LINGUIST List 9.360

Wed Mar 11 1998

Sum: O-words: colloquial/slangy words ending in <o>

Editor for this issue: Brett Churchill <brettlinguistlist.org>


Directory

  • Mikael Parkvall, O-words: colloquial/slangy words ending in <-o>

    Message 1: O-words: colloquial/slangy words ending in <-o>

    Date: Sun, 8 Mar 1998 17:45:15 +0100 (MET)
    From: Mikael Parkvall <parkvallling.su.se>
    Subject: O-words: colloquial/slangy words ending in <-o>


    Some time ago, I posted a question on colloquial/slangy words ending in <-o> on both Linguist List and Nordlingnet. To exemplify, I gave the English words "weirdo", "wacko", "wino", "psycho" and "fatso", asking people on the lists about the use in various languages of words formed along similar lines. I am happy to say that I received no less than 39 answers, for which I am of course most grateful. The names of those to whom I am indebted can be found at the end of this posting. Given the large amount of information I received, the following summary is pretty long. I have organised it the following way:

    1. An O-word taxonomy 1.1 Apocoped O-words 1.2 Paragoge O 1.3 "Agentive" O-suffixation 1.4 "Parasitical O" 2. Formal characteristics of O-words 2.1 Phonology 2.2 Semantics 2.3 Morphosyntax 3. Is the O productive? 4. O-ism in various languages 4.1 American English 4.2 British English 4.3 Australian English 4.4 German 4.5 French 4.6 Danish 4.7 Icelandic 4.8 Dutch 4.9 Swedish 4.10 Finland Swedish 4.11 Finnish 5. History and O-rigins 6. Further reading

    1. An O-word taxonomy Many of the respondents felt that there are several kinds of O-words, and some even made taxonomies of the various caterogories in one or more languages. Probably about half of those who responded underlined the differences between words such as "psycho", "typo", which are evidently clipped verions of "psychotic/psychopath" and "typographical", and other items, where the "o" is a paragogic vowel, and has no etymological raison d'\234tre. This distinction seems reasonable, although the two groups may have increased the frequency of each other. Belonging etymologically to the apocope group, "psycho" may have caught on partly because of its semantic and phonological resemblance to the paragogic group, as suggested by Larry Rosenwald. The most elaborate taxonomy was sent to me by Jannis Androutsopoulos. I have adapted it somewhat, but what follows is in essence derived from Jannis' proposal:

    1.1 Apocoped O-words <o> as an etymological adjoining vowel that has become final through clipping of the first constituent of (usually) an Adj+N compound of (usually) Latin or Greek origin. Examples: "Psycho" (< "Psychopath"), "\214ko" (< "\214kologe") (German), "klepto" (< kleptomaniac), "nympho" (< nymphomaniac), "slo-mo" (< slow-motion) (English), "homo" (< "homosexuel") (Danish and other languages), "aso" (< "asociaal") (Dutch), "skitso" (< [?]) (Finnsih), "clito" (< "clitoris"), "m\233galo" (< "m\233galomane"), "h\233bdo" (< "h\233bdomadaire") (French), "mongo" (< "mongoloid"), "alko" (< "alkoholist") (Swedish). This group will henceforth be referred to as O1.

    1.2 Paragoge O <o> as a suffix of clipped words (nouns or adjectives). Examples: "Fascho" (< "Faschist"), "Anarcho" (< "Anarchist") (German), "Gibbo" (< Gibson), "Jacko" (< Jackson) (British English) "paddo" (< "paddestoel") (Dutch), "ap\233ro" (< "ap\233ritif"), "facho" (< "fasciste") (French), "str\230t\243" (< "str\230tisvagn"), "p\250k\243" (< "p\250kalegur") (Icelandic), "neggo" (< negativ), "despo" (< desperat) (Swedish). This group of words will henceforth be referred to as O2.

