Green’s Dictionary of Slang News

Everyone’s A Critic: Redeeming Cornelius Crowe

James Lambert

Spare a thought, gentle reader, for the poor maligned Cornelius Crowe, an unsung hero of Australian slang that has been abolished to the realms of anonymity (have you ever heard of him?) at the hands of misdirected critics, carping reviewers, and various other detractors.

The Crowe I crow about was an Australian police officer of Irish descent, born in Tipperary in 1853 and passing from this world in the seaside Melbourne suburb of Sandringham in 1928. His claim to lexicographical fame – such as it is – is his 1895 publication An Australian Slang Dictionary. Sadly, I don’t have a copy, nor have I been able to locate a copy for sale. Barry Humphries – creator of Bazza McKenzie and one of Australia’s foremost slang promoters – has a copy, which clearly he cherishes as it has been with him since his late teens ‘despite the depredations of time, travel, and multiple divorce’, and, of course, the GDoS reference library possesses a copy of this rarity. If I do ever get a copy for myself, you know where it’s going – straight to the pool room!

Thankfully, modern technology allows one to get up close and personal with the good Cornelius’s dictionary through a PDF version, and I have scrutinised its 105 pages with a fine-toothed comb in order to come to an understanding of its contents and the author’s intentions.

Crowe produced his dictionary on the back of his publishing success of the year before, The Duties Of A Constable (1894). This booklet, set out in question and answer format, gave his first-hand account of what a constable does, and verily should do, when confronted with various species of lawbreaking. His next publishing venture, his dictionary, was advertised in Melbourne’s Age newspaper as early as March of 1895, and was on sale in all good bookstores by July of that year. An advertisement of the day ran:

AUSTRALIAN SLANG DICTIONARY. Constable Crowe, whose ‘Police Manual’ proved so successful, has just published an Australian Slang Dictionary, which should prove interesting to the curious, and instructive to the unsophisticated. It contains several thousand slang terms and specimens of the ‘lingo’ adopted by the larrikin and criminal classes of the colonies, as well as a great many everyday expressions not to be found in the standard dictionary. As the methods and habits of ‘the submerged tenth,’ it should prove of utility, and as a library curiosity it will no doubt find its way into the hands of many a purist in language, to whom a slang expression in ordinary conversation is an abhorrence. The price is a modest shilling. The Fitzroy City Press, 5 July 1895, p.3          

Crowe’s stated aim is to record lexical items (grouped under the catch-all moniker of ‘slang’) that do not appear in the standard dictionaries of the day, especially focussing on, but not exclusively so, the language of the underworld. The reason for this is to ‘prevent criminals, rogues and gamblers from conversing with impunity in the presence of the police and public’ (Crowe 1895: 1). Crowe was, after all, a career policeman and reducing crime and foiling criminal activity was his game. How useful his dictionary was in that particular enterprise is unknown; perhaps not much, but, at least it was worth a try.

One can imagine that Crowe might have expected, or at least hoped, that his dictionary would be as well received as his policing manual had been. Sadly, nothing could be further from the truth. Almost right off the bat his dictionary was subjected to uncomplimentary reviews. A. G. Stephens, the new publications reviewer for The Bulletin, Australia’s foremost weekly magazine at the time, was exceedingly harsh:

The Australian Slang Dictionary (paper, 1s. 3d.), by Cornelius Crowe, is an amazingly ignorant production. The author has mixed an olla-podrida of linguistic scraps picked up everywhere – most of which are either not Australian or not slang – and enriched it with little bits of his own. The spelling is atrocious, and the definitions are worse – e.g., ‘Pyjands, a kind of loose drawers;’ ‘Axe to grind, disseminating (!) for personal ends;’ and so on. The collection, however, is not without interest – one likes to know that ‘japanned by the Salvos’ means ‘converted by the Salvation Army;’ it has a good deal of unconscious humor – ‘bubble company’ is defined as ‘land and mining syndicates;’ and it will doubtless find the readers it deserves. The Bulletin (Sydney) 17 Aug. 1895, inside cover.

