People Who Curse Have Better Vocabularies.

A new study has found that those who have a healthy repertoire of curse words at their disposal are more likely to have a richer vocabulary than those who don’t. Now, that’s my kind of study. Found via the Facebook feed of Slavomír Čéplö (aka bulbul), who comments: “Aivan niin, perkele!” Which is “Quite so” in Finnish plus the most famous Finnish curse word.

The actual study is Kristin L. Jay and Timothy B. Jay, “Taboo word fluency and knowledge of slurs and general pejoratives: Deconstructing the poverty-of-vocabulary myth” (Language Sciences 52 [November 2015]: 251–259); the abstract is here. (This being Elsevier, you’ll have to shell out $55.20 for the fucking article.)


  1. The same authors have published “A child’s garden of curses: a gender, historical, and age-related evaluation of the taboo lexicon” (Am. J. Psychol. 126(4):459-75, 2013):

    Child swearing is a largely unexplored topic among language researchers, although assumptions about what children know about taboo language form the basis for language standards in many settings. The purpose of the studies presented here is to provide descriptive data about the emergence of adultlike swearing in children; specifically, we aim to document what words children of different ages know and use. Study 1 presents observational data from adults and children (ages 1-12). Study 2 compares perceptions of the inappropriateness of taboo words between adults and older (ages 9-12) and younger (ages 6-8) children. Collectively these data indicate that by the time children enter school they have the rudiments of adult swearing, although children and adults differ in their assessments of the inappropriateness of mild taboo words. Comparisons of these data with estimates obtained in the 1980s allow us to comment on whether swearing habits are changing over the years. Child swearing data can be applied to contemporary social problems and academic issues.

  2. An excellent companion piece (and a daring research topic) — thanks!

  3. Athel Cornish-Bowden says:

    Google Translate tells me that “perkele” means “the devil”, which seems quite mild for the most famous curse word in Finnish. However, maybe it also means something else that Google Translate was too delicate to reveal?

  4. Athel Cornish-Bowden says:

    (This being Elsevier, you’ll have to shell out $55.20 for the fucking article.)

    Don’t you mean

    (This being fucking Elsevier, you’ll have to shell out $55.20 for the article.) ?

  5. ‘Perkele’ is indeed ‘devil.’ The other big curse word in Finnish is ‘Saatana,’ whose meaning you can guess. They use devilry rather than sexy stuff for their oaths. There must be some historico-cultural reason for that, but I don’t know what is is.

  6. Remember that someone from another culture might say “What’s the big deal with ‘fuck’? It just means ‘have sex’!” The force is not in the literal meaning but in what the culture makes of it.

  7. The original referent of perkele is the Baltic thunder-god, who was also responsible for rain, mountains, oak trees, and the sky, thus corresponding to Thor or Zeus. His names in the Baltic languages were Lithuanian Perkūnas, Latvian Pērkons, Old Prussian Perkūns, Yotvingian Parkuns, all < Proto-Baltic Perkūnas *< PIE *Perkʷunos < *perkʷos ‘fir, oak, forested mountain’. In the modern Baltic languages the words mostly mean ‘thunder’, but in Finnish the god was identified with the Devil (capitalized) during Christianization. Consequently, perkele still means both ‘devil’ and ‘the Devil’, but is much more frequently used as a swear and as an adjective roughly equivalent to ‘bloody, fucking’.

  8. marie-lucie says:

    PIE *Perkʷunos < *perkʷos ‘fir, oak, forested mountain’

    I am not familiar with Baltic etymology, so I am wondering if PIE *perkʷos has anything to do with Latin quercus ‘oak”. I am thinking of the words for “five”, from *penkw- but ending up as Latin quinque .

  9. David Marjanović says:

    Yes, m-l! 🙂 The sound change from *p…kʷ to *kʷ…kʷ is regular, and it’s an Italo-Celtic innovation.