    1.3 "Agentive" O-suffixation Agentive - or at least prototypically nominalising - O-suffixation. Henceforth referred to as O3. Examples include "cheapo" (< cheap), "sleazo" (< sleazy), "sicko" (< sick) (American English), "sado" (< sad), "provo" (< provisional) (British English), "h\228irikk\246" (< "h\228irit\228"), "juoppo" (< "juoda") (Finnish), "Banalo" (< "banal"), "Labilo" (< "labil"), "Normalo" (< "normal") (German), "pervo" (< "pervers"), "dummo" (< "dum"), "pucko" (< "puckad") (Swedish). This class is represented in most of the Germanic languages considered here, as well as in Finnish. The O-word is formed from an adjective or a verb (which is shortened if polysyllabic), resulting in a noun denoting a person having the quality described by the adjective, or a person habitually performing the action described by the verb. Compared to agentive suffixes proper (e. g. "-er", "-or", "-ary", "-ster"), Chuck Bigelow feels that the "-o" suffix appears to have more of the copular and less of the active agentive sense ("wino" 'one who is wined [winish? winey?]', "weirdo" 'one who is weird', "fatso" 'one who is fat'). This class is not entirely homogenous, since the semantic value of the O varies from one word to the other, even in the same language. Diagnostic for inclusion has been either some kind of agentive-like meaning, or a nominalising function for those words which were not originally nouns.

    1.4 "Parasitical O" "Parasitical O"; an ending with is neither etymologially or semantically motivated, and which is added to a word that has not undergone any other phonological change. "geilo", "tollo", "nullo", "kompletto" (German). Henceforth referred to as O4. "Jacko" (< Jack), "Johno" (< John) (British English), "fuldo" (< "fuld"), "festo" (< "fest") (Danish), "sk\237t\243" (< "sk\237t") (Icelandic).

    2. Formal characteristics of O-words

    2.1 Phonology In most cases represented here - regardless of host language - the o's occur on monosyllabic stems, or on stems made monosyllabic through apocope, and this has probably (whether deliberately or not) been a requirement for inclusion in the o-group for me as well as for my respondents. The O-words are thus more or less by definition bisyllabic. It is also notable that a very large proportion of the O-words consist of CV sylllables only - presumably the least marked of all syllable types. Though there are indeed exceptions (especially words with complex initial onsents, i. e. with a CCVCV structure), the prototypical O-word has the structure CVCV. I haven't consulted reverse dictionaries for other languages, but so far as Swedish is concerned, it is potentially significant that my dictionary lists no indigenous words with this phonological structure (i. e. (C)CVCV, where V2 is /u/, that is <o>). The stem doesn't normally undergo any changes apart from shortening (or other predictable processes, such as vowel harmony in the case of Finnish), but in a small number of cases, it is slightly altered, usually through epenthesis, as in "fatso" (< fat) or consonantal changes, as in Swedish "mysko" (< "mystisk") 'strange, odd, queer'.

    2.2 Semantics It is obvious that a remarkably large proportion of the O-words have negative connotations, and indeed, as one American respondent said with regard to the productivity, "kids can and do make up insults using it". Apparently, as Henny Klein put it (with regard to Dutch), "especially personal characteristics can be expressed in this way", and Joe Hilferty pointed out that many of the O-words (e. g. "weirdo," "wacko," "psycho," "schizo", "wino" "fatso") refer in one way or the other to states of mental or physical abnormality. In short, it is "unusual to see the -o attached to words that aren't insulting, derogatory, or negative in some way" (David Weiss). There are of course exceptions, and even though most O-words are disparaging, the Swedish O-set includes "hygglo" (< "hygglig") 'friendly, nice; one who is friendly, nice' and "schyssto" idem (utlimately from French "juste"). Interestingly, however, these seem to often (though by no means exclusively) be used ironically.

    2.3 Morphosyntax Apparently, the vast majority of O-words are nouns, although some English examples (e. g. "wacko" and "psycho") can function either as nouns or as adjectives. A much smaller group, it seems, function exclusively, or at least primarily as adjectives, including "nutso" 'crazy', "blotto" 'drunk', "schizo" 'schizophrenic/schizoid'. The most interesting cases, in my view, are those where o-suffixation not only changes the phonological and semantic/stylistic, but also the syntactic properties of the word. This is the case of a subset of the O-words in some languages. In at least American and British English, Dutch, German, Swedish and Finnish, there are clear cases of a nominalising O. In the Germanic languages, the O-word is derived from an adjective, the derivate always denoting 'a person having the quality X', whereas in Finnish, nominal O-words seem to derive more often from verbal stems. Some Germanic examples:

    American English: "sleazo" (< sleazy) 'sleazy person', "cheapo" (< cheap) 'cheapskate', "queero" (< queer) 'homosexual', "sicko" (< sick) 'perverted person', "weirdo" (< weird) 'crazy person', "wacko" (< whacked) 'crazy person' Dutch: "aso" (< asociaal) 'not behaving in a socially acceptable way' German: "Judo" (< junger Liberaldemokrat) 'young liberal-democtrat', "Juso" (< junger Sozialdemokrat) 'young social-democrat', "Banalo" (< banal) 'banal', "Labilo" (< labil) 'emotionally unstable person', "Normalo" (< normal) 'normal person' Swedish: "dummo" (< dum) 'stupid person', "fetto" (< fet) 'fat person', "pucko" (< puckad) 'stupid person', "schyssto" (< schysst) 'nice person', "hygglo" (< hygglig) 'nice person' (Swedish).

    One particularly noteworthy characteristic of the Swedish O-words is that they are all neuter, whereas most animates, and virtually all +HUMAN nouns in Swedish are non-neuter. "Mongo" (< "mongoloid") is thus "ett mongo" (INDEF) and "mongot" (DEF) rather than *"en mongo" and *"mongon". It is tempting to assume that a gender that is normally associated with inanimates is used in order to render the word more disparaging.

    3. Is the O productive? Most people seemed to agree that the o-affixing is indeed productive in American English, although opinions differed somewhat with regard to the degree of productivity. One respondent called it "somewhat productive", whereas others emphasised that there is a core of more or less lexicalised items (some of which will even be found in dictionaries), and a more unstable periphery where the o-items are variably used and variably acceptable. A particularly interesting feature is the limits of productivity. Although informants may differ considerably, there seem to be potential coinages that no native speaker would actually produce, except perhaps tongue-in-cheek. For Swedish, the impossible group would include *"maso" 'masochist' (although "sado" would be acceptable) and *"smalo" 'thin person' (but "fetto" is perfectly well-formed), for American English *"cracko", *"snowo", *"dopo" or *"drunko" (though one can be a "wino"). The relative newness of most Swedish O forms make it temping to assume that the suffix is becoming more and more productive, but a survey made by some students of mine indicated that the productivity is in fact declining, with my generation (I was born in 1971) probably representing the peak of O popularity. For the coining of nick-names, the variation between O usage and other morphophonemic processes is obviously lexically determined. John Atkinson (from Australia) wonders why he is always "Acko", when his friend Douglas is always "Dougie". Ian MacNamara and Macquarie Island are "Macca", or sometimes "Maccas", but never *"Mackie" or *"Macko", despite the widespread o usage in Australian English. "On the other hand", says John, "I could be either Johnno or Johnnie - in this case, the former is rather jocular, while the latter is more diminutive (thus more likely to be used for a child". The same applies to derivations indicating a person's occupation or avocation; thus there are "garbos" and "rabbitohs" in Australia, but also "truckies" ("trucker" in American English) and "wharfies" 'wharf labourers' ("dockers" in American English). Some occupations do not seem to take either suffix, so that a person who paints houses is always a painter, never a *"painto" or a *"paintie". There are some attestations Semantic differences resulting from varying choices of agentive suffixes. One minimal pair is "bikie", which is used in Australia for a member of a motorcycle gang, while a "biker" is a law-abiding citizen who rides a motorbike for recreation (This distinction is apparently only made by the motorbike riders themselves, however). Simlarly, the job of a "rabbitoh" 'person who hawks rabbit meat from door to door' is completely different from that of a "rabbiter" 'someone who hunts or traps rabbits for their skins or meat (i.e., professionally rather than as a sport), or who is paid by the landowner to exterminate them'. John Atkinson says he suspects the choice of the "-o" or "-ie" suffix in Australian for a newly invented word to be to some extent phonetically driven, but was unable to guess what the rules may be. In some cases, the two suffixes may be in free variation, and it is not clear what difference there would be between a "weirdo" and a "weirdie", between a "fatso" and a "fattie", or between "wacko" (used adjectivally) and "wacky". In German, where the formation of O3 words is indeed productive, it only yields designations for socio-cultural groups and/or character traits, but not for characteristics relating to intelligence or physical constitution. Counterparts to words like "weirdo" would rather take the "-i" suffix in German, according to Jannis Androutsopoulos.