This must have hurt. A positive review in the widely-read and enormously well-respected Bulletin would have cemented Crowe’s dictionary as a classic. However, as reviews go this is pretty damning. At the same time, the review istelf is pretty terrible. The reviewer complains that Crowe’s dictionary is ‘ignorant’, but then offers practically no evidence of this (obviously pyjands is the result of a typographical or transcription error for the word ‘pyjamas’, and thus indicates a lapse in careful proofreading, not ignorance; but axe to grind can mean ‘a proposition or point of view to advocate out of self-interest’, and so while Crowe’s definition is a little clumsy, it is not wrong nor ignorant). As for the dictionary being an ‘olla-podrida’ (a hodgepodge mixture of various things), good lord! All dictionaries are a miscellaneous assortment or collection of words covering a wide range of topics, linguistic subsets, etc. Hardly a valid criticism to level at a dictionary of slang.

This leaves us with the only other criticism in the Bulletin’s damning review not yet addressed, namely that most of the lexis is not Australian (a claim, it might be added, strongly undermined by the fact that GDoS records at least 250 slang terms of Crowe’s that are also cited from The Bulletin!).

The unAustralianness of Crow’s lexis was also bemoaned by lexicographer A.A. Morris in the introduction to his own dictionary of Australianisms, who apoplectically complained that by his estimation ‘not one word in fifty in his [Crowe’s] little book has an Australian origin, or even a specially Australian use’ (1898: xi–xii). Indeed, it is this issue, the ‘Australianness’ of Crowe’s wordlist, or rather the lack thereof, that has become the enduring critique of his dictionary by all subsequent commentators, and it is the reason Crowe’s dictionary has been discarded as a piece of lexicographical trash not worthy of serious consideration – in my opinion wrongly, as we shall see.

Typical of the type of criticism that has become the sole lens through which to view Crowe’s dictionary, is the treatment of Sidney J. Baker, for long the most widely respected authority on Australian slang. Baker laments the fact that previous Australian slang dictionaries, including Crowe’s, had ‘a good deal of slang […] that was not and never has been Australian’ and that this ‘must be regarded as not only bad workmanship, but doubly unfortunate in that it encouraged unbelievers to persist in their conviction that our [Australian] slang has little originality in it.’ He goes on to complain that this caused Australian slang collectors (such as himself) ‘considerable damage in the eyes both of Australians and the English’ (1945: 26–27). What evidence Baker had of this ‘considerable damage’ remains a mystery, although, with the benefit of hindsight, I feel Baker was at best exaggerating the consequences.

Returning to the subject in 1953, Baker reductively sums up Crowe’s dictionary by stating that ‘many hundred expressions listed in’ it ‘are of English origin’ (Baker 1953: 233).

A 2005 PhD thesis, by Judith Smyth Robertson, on early Australian lexicography provides a more substantial investigation into Crowe’s dictionary, but essentially comes to the same conclusion. Robertson compares the entries in Crowe’s dictionary against some earlier British and American dictionaries and finds a large overlap, concentrated largely on Hotten’s Slang Dictionary of 1865 (or the 1874 edition) and Trumble’s Slang Dictionary of New York, London and Paris of 1881 (a work in itself largely copied from Matsell 1859). According to her calculations, Crowe’s dictionary ‘contains 2688 defined terms’, of which ‘2574 also appear in earlier dictionaries’. The conclusion Robertson draws from this is that basically 95.8% of Crowe’s dictionary, being copied from other dictionaries, is not Australian slang, and accordingly the rest of her analysis treats in detail only those terms she identifies as Australianisms.