    -quos < -cus is also regular and happened within Classical Latin. In equos < ecus < equus, the [kʷ] was quickly levelled in from the rest of the declension; nonetheless, all three forms are attested in this chronological sequence.


    What the everloving fuck. Elsevier articles are supposed to cost 30 35 bucks, and Springer tops out around 40.

  10. Trond Engen says:

    It’s also known from ermanic, apparently, although with a semantic shift from “oak” to *furhWo- “pine”. Even the name of the god is attested with the poetic theonyms Fjörgyn “the mother of Þórr” and Fjörgynn “the father of Frigg”. It seems to me that the detivation with -n- must originally have been adjectival, being used for storm gods and mountainous forests in various branches of Indo-European.

  11. What the everloving fuck. Elsevier articles are supposed to cost 30 35 bucks, and Springer tops out around 40.

    You pay extra for curse words.

  12. The sound change from *p…kʷ to *kʷ…kʷ is regular, and it’s an Italo-Celtic innovation.

    How is it regular? In Italic (as manifested in Latin) it only appears in quinque, quercus, and coquō.

    Celtic and some Italic languages did the opposite, regularly changed labiovelars to bilabials.

  13. Trond Engen says:

    I just learned from Wikipedia that Fjörgyn is a synonym of Jorð “Earth” and probably an alternate name for the same godess. This is obviously fertilty cult. But why a word for wilderness as metonymic for the fertile earth? Could Fjörgyn be the virgin earth being impregnated by the Sky god, later reappearing as Gerð “Yard; fenced field”?

  14. Trond Engen says:

    Going out on a limb here. could *perkW- be a compound meaning “before this”, i.e. “before mankind” or “outside the realm of mankind”?

  15. marie-lucie says:

    (*p to *kw) Celtic and some Italic languages did the opposite, regularly changed labiovelars to bilabials.

    Indeed that change seems to be more common in the world than the opposite.

  16. it only appears

    To refute the claim that shmoos are regularly white, it’s not enough to show that only three shmoos exist: you have to actually exhibit a red, green, yellow, blue, or black shmoo. Unless you can provide a Latin word of the form p…qu…, then the sound-change is regular, no matter how small.

    Elsevier articles are supposed to cost 30 35 bucks, and Springer tops out around 40.

    That’s how oligopoly works: one member raises prices (possibly only on some of their products to begin with), and after a discreet interval the rest follow. This maneuver can be repeated indefinitely, requires no illegal collusion as each just observes the others’ published prices, and allows them all to claim that they are merely charging the market price — completely ignoring the fact that there is no actual market here, because the whole thing is propped up as a state monopoly by the copyright laws. Which themselves got to their current insane length and strength by exactly the same process: the first country to switch from life+50 to life+70 was Portugal, under the influence of the Mouse, and all the others could then go along under the banner of “harmonization”.

  17. marie-lucie says:

    JC: a Latin word of the form p…qu…

    What about pecus, pecoris ?

  18. The Baltic thunder-god Perkūnas has nothing to do with ‘oak’. These words are homonymous.

  19. pecus

    Nope, it’s < PIE peḱu- (Skt páśu, Gothic faihu, where ai = long e), so the /k/ was palatalized, not labialized.


    So they may be, but the oak and the thunder-god are associated throughout the Indo-European-speaking parts of Europe; his cult in Lithuania was associated with a sacred oak and an eternal flame.

  20. Trond Engen says:

    Fjörgyn “Outside-ness” or “Before-ness” = “Nature” as opposed to Gerð “Insideness” = “Culture”! For the meaning “mountainous forest, ref. Urwald “primordial forest” > “big, impenetrable forest”. The tree name could be a back-formation after the meaning Ur- was lost.

  21. The things is the word ‘oak’ isn’t attested in Baltic, but ‘thunder-dog’ has parallels in Slavic, thus suggesting a different etymology. See Mallory-Adams (1997) for more details:

  22. I thought David had erred on both the regularity and the direction of change; if you say that there are no p…qu… words in Latin, then I’ll concede on the regularity (De Vaan says procul might be from pro-kwo-, but that could easily postdate the sound change.)