    4. O-ism in various languages

    4.1 American English In addition to the words I myself used as examples in my query ("weirdo", "wacko", "wino", "psycho", "fatso"), a number of additional examples from English usage were suggested, including the following: "apeo", "bimbo", "bizarro", "blotto", "bozo", "creepo", "daddyo", "dago", "dino", "dorko", "dumbo", "gringo", "jerko", "Jimbo", "Kebo", "keeno", "kiddo", "kinko", "lardo", "lezzo", "lesbo", "limmo", "lingo", "Manbo", "medico", "narco", "nerdo", "nutso", "pervo", "pinko", "presto", "queero", "Sluggo". Both Chuck Bigelow and Kevin Caldwell mentioned the Marx brothers, all of whom had names that by phonological criteria would fit into the o-category. Chuck also gave me the supposed (adding that they might be folk-etymological) derivations of their names

    "Groucho" = "he who is grouchy" "Harpo" = "he who plays the harp" "Chico" = "he who chases women ('chicks')" "Gummo" = "he who wears rubber-soled shoes" "Zeppo" = [I forget the supposed derivation]

    For the Marx brothers, the "-o" suffix maybe have been chosen by analogy with the Italianate names of popular Vaudeville clowns, ending in the usual Italian masculine suffix, "-o". The same explanation could be valid for Kevin's mention of children doing magic tricks calling themselves things like "Kevino the Magnificent" or "the Great Kevinini". In the latter case, I myself suspect that Erikh Weiss, alias Harry Houdini could be the main influence today.

    4.2 British English I'm not sure to what extent British O-ism differs from that of American English. In any case, Sean Jensen pointed out to me the that the O is attached to Christian names (e. g. "Jacko", "Johno"), but that British surnames are particularly likely to attract O's. "Gibbo" (< Gibson), "Jacko" (< Jackson) where the examples he gave. Other O-words used in Britain (most of which are not exclusively British) include "lotto" (< lottery), "promo" (< promotion), "demo" (< demonstration), "bucko" (< buck, 'young man'), "sado" (< sad; a loser), "provo" (< provisional; 'IRA member'), "slo-mo" (slow-motion).

    4.3 Australian English According to Keith Battarbee, John Atkinson, Shelly Harrison and George Huttar, the <-o> is very common in Australian English (and even more than in the U. S.), witness examples as "reffo" 'refugee', "smoko" 'cigarette break', "milko" 'milkman', "rabbitos" 'person who hawks rabbit meat from door to door' (a job now extinct since the introduction of myxomatosis), "Freo" 'Freemantle', "garbo" 'garbageman' and "arvo" 'afternoon'.

    4.4 German >From Germany, O-words of several kinds are reported (Jannis Androutsopoulos' taxonomy referred to above was based on German). Other respondents suggested "Macho", "Realo" (< "Realist", i. e. member of the "realist" rather than the "fundamentalist" wing of the Green Party; a member of the latter would be dubbed "Fundi"), "Fascho" (< "Faschist"), "Juso" 'young social-democrat', "Judo" 'for young liberal-democtrats'. Note that for the two last examples, the o is etymologically motivated in the former case, but not in the latter. Ewald Lang therefore suggests that "Judo" has been formed in analogy to "Juso".