None of this type of analysis – either from Crowe’s contemporaries or later scholars – sits well with me. Actually, such appraisals are simply misguided. Crowe was never trying to only record Australianisms, so to criticise him for not doing so is like complaining that an oil painting of roses smells like oil paint, not roses. What Crowe was recording was slang used in Australia, not slang coined in Australia. In fact, Crowe is completely upfront about this himself in the Preface to his dictionary wherein he states:

Although I have entitled the book the ‘Australian Slang Dictionary’, I would ask the reader to bear in mind that but few of the terms it contains have been invented by Australian criminals; the most of them have been brought into use by the criminal classes who have emigrated here from different parts of the world. (Crowe 1895: 1)

That the slang used in Australia in the late 19th century was replete with especially British, but also American, terms should hardly be surprising. Indeed, it would be miraculous if it were not. This should be obvious when one stops to think of the make-up of Australian society at the time. The non-Indigenous section of Australia society was essentially the result of immigration and hence was composed of migrants and the offspring of migrants. The gold-rush era (1850s–1870s), during which Australia’s population quadrupled, particularly saw an enormous influx of migrants from across the world, with an especially large number of American migrants. At the same time, the country had quite unashamedly instituted an unofficial White Australia Policy and between 1860 and 1900 there were over 400,000 ‘assisted migrants’ to Australia: principally skilled workers from Europe, especially from the UK. This is a large number of immigrants for a country that had a population of under 3.5 million in 1895. What this means is that in Crowe’s day a considerable portion of the Australian population were not even native-born speakers of Australian English. Many were speakers freshly, or only recently, arrived from Britain and they would have brought with them their own speech habits. Thus, Australian English, while going in its own direction and producing its own conventions, norms, accent, and, of course, its own slang terminology, was all the while being continually enriched with new material via immigrants arriving at a fairly high rate.

Given this societal composition, it makes sense that a dictionary of the slang in use in Australia in Crowe’s time would have a fair amount of non-Australianisms – especially Britishisms – in its make-up, as a matter of course. It is unavoidable that a mixture of local and non-local terms would be in common usage. Indeed, Australian slang was a veritable ‘olla podrida’, as it still is today. In this light, we can see that Crowe was being sensible to not restrict his dictionary solely to Australianisms. To do so would have undermined the stated aim of his book.

In an addendum, Crowe notes that ‘some of the words in this Dictionary were taken from examples of prison slang given by Mr. Michael Davitt’ who ‘had ample scope for observation during his political retirement at Newgate, Millbank, Dartmoor and Portsmouth [various British prisons], and his prison diary gives numerous specimens of the slang which he found was used by the thieving fraternity’ (1895: 105). Davitt was an Irish republican activist, writer, and lecturer who agitated for Home Rule, and served seven and a half years in British prisons where he endured solitary confinement, hard labour, and poor rations that permanently damaged his health. He wrote a two volume book detailing his time in prison, from which many citations have been added to GDoS []. Obviously, Davitt’s lexis was British in origin, but that does not prevent any of the slang he used also being used in Australia.

Interestingly, in addition to his use of Davitt, Crowe goes on to acknowledge the ‘assistance from many other persons and sources’ and especially expresses his ‘indebtedness to Detective D.G. O’Donnell and J. O’Sullivan, of Fitzroy’ (1895: 105), fellow detectives from his local precinct. This is significant as it shows that Crowe did not rely only on his own knowledge and research, but also had input from other first-hand participants in the law and order profession.

Previous commentators, by focusing on the small amount of Australian coinages in Crowe’s dictionary, have largely assumed that the lexis which Crowe lifted from other dictionaries (and there’s no mistaking that he simply copied many, many entries verbatim from other sources) formed no part of Australian slang or Australian English. But, is this a valid conclusion? I think not. If we take the meaning of ‘Australian slang’ to be the slang used in Australia, irrespective of its point of origin, then what Crowe provides is a snapshot of the slang that was in use in his day.

