottish term, to the extent that Thomas Carlyle, writing about it in 1867, goes as far as to call it ‘intensely vernacular’ and limits it to one particular region:

White bonnet..is the Annandale name for a false bidder merely appointed to raise prices, works so for his five shillings at some poor little Annandale roup.” [Roup is a Scottish and northern English word for a public auction.]

Meanwhile, all the early evidence for bonnet used sans white centres around London, more specifically London gambling houses, which poses a further problem for the white bonnet theory. All early use of this sense of bonnet centres around the ‘pretended player in a gambling house’ side of things, which, while conceptually in the same area as the fake auction bidder, adds a semantic difficulty alongside the geographical one; you would have to accept that the term was simultaneously shortened and reapplied in a different regional and social context.

Finally, even if there are still takers for the white bonnet theory, it does little to actually solve the mystery, instead just pushing it one stage backward. Why give a fictitious auction bidder the name of a type of hat, and what’s more, a hat of a specific colour? John Jamieson, in his 1808 Etymological Dictionary of the Scottish Language, has a stab:

“This metaph[orical] term seems to signify a marked person, or one who deserves to be marked… The term is most probably a literal translation of a Fr[ench] phrase, the meaning of which is now lost. For the expression, Bonnet blanc, ou blanc bonnet, is still proverbially used to denote things that are exactly alike, and which may be indifferently put one for the other.”

We could kindly call this fairly speculative, and there are certainly enough holes in it to make us hesitate to pronounce the ‘lost French phrase > white bonnet as fake bidder > bonnet as fake gambler’ sequence the obvious winner of the bonnet conundrum competition.

2. The ‘cheat’ verb theory.

Jonathon Green, in his 2010 Green’s Dictionary of Slang, relates both the gambling cheat and sham bidder senses back to a use of bonnet as a verb, with the meaning ‘to cheat’. This feels plausible, but unfortunately the earliest evidence found for this verb is from 1843, over a decade later than bonnets begin to plague London gambling houses as nouns. However it is of course easily conceivable that the verb was in spoken use much earlier than the first time it found its way into print, though the same could be said of the noun. Either way, a verdict of ‘not proven’ has to be returned.

3. The ‘concealment’ theory.

OED gives at this sense an earlier, 1819, quotation in square brackets (this is the OED’s standard way of presenting quotations which do not themselves reflect the sense described by the definition but which may have some relevance to its development). From a glossary of ‘flash language’ (i.e. thieves’ slang, or cant) at the end of the Memoirs of swindler and pickpocket James Hardy Vaux, it gives this definition of bonnet:

“A concealment, pretext, or pretence; an ostensible manner of accounting for what you really mean to conceal.”

This is clearly a different, and much more abstract, meaning, but it does seem to provide a link to the second, ‘function’, category of metaphorical resemblance to a bonnet mentioned above: a figurative covering over what someone is really up to. It’s then feasible that the meaning was transferred to a person pretending to be what they are not.

Feasible, but far from certain. Vaux’s definition appears to be the only recorded instance of this meaning of bonnet, with the exception of a few later dictionaries or glossaries which repeat Vaux’s text word for word: evidence for natural use in context, such as that which abounds for the fake gambler or bidder sense, is completely lacking. And although he was once active in criminal circles in London, by 1810 Vaux had been transported to Australia (for the second of what proved to be three times) for stealing some jewellery, and it is not completely clear how far his work covers true London underworld slang and how much of it actually reflects the words in use among transported convicts in Australia (although there was of course a fair amount of cross-fertilization between the two); the glossary originally sprung from an earlier short dictionary written by Vaux, for the use of magistrates in the penal colonies wishing to decode what the convicts were talking about

4. The ‘blinding the eyes’ theory.

This one has the support of a couple of major late-19th-century dictionaries. Both the Century Dictionary (1889) and J.S. Farmer and W.E. Henley’s Slang and its Analogues Past and Present (1890) make variations on the same bald statement, unqualified by any doubt or uncertainty. This from the Century Dictionary:

“So called because such a person figuratively bonnets or blinds the eyes of the victim”

The reference here is to what is covered in OED as sense 3 of the verb bonnet: ‘To obscure the vision of (a person) by suddenly or unexpectedly pulling or pushing his or her hat down over the eyes’. Again the metaphor relates to the function of bonnets as a covering, and more importantly, much more closely to the fact that bonnets are a type of hat. Unlike the verb meaning ‘to cheat’ suggested in theory 2, this one is first attested in 1828 and so is able, chronologically speaking, to support the conjectured figurative conversion to a noun, albeit only by virtue of being three years earlier. It is also found in the right geographical location: the first quotation is from George Smeeton’s book Doings in London, and furthermore describes a situation in which someone’s winnings from a gambling game are stolen after their hat is pulled down over their eyes.

Even so, that figurative conversion to a noun, relying as it does on both a change in part of speech and an acceptance that a hat-based prank is the metaphorical source, feels a bit of a stretch, and it is tempting to imagine that these two august 19th-century works were indulging in a bit of after-the-fact rationalization (aka ‘guessing’) with regard to the origin of this sense.

In the end, the strongest we can say of any of these theories is ‘perhaps’, alongside an acknowledgement that the ‘blinding the eyes’ theory probably has the most going for it. More, certainly, than what can only be regarded as something of a shot in the dark, given by Ernest Weekley in his 1921 Etymological Dictionary of Modern English, and also quoted in Eric Partridge’s Dictionary of Slang and Unconventional English (1937):

“In the sense of accomplice, etc…there is perh[aps] a reminiscence of F[rench] deux têtes dans un bonnet, ‘hand and glove’.”

(Hand and glove is a rarer variant of the phrase hand in glove, ‘intimately associated or connected’.)

In fairness, however, even this too includes the key word which has to apply to all the suggested origins of this sense: ‘Perhaps’.


The opinions and other information contained in the OED blog posts and comments do not necessarily reflect the opinions or positions of Oxford University Press.