    4.5 French

    In French, the O1 group is the largest by far. Such words generally seem to have negative or popular connotations, which may be due to many of the adjectives or nouns in both the standard and the slang register ending in /-o/ having similar properties; examples given by Roger Billerey are: "lourdaud" 'clumsy', "pataud" idem, "costaud" 'strapping', "salaud", "saligaud", "salopiot" 'bastard', "blaireau" 'annoying, ill-tempered person' (lit. 'badger'). In many French insults, especially as used by children, <-o> alternates with <-os>. Examples hereof are "cr\233tino" or "cr\233tinos" (< "cr\233tin"), "gravos" (< "grave" in its slangy meanings of "deranged" or "stupid and/or boring"). Another interesting case reported by Roger Billerey is the slang adjective "craignos", coined after the slangy expression using the verb "craindre" 'to dread', "\231a craint" ("it sucks", lit. "it dreads"). It seems like o-suffixation in French rarely affects the categorical status of the words involved, although several of the words are multi-functional (cf. however "lourdaud" above, which is derived from "lourd" 'heavy'). Claire Saillard suggests that o-suffixation in French is allowed only with words that are "1) suffixated (and thus considered 'too long'), 2) ambivalent as to their syntactic category".

    4.6 Danish Tore Kristiansen reported results from a pilot study of slang usage among the youth of N\230stved, Danmark (part of the UNO project [="Ungdomssprog i Norden", i. e. Youth language in the Nordic countries]). I was given a large number of interesting words collected among N\230stved teenagers. Some of these are apparent loans from colloquial English ("daddio" 'father', "weirdo" 'unintelligent person; strange, odd, queer', "fatso" 'fat person') and others from Romance languages ("amigo" 'friend', "bambino" 'child', "dinero" 'money', "bueno" 'something good'). Some are obviously derived from standard Danish, such as "brillo" 'glasses' (< Da "briller"), "festo" 'party' (< Da "fest"), "fuldo" 'drunk' (< Da "fuld"), "tynno" 'thin; thin person' (< Da "tynd"), "godto" 'something good' (< Da 'godt'). The usual apocoped "homo" 'homosexual man or woman' also appeared, as well as some items for which I saw no apparent etymology.

    4.7 Icelandic The most well-known Icelandic o-word appears to be "str\230t\243" (< "str\230tisvagn") '(city) bus'. Other examples provided by my colleague P\233tur Helgason were "p\250k\243" 'out, not up-to-date (of a person)' (< "p\250kalegur" 'impish'), "hall\243" idem (< "hall\230rislegur" idem; cf. "hall\230ri" 'year of famine'), and "kamm\243" 'friendly' (< Danish "kamrat" 'comrade'). In Icelandic, even place-names may receive the o-suffix after the expected apocope. \216ystein Alexander Vangsnes reports the use of "Grund\243" and "Patr\243" for the comunities of Grundafj\246r\240ur og Patreksfj\246r\240ur respectively. He also quoteed a recent Icelandic novel ("Thetta er allt a\240 koma" by Hallgr\237mur Helgason), where a character uses "sk\237t\243" instead of "sk\237t" 'shit'.

    4.8 Dutch In Dutch too, the <-o> displays some productivity, especially among younger speakers. Some, such as "weirdo", are simply loaned from English, but others are formed from Dutch roots. Henny Klein provided me with the example "paddo", derived from 'paddestoel', a narcotic mushroom (toadstool). The o-form of this word has now gained enough currency to be used even in newspapers. Henny also provided the above-mentioned "aso" (< "asociaal").

    4.9 Swedish The Swedish set of O-words includes members of the groups O1-O3, but few obvious cases of O4 ("schyssto" 'nice, friendly' could count either as O3 or O4). Examples of O1 words are "alko" (< "alkoholist") 'alcoholic', "favvo" (< "favorit") 'favourite', "homo" (< "homosexuell") 'homosexual', "kollo" (< "koloni") 'children's summer camp', "lyllo" (< "lyckost") 'exceptionall lucky person' (lit. 'lucky cheese'), "mongo" (< "mongoloid") 'crazy person', "n\246rdo" (< "n\246rd", i. e. English 'nerd') 'nerd', "pancho" (< "pension\228r") 'old person', and "psyko" (< "psykopat") 'crazy person'. O2 items include "aggro" (< "aggressiv") 'aggressive (person)', "despo" (< "desperat") 'desperate (person)', "hysto" (< "hysterisk") 'hysterical (person)', "intello" (< "intellektuell") 'intellectual (person)', "neggo" (< "negativ") 'negative (person)', "pretto" (< "pretenti\246s") 'prenentious', and "v\228rdo" (< "v\228rdel\246s") 'worthless' The O3 group, finally, contains derivations such as "dummo" (< "dum"), 'stupid person', "fetto" (< "fet"), 'fat person', "fyllo" (< "full"), 'drunkard', "hygglo" (< "hygglig"), 'nice (person)', "pervo" (< "pervers"), 'pervert', "pucko" (< "puckad"), 'stupid (person)', "slappo" (< "slapp"), 'lazy person', "slemmo" (< "slem(mig)"), 'disgusting person', and "svullo" (< "svulla"), 'person who eats a lot, fat person'