Actually, by focussing on the overlap of Crowe and his lexicographical sources (i.e. what Crowe lifted from them), no one seems to have thought to examine what Crowe didn’t take. As it turns out, quite a lot! In fact, Crowe was extremely selective. I checked Crowe’s entries against four notable dictionaries he clearly used as source material:

  1. Matsell Vocabulum 1859 (American)
  2. Hotten Slang Dictionary 1874 (British)
  3. Trumble Slang Dictionary 1881 (American)
  4. Anonymous Sydney Slang Dictionary 1882 (Australian)

The following table (based on a sample – the letters A, B, and C – of the first three dictionaries, and the whole of the Sydney Slang Dictionary) shows just how selective Crowe was.

Crowe did not take 62.5% of Matsell’s terms, 70.6% of Hotten’s, 54.5% of Trumble’s, nor 74.5% of the Sydney Slang Dictionary.

Crowe did not just copy previous dictionaries holus bolus, but rather extracted only specific items. If we take time to ask why, then surely the only answer can be that his selection of terms was based on attaining his dictionary’s stated objective. In other words, he selected those terms which he thought or believed were being used in Australia, and therefore of some use to his lexicon’s goal – providing the Australian public with a useful set of terms in use amongst the underworld.

In the case of the Sydney Slang Dictionary, his only Australian source book, Crowe only took 269 entries (25.5%) of the 1053 in the original. That he rejected 74.5% as not appropriate for his dictionary reveals that he was no uncritical copyist. So, where the Sydney Slang Dictionary has entries for dead beat ‘one who evades his debts’, dead lurk ‘entering a house during church service’, and dead sucker ‘a stealer of coats and umbrellas’, none of these were chosen by Crowe for use in his dictionary, presumably because he and his colleagues had never heard them used by the criminal class of their day.

Instead, in the same section (i.e. terms beginning with dead) Crowe has eight terms.

Of these, eight terms, six are recorded by Hotten (1874), and one occurs in Matsell (1859). But, when one examines the entries in GDoS, it is plain to see that, with the exception of dead swag, all the terms in Crowe’s dictionary were indeed in use in Australia prior to the publication of his dictionary in 1895. Sure, some of them were used earlier in Britain, but nevertheless the citations prove that these terms were also genuine members of Australian slang at the time, and as such may have indeed been known to Crowe and his fellow police officers who had first-hand experience dealing with underworld characters of the day.

Although I have searched high and low, I have found scant evidence for the term dead swag: the sole exception being an anonymous column in Melbourne’s Argus newspaper of September 1901. But, as the saying goes, absence of evidence does not equate to evidence of absence. Crowe appears to have lifted his dictionary entry for dead swag straight out of Matsell’s dictionary, or else Trumble’s: Crowe’s definition is exactly the same as theirs. But, presumably Crowe chose to copy this term (and not many others on the same page) for a reason, i.e., he knew it. It may have not been very common in Australia, but dictionaries are not obliged to only record common terms, in fact, they benefit by erring of the side of inclusion. Clearly the more uncommon a term is, the less likely a general reader (or listener) is to know it, and the more likely they will want to look it up in a dictionary.

Another point which must be considered when assessing Crowe’s dictionary is that today we only have an imperfect record of the language of the 1890s. In fact, what we can see today is merely a subset of what language was actually in use at the time, run through a number of sieves each of which exclude some of the whole. The sifting process goes like this: first, not all words in colloquial use in the 1890s would have necessarily made it into print; second, not everything that was in print back then is now extant (held in libraries, or even privately); third, only a fraction of total library holdings are readily available in searchable format. On this last point, although modern searchable databases – such as the enormously vast and vastly important Trove database – give the modern linguistic researcher access to so much more than was ever available before today, still not every extant 19th century book, magazine, or newspaper has been scanned in and databased – not by a long shot. As for manuscript materials, very little to none of this has been transcribed, databased, or made available.

Happily, each day more and more material comes online somewhere, and the fact that we cannot now easily find evidence for a certain word or phrase, does not mean we won’t be able to find evidence for it next year, even next week, if we look again. In particular, the massive Trove holdings have proven to be an indescribably valuable boon to Australian lexicography, and with this tool I have been able to track down much evidence that Crowe’s dictionary entries were indeed reflective of Australian slang at the time, even if not in every instance.