    4.10 Finland Swedish Maria Luoma and Jan K. Lindstr\246m informed me that most of the Swedish examples I gave would not normally be used in the Swedish-speaking parts of Finland, the slang register there drawing on Finnish rather than on standard Swedish. To the extent that the Swedish O-words are used (e. g. "pucko" 'stupid person' and "fyllo" 'drunkard'), they seem to be perceived as influences from Sweden.

    4.11 Finnish Jan K. Lindstr\246m gave a few examples from Finnish slang whose phonology is reminiscent of the O-words (although the <o> would in some cases be realised as a front rounded vowel due to vowel harmony). His list includes apocoped items such as "skitso" (< schizophrenic), but also a number of derivations, mostly deverbal: "juoppo" 'drunkard' (< Fi "juoda" 'to drink'), "sy\246pp\246" 'person who eats a lot' (< Fi "sy\246d\228" 'to eat'), "happo" 'drunkard (lit. 'acid'), "h\228irikk\246" 'troublemaker' (< Fi "h\228irit\228" 'to disturb'), "\228lykk\246" 'intelligent person' (< Fi "\228ly" 'intelligence'). With the exception of the two last examples, all are bisyllabic. Most would not be considered slang, although somewhat colloquial. Finnish slang words are often two-syllabic and ended by /u/ or, in front vowel words, by /y/. This, Jorma Toivainen tells me, is even more common in the baby talk register, particularly in the western Finnish dialects of Ostrobothnia and Tavastland.

    5. History and O-rigins Age Gisle Andersen recommended that claims about the age of these words be forwarded with caution. Though usually assumed to be quite recent, at least one o-word was attested as early as in 1883, namely "bucko" 'strong, dominant man'. As for "weirdo", the Oxford Dictionary of Modern Slang gives 1955 as the year of first attestation. Joe Hilferty even thinks he recalls having encountered O-words in Shakespeare, possibly in "Othello" (the title of which has three syllables, alas...). Even if the bulk of O items are not that old, the fact that the Marx brothers adopted their stage names sometime in the first two decades of the 20th century could perhaps be an interesting clue in determining the age of the O-words. As for Swedish, I personally perceive most O-words as fairly recent in Swedish (with one or two exceptions), and indeed, the majority of them are virtually unknown to middle-aged speakers. However, students of literary history at Uppsala university were allegedly referred to as "litton" (where the "-n" is a plural suffix) in the 1950s.

    Origin Latin compunds and apocope One common suggestion is that the O1 words ending in <-o> through apocope, such as "psycho", "typo" or Dutch "aso" (< "asociaal") 'someone who displays socially unacceptable behaviour' constitute the original source. The pattern would then have generalised to other lexical items, perhaps along with the negative connotations associated with making a typographical error, being psychologically disturbed, or not being well behaved. The more popular connotations of a form such as "hippo" or "auto" as opposed to the more "learned" "hippopotamus" and "automobile" are also mentioned by many of my respondents as a reason for the general semantics of O-words. Ewald Lang also drew my attention to the O turning up in Latinate constructions ("Sino-Tibetan languages", "trait\233 franco-allemand", "anglo-amerikanische Literatur", etc.), where there would be no <o> in the original Latin paradigm. This suggests a generalisation that would increase the number of potential O1 items. Jannis Androutsopoulos suggests that the O1 class are the starting point of German o-ism. English influence would then have contributed to the spread and the emergence of the O2 and O3 categories. In other words, although the O1 items in a way can be seen simply as shortened compounds, they have not only phonologically, but also stylistically/semantically developed in the direction of the other classes in the sense that they belong to a colloquial rather than a learned register, and that they furthermore quite often denote persons of allegedly aberrant psychological or criminal behavior.

    Romani For the Swedish set of O-words, a Romani origin is often claimed, and incidentally, this was done in a newspaper article the same day I posted my query (Sydsvenska Dagbladet, 20th January). Romani has indeed left a heavy imprint on Swedish slang (perhaps even more so than on the slang vocabularies of other European languages), and <-o> being the typical ending of masculine nouns and adjectives, the assumption is that Romani-derived slang words ending thus would have set a pattern for other parts of the colloquial register.

    Irish For American English, Shane O'Driscoll suggests an Irish etymon, namely "o" which in 'Gailge' Irish Gaelic means young and is derived from "o/g" (/ indicating fada appearing above the o) (as in Sean'o = little/young John). He supports this hypothesis by adding that when words such as "wino", or "wierdo", are used in Hiberno-English, they are often preceded by "little". Given the large proportion of Irish-descended Americans, the <-o> would be likely to have entered American English from Ireland. This diminutive/vocative usage of <-o> could perhaps be associated with the what Roger Billerey reports as a "popular language 'tic' or mannerism in the U. S. in the 1950s". The <-o> suffix was then frequently added to terms of address, such as "daddy-o", "Johnny-o", etc. Plenty of examples of this are to be found in Alfred Hitchcock's movie "Vertigo" - which incidentally itself ends in an <o> - from the early 1950s. This suffixation does not change the part of speech of the suffixed word (usually a noun or proper noun), however.

    A Romance origin? Mock Spanish and covert racism An interesting approach is that suggested by Jane Hill in her article "Mock Spanish: A Site for the Indexical Reproduction of Racism in American English" (available at <http://www.cs.uchicago.edu/l-c/archives/subs/hill-jane/#intro>;). Hill presents examples of the affixation of Spanish grammatical elements to English words, e.g. "no problemo", "el cheapo.", and writes: "...the definite article _el_ and the masculine-gender suffix -o are used with English words to give them a new semantic flavor, ranging from jocularity to insult, or to enhance an already somewhat negative connotation of the English word." (this was brought to my attention by Raimund Schiess). Other respondents added examples like "el stupido" and "Hey, here comes el dorko!", suggestin a connexion between the intended deprecating (humorous) effect and the exoticness of the <-o> ending. In pretty much the same vein, Roger Billerey discusses morphophonosymbolism, i.e. the idea that certain sounds have been integrated by certain cultures as evoking certain ideas. "It seems to me", he writes, "that a common thread here is the fact that most of the words listed above and in your examples have certain negative and/or ribald connotations, which might be linked to the presence of the sound [o], in English, French and maybe Swedish (according to your examples)". The relative rarity of word-final <o> in these languages, he suggests "might account for a certain 'outlandish ring' to the derived -o words, which in turn acquire a particular 'perceptional behavior'". The apparent absence of similar constructions from Romance languages other than French might be taken as a support for this view. Jannis Androutsopoulos proposes a deliberate wish to make things sound Romance as the main factor behind the words in class O4 in the taxonomy above. As opposed to the above American examples, presumably no contemptuous value is intended here. The same could perhaps be true for the frequency of O-like words in brandnaming, e. g. "Polo", "Rolo", "Omo", "Flymo". Peter Menzel specifically suggested Italian, where more or less any adjective or past participle can function as a (provided the referent is masculine) o-final noun. Thus, words such as "maleducato" 'one who is uneducated' are generated. Additional nominalising material is sometimes required, as in "tardivo" 'one who is usually or always late'. Some of these forms are found in English literature from the Renaissance and on, and this may have sparked o productivity in English. The suffix "-oso" 'he who is...', as in "generoso" 'generous', "spiritoso", 'one who is funny', "magroso" 'one who is skinny' are perhaps even more reminiscent of O3 usage in English and other European languages.

    Other possible origins A possibly related phenomenon was reported by Sean Jensen: In the 18th and 19th centuries in London, "-o" was a favourite suffix of costermongers when crying their wares: "Cockles & Mussels alive, alive-o!".