The following set of terms have been held up as proof of Crowe’s faulty lexicographical method (Robertson 2002).

  • bit: a tool used by burglars1
    • bower: a prison
    • copbusy: to pass stolen goods to a confederate
    • croaker: a newspaper
    • goaway: a train or tram
    • jade: a long prison sentence
    • lion: to frighten or intimidate
    • polisher: a jailbird
    • roofer: a hat
    • : to convict

For this set of terms, evidence exists for both chop up and jade, though for both the evidence is after their appearance in Crowe’s dictionary. For the others, so far no evidence has been discovered, but, with the exception of copbusy, all other words are difficult to search for as Trove’s massiveness means that any search returns an enormous number of hits. In this instance, the sheer size of Trove is a downside rather than an upside. You might think that jade is not such a common term, but Trove returns over 255,000 hits, and if advertisements are excluded there are still over 100,000 hits. Similarly, there are over 100,000 hits for croaker, which used to be the common term for an inveterate complainer, subsequently replaced in Australian English with the term whinger. Even when you search for ‘croaker’ and ‘newspaper’ together, you get over 9000 hits and to wade through each of these would take a month of Sundays. And, if you think lion as a verb, especially in the participial forms lioned and lioning, might have a more feasible number of hits to navigate, think again. Unfortunately, searches for these terms give one an incalculable number of scanning errors (known as scannos in the trade) for words ending in -tion, -tioned, -tioning, especially as this is a great place to hyphenate long words over a line break (scannos being another foible of Trove, which, to be fair, permits and encourages its users to correct such OCR errors).

In the end, Crowe’s dictionary may indeed have had a certain amount of terms that were not ever used in Australia. Maybe. But, so far, my recent investigations lead me to believe that much of what Crowe recorded did have currency in his day, and that, contra the prevailing view of Crowe, there was no significant fault with his lexicographical method. Sure, Crowe’s dictionary was hastily put together and does have a number of poor definitions, typos, and the like. He was no practised or trained lexicographer, but an amateur at the game, and in light of this, I believe that not only did he do a relatively good job but that his dictionary needs careful reconsideration and reappraisal. We won’t know how many entries in Crowe’s dictionary accurately represent Australian slang of its day without examining each and every case, and there are 2961 defined terms in the dictionary, so quite a big job. I’ve made a start, but there’s a long way to go.

The following are some of the terms that Crowe included in his dictionary and that GDoS now reveals indeed are, or at least were once, part of Australian slang:

  • absquatulate: to disappear; to decamp [US since 1833; in Australia from 1841]
  • agitate the communicator: to ring the bell [in Australia from 1859]
  • Alfred David: an affidavit [British since 1865; in Australia from 1885]
  • back slums: the low disreputable portions of a city [British since 1821; in Australia from 1834]
  • barber: a hotel barber [Australian original, first recorded in Crowe 1895]
  • barking irons: firearms [British since 1764; in Australia from 1826]
  • battler: a punter trying to make a living on the turf [an Australian original, from 1886]
  • block and tackle: a watch and chain [Australian original, since 1895]
  • bog orange: a potato [British since 1864; in Australia from 1868]
  • bone box: the mouth [British since 1788; first Australian use in Crowe]
  • chopper: a blow or punch [British since 1793; in Australia from 1835]
  • chop-up: the divvying up of stolen goods [Australian original, first recorded in Crowe]
  • cracksman: a burglar [British since 1790; in Australia from 1827]
  • federating: making love [Australian original, since 1886; the proposition for the Australian colonies to federate into a commonwealth (occurring finally in 1901) was constant news fodder of the era]
  • hotel barber: a hotel guest who robs other guests [Australian original, since 1871]
  • lushington: a drunkard [first recorded in Vaux 1812 and persisting in Australia up to the 1930s]
  • out for an airing: of a horse, running in a race but not trying to win [Australian original, since 1890]
  • piece of blue paper: a summons [Australian original, since 1879; the origin of the later bluey ‘a summons’]
  • schlog it on: to put up the price [had a brief existence in 1895 due to a much reported court case]

Australian slang has a rich history that has by and large been ignored as a consequence of Australian lexicography’s tunnel vision fixation with Australianisms. Naturally, Australia’s penal beginnings meant that underworld slang, the thieves’ patter, the flash tongue, had a healthy start in the fledgling colony, especially in the big cities of Sydney and Melbourne. A columnist writing in 1869 decries the prevalence of the flash language in Melbourne:

Perhaps in no other city is this terrible language spoken with such facility as in Melbourne. The reason is obvious. During the last twenty years has been pouring into the city a crowd of released convicts, redeemed scoundrels, adventurous vagabonds, all of whom speak this hideous tongue with facility. The Australasian (Melbourne) 17 July 1869, p. 8.

Of course, Australians were not all of the criminal class. However, I believe the desire to disavow the penal stamp has in part contributed to the disinterest in slang that was not home-grown. Crowe’s dictionary gives a valuable insight into the sort of slang that was in use in his day, and far from ignoring the lexis he records, much better would be to investigate it further so that a more complete picture of Australian English can be drawn.


1. Robertson (2002) defines bit as ‘jemmy, a crowbar’, but Crowe, and his source, Trumble, both define it as ‘a burglar’s instrument’. Trove has many examples of the term centre bit as a burglar’s tool, alongside the jemmy, but I could not find bit alone. I believe Robertson has mistaken the meaning in this case.  On the same page, Robertson also claims that Crowe defined japanning as ‘stealing cash-boxes’, but this is incorrect. Crowe instead has japanned ‘a thief converted by the chaplain is so spoken of’, a term going back to Matsell (1859). Where Roberton got her definition I do not know: it is not in Matsell, Hotten, Trumble, or the Sydney Slang Dictionary, nor is it in Vaux or Farmer and Henley. It is, however, a legitimate Australian slang term, albeit obsolete.


Baker, Sidney J. 1945. The Australian Language. Sydney: Angus and Robertson.

Baker, Sidney J. 1953. Australia Speaks. Sydney: Shakespeare Head Press.

Crowe, Cornelius. 1895. The Australian Slang Dictionary, containing the Words and Phrases of the Thieving Fraternity, together with the Unauthorised, though Popular Expressions Now in Vogue with All Classes in Australia. Fitzroy: Robert Barr.

[Hotten, John Camden]. 1874. The Slang Dictionary: Etymological, Historical, and Anecdotal. London: Chatto and Windus.

Morris, A.A. 1898. Austral English: A Dictionary of Australasian Words, Phrases and Usages. London: Macmillan and Co.

Robertson, Judith. 2002. “The perils of lexicography.” Ozwords, 9(1): 1–­3.

Robertson, Judith. 2005. Australian Lexicography 1880–1910: An Evalutation. Doctoral thesis, Australian National University, Canberra.

The Sydney Slang Dictionary: Comprising All the Slang Words and Phrases in use in Sydney and in the Shadows of Life. 1882. Sydney: H. J. Franklin.

[Trumble, Alfred]. 1881. The Slang Dictionary of New York, London and Paris. New York: National Police Gazette.

Vaux, James Hardy. 1812. “A new and comprehensive vocabulary of the flash language.” In Memoirs of James Hardy Vaux (vol. 2). 1819. London: W. Clowes.

What Makes a Bonzer Etymology?

Exploring the mysteries of bonzers, boshters, boskers, and bontodgers

James Lambert



[James Lambert is the Contributing Editor of Green’s Dictionary of Slang. In this special guest blog he has been looking into the etymology of one of Australia’s quintessential slang words: bonzer and with it a variety of associated terms. The conclusions he draws, based on extensive research, take our knowledge of this unique group of superlatives to a new level. Now read on. JG]


The Etymologist’s Art

Etymology is the science/art of determining the origin of words. As a discipline, it requires recourse to various linguistic subfields, especially phonology and semantics, but also an equal amount of historical research.