    6. Further reading I haven't made a thorough literature search on this subject, but the Danish journal "K&K (Kultur og Klasse)" had a paper by Claus Kristiansen with the promising title " Mondo weirdo video" in no. 1 (1991), pp 98-110. Unfortunately, this very volume was stolen from our university library. A search for "weirdo" in the MLA and LLBA databses also yielded a reference to William Lee's "New Words and Phrases in English" in "Praxis des neusprachlichen Unterrichts" 38 (2), 1991, pp 169-170, but I have as yet not been able to consult that either. Mayrene E. Bentley suggested Mary Ellen Scullen's (1997), "French Prosodic Morphology". (Bloomington: Indian University Linguistics Club), where French O-words are discussed. There appears to be a thesis on Icelandic O-words written at the University of Iceland, to which I don't have the exact reference. Wayles Browne and Eir\237kur R\246gnvaldsson also mentioned the following article on the subject: Oscar F. Jones, "Icelandic Neologisms in -\243", _Word_ (New York) vol. 20, no. 1, April 1964, 18-27. Toivainen, Jorma (1976): "A Hypocoristic Geminate Consonant Suffix in Finnish" (Salzburger Beitr\228ge zur Linguistik 2, pp 17-24. G\252nter Narr, T\252bingen) contains a list of Finnish baby-talk and slang words, some of which seem to belong in the O group. G\246sta Bergman's "Slang och hemliga spr\229k" (Falk\246ping 1964) allegedly discusses the Swedish O-words and their Romani connexion. A short note on some Swedish O-words, "Svenskans nya ord", by Eva B\228ckstedt, can be found at <http://www-b.svd.se/svd/lasvart/nyaord.html#Pretto>;. I recently made some of my students investigate the acceptability of a number of Swedish O-words; the two short reports are available for anyone who reads Swedish.

    Gisle Andersen <gisle.anderseneng.uib.no> Jannis Androutsopoulos <androutsnovell1.gs.uni-heidelberg.de> John Atkinson <johnamech.eng.usyd.edu.au> Johanna Barddal <johanna.barddalnordlund.lu.se> E. Bashir <ebashirumich.edu> Keith Battarbee <keibatmailhost.utu.fi> Mayrene E. Bentley <bentle16pilot.msu.edu> Chuck Bigelow <bandhmaui.net> Roger Billerey <billereysilcom.com> E. Wayles Browne <ewb2cornell.edu> Kevin Caldwell <kdcaldwinterserv.com> John Davis <davisjustice.usdoj.gov> Scott DeLancey <delanceydarkwing.uoregon.edu> Joachim Grabowski <fpjgrummelplatz.uni-mannheim.de> Shelly Harrison <shellycyllene.uwa.edu.au> P\233tur Helgason <peturling.su.se> Joe Hilferty <hilfertylingua.fil.ub.es> George Huttar <george_huttarsil.org> Sean Jensen <seanjseanj.demon.co.uk> P\228ivi Juvonen <paiviling.su.se> Henny Klein <e.h.kleinfarm.rug.nl> Tore Kristiansen <tkcoco.ihi.ku.dk> Ewald Lang <langzas.gwz-berlin.de> Jan K. Lindstr\246m <jklindstling.helsinki.fi> Maria Luoma <mluomaaton.abo.fi> Peter Menzel <pmenzelclub-internet.fr> Shane O'Driscoll <sodris.cl4compapp.dcu.ie> Dave Robertson <droberttincan.tincan.org> Eir\237kur R\246gnvaldsson <eirikurrhi.hi.is> Henrik Rosenkvist <henrik.rosenkvistnordlund.lu.se> Larry Rosenwald <lrosenwaldwellesley.edu> Claire Saillard <claire.saillardlinguist.jussieu.fr> Raimund Schiess <schiessstud.uni-frankfurt.de> Jennifer Spenader <jenniferling.su.se> Jorma Toivainen <toivaineutu.fi> Harald Ulland <harald.ullandroman.uib.no> \216ystein Alexander Vangsnes <vangsneslili.uib.no> Doug Walker <dcwalkeracs.ucalgary.ca> David Weiss" <david_weissgbinc.com>

    For the Swedish O-words, I also used my own native-speaker intuition.

    Any further comments on the subject are most welcome.

    Mikael Parkvall Stockholm