I say science/art quite deliberately as the wide-ranging field covers both ends of the spectrum. But, in all honesty, the watchword of all good etymologists is ‘common sense’. That common sense, however, must be predicated on cold hard evidence, without which etymological speculations are nothing more than speculations, and very often very poor speculations at that, not only because they lack any solid evidentiary basis, but because they often fly in the face of what evidence there is. In other words, etymology works best when the etymologist has a very deep understanding of the history and development of the language or languages involved (there is frequently more than one) and is also able to draw on considerable linguistic evidence.

In this blog we will examine some of the essential tools of the etymologist and see how they can be applied to sift through different and competing suggestions so that unverified and unverifiable speculations are sidelined and we are left with more realistic and likely theories.

One of the major analytic heuristics is analogy. The basic idea here is that if a certain sound or sense or spelling development is known to have occurred with a certain word or set of words some time in the past, then it is reasonable to assume that the same development could happen again. For example, take the relatively new word newb, meaning ‘a newcomer or novice’, which first appears in the 1990s. We can suggest a tentative etymology, namely, that newb is a clipping of the word newbie, which has the same meaning. Now, it doesn’t take an etymologist, nor a rocket scientist for that matter, to intuitively know that this suggested derivation is 100% correct – even to the untrained eye it is clear that newb is obviously a shortening of newbie. However, one of the reasons we can be so sure is analogy. For when we ask ourselves if there are any analogous examples, the answer is a resounding yes, there are countless analogous examples: pug ‘a boxer’, clipped from pugilist (1858); pub, clipped from public house (1800), rhino and croc clipped from rhinoceros (1858) and crocodile (1884); gym, from gymnasium (1871), and so on. One of the earliest examples being gent, a clipping of gentleman dating back to the 1600s.

In fact, the process of foreshortening words is a common formative process in the English language (and other languages besides), and as English speakers, we know this. This knowledge is part of our overall comprehension of our language and it is on the basis of our knowledge of these many analogous examples that we can be confident that newb is a shortening of newbie.

You will have noticed that I have given the examples of clipped words followed by a year in parentheses. These are, as you will of course have already discerned, the year of earliest record for each of the terms. This brings us to the next tool of the etymologist, namely, chronological precedence. Quite simply put, if a certain term is meant to be the origin of another term, then the first (called the etymon) must have existed prior to its derivative. This is just common sense. Importantly, all the examples of clippings I have given above antedate the appearance of newb, thus showing that the process of forming neologisms by clipping was already in place in the language.

Nevertheless, as common sensical as this may seem, amateur etymologists often put forward theories that completely ignore this essential point. A good example of a chronologically impossible (and therefore incorrect) etymology is for the birdwatcher’s term jizz, meaning ‘the overall structure of a bird’. The erroneous theory goes that this word was originally an acronym GISS, standing for General Impression, Shape, and Size, which was used by fighter pilots in the Second World War in reference to a method of recognising aeroplane types. The fatal problem with this theory is that the birdwatcher’s term was first used in the 1920s, considerably before WWII.

This leads us to the next implement in the etymologist’s toolbox, namely documentary evidence. As etymologies are by and large dealing with the speech habits of the past, sometimes the very ancient past, our only access to the way words were formerly used is if they happen to survive in historical texts. This is one essential that good etymologists always rely on. For example, with the jizz/GISS etymology discussed above, we know it must be wrong because there is documentary evidence that jizz was used in the 1920s (it occurs in the Manchester Guardian of the 6th of December 1921), and there is no documentary evidence that GISS was used before WWII. However, if, for instance, there were no surviving copies of the Manchester Guardian from 1921 – if say they had all destroyed during the Manchester blitz of 1940 – then the GISS derivation would have seemed eminently plausible, rather than entirely impossible.

This of course shows a weakness in etymological research – that is, it can only be as good as the documentary evidence allows. Etymologists, and lexicographers as well, are bound by the resources available to them. One can never see every use of a word, especially as most language takes place in speech and is for all practical purposes lost forever once it is spoken. Only the thinnest minuscule modicum of the daily language of billions gets written down or electronically recorded, and then only a paltry sampling of what is recorded ever passes under the gaze of the lexicographer or etymologist.

So, to conduct high quality etymology, one must have recourse to historical texts, and the more the better. Thankfully, these days more and more historical texts are being digitised and made available, and the task of the etymologist is becoming ever more easier with respect to accessing documentary evidence upon which they can conduct their careful and often painstaking detective work. And, as we shall see below, this can either make or break an etymology.

Another important string to the etymologist’s bow is historical knowledge, both linguistic and sociocultural; a knowledge of who was in contact with whom, of what languages or dialects or lingos were formerly in use and when and where they were. An absence of this can result in highly improbable folk etymologies, which, while often providing a pleasing story, are utterly ridiculous. A classic example of this I once experienced was the origin of the word pothole put forth by a tour bus driver in Bath, England. His theory, presented as absolute fact, of course, was that during the Roman occupation of England (roughly 43–84 AD) the native Celts needed clay to make pots and they took advantage of a ready supply of good clay by excavating it from Roman-built roads, leaving ‘potholes behind. This explanation was generally received with much interest and ‘oohing’ from the bus occupants, while I bit my tongue reasoning that one should never let the truth get in the way of a good story and that to deflate the tour guide in front of everyone was just bad manners. Yet, anyone with a modicum of historical knowledge will see the gross impossibility of the tale. First, during the Roman occupation of England, Celts spoke Old Brythonic and the Romans spoke Latin. The words pot and hole are both Germanic words first brought to England around 450 AD, long after the Romans had left. Second, the word pothole itself is only recorded since the beginning of the 19th century. So, as a result of ignoring historical knowledge, the tour guide’s etymology of pothole was impossible by a distance of some seventeen centuries. Third, Roman roads did not use clay, or used clay mixed with rubble as one of the lowest layers, not easy to dig up, nor very useful for potting. (Actually, the proposed etymology also fails from an absence of documentary evidence, just for good measure).

Finally, the last arrow in the etymologist’s quiver is Occam’s razor, also known as ‘the law of parsimony’. For our purposes, this simply means that when a suggested etymology requires too many steps or too many assumptions, then it is probably wrong, and suggestions with the least complexity have a greater chance of being right. This is a rule of thumb, not a hard and fast rule, but it often comes in handy. A great example of a suggested etymology that violates Occam’s razor is the famously preposterous etymology for the word hoodlum first outlined in the fourth edition of Bartlett’s Dictionary of Americanisms (1877), and restated thereafter by many authorities. Repeating a story he had heard, Bartlett posits that the word hoodlum originated by a printer’s error for the term noodlum, which is backslang for Muldoon, the name of the leader of a San Franciscan gang of street Arabs. This makes the assumption that there was a backslang version of Muldoon’s name (no evidence for this), and then the assumption that a printer made an error and typeset hoodlum instead of noodlum (no evidence for this), and then the assumption that from this single error the word entered general slang (there are, at least, a few examples of a single text contributing a new word to the slang lexicon, so this last assumption is not entirely fanciful). In any case, this piling up of assumption on assumption would make William of Occam spin in his grave, and makes the suggested etymology highly unlikely indeed. An alternate etymology suggests that hoodlum may be from German dialect words, such as hudelum ‘disorderly’, Hudellump ‘slovenly person’, and is to be preferred as it relies on less assumptions.

These principles of etymological investigation work in concert with one another, but depending on the derivation suggested often only one or two may apply. In the following, I will begin by making it clear when I am invoking one of the principles, but then will tail this explicitness off so as not to belabour the point and clutter up the text.