  23. Trond Engen says:

    Yeah, there’s nothing wrong with the “rock” etymology, and the daughters of *perkW- could be back-formed from “mountain tree” independently in some languages after the meaning “mountain” was lost. The identification of the thunder god with the oak tree is still there, but I guess it might be explained by homonymy.

  24. There was, indeed, a later change /kʷ/ > /p/ in parts of Celtic (in which all original /p/ had been lost) and non-Latin Italic, thus partly undoing the earlier change and violating Dollo’s law of irreversibility in evolution, but indeed because language evolution is mostly neutralist, there’s nothing to prevent it.

  25. violating Dollo’s law of irreversibility in evolution

    Its linguistic analogue would be violated if, say, French developed back into Latin. The reversal of a single trait can well happen in living things as well as languages.

  26. (saved too soon)

    … Why, the very basketball team of Oaklahoma City is named the Thunder.

    In any case, I don’t actually know that there are no p…qu… words in Latin, just that I haven’t found any, excepting of course those with prefixes in p- and qu- in the root, like postquam, proloquor, etc.

  27. marie-lucie says:

    Dollo’s law of irreversibility in evolution

    I had to look this up as it was totally unfamiliar. From what I understand of the Wikipedia article, it concerns the evolution of biological structures or parameters within those structures, such as whether an originally coiled shell, once coiling has disappeared, can become coiled again (of course this is about the general type of shell, not about an individual shell). It seems to me that irreversibility in such cases concerns much more complex structures, in which some patterns may be more or less active or useful (sorry, I don’t know which technical terms would be appropriate here). In phonological evolution, some changes are more likely than others (such as lenition of consonants between vowels, rather than fortition) because they are caused by some natural tendencies of the human vocal apparatus, but those tendencies can be counteracted by a speaker’s deliberate attention to their pronunciation for a variety of reasons, as well as by a number of other factors (a consonant’s position in the word, influence of stress, and more). The range of variation and later change possible when starting from a specific proto-phoneme can be quite large, but there is no reason for a previously discarded phoneme or pattern to be reinstated in a language variety, for various reasons (such as borrowing or other foreign influence) which have nothing to do with the previous structural usefulness or coherence or other quality of the disappeared pattern. So for instance, the appearance of p in a Celtic language which had lost it centuries ago does not have to be a “reintroduction” related to that loss. It could be argued that the previous loss of p left a “structural gap”, an apparent weakness which demanded to be rectified, but the language in question apparently did quite well without it for quite a while.

    As for the Celtic-Italic interchange p / kw (here meaning a change which could occur in either direction), I suppose that there was originally some allophonic, sociolinguistic or dialectal “free” variation between the two sounds, which stabilized as one or the other phoneme in different dialects/languages?

  28. m.-l.: the *p> kʷ in Italic is conditioned by the following kʷ, but the *kʷ>p sound changes in Celtic and the Italic dialects are unconditioned, I believe. That complicates the model of simple free variation between the two sounds.

  29. marie-lucie says:

    Y : That complicates the model of simple free variation between the two sounds.

    Not necessarily, if the two changes happened at different times.

  30. m-l:

    Eh, it’s more complicated, partly because the various changes may have happened at periods of unity or in parallel in different groups. But
    the relative timing was:

    1) p … kw became kw … kw in throughout Celtic and Italic, either during the hypothetical Italo-Celtic period or independently.

    2) p was lost throughout Celtic.

    3a) In Oscan and Umbrian (so-called P-Italic), kw > p everywhere. Latin and Faliscan (so-called Q-Italic) remained unchanged.

    3b) (before or after 3a). In Gaulish and Brittonic (so-called P-Celtic), kw > p everywhere. Irish and Iberian (so-called Q-Celtic) remained unchanged.

    It’s practically certain that 3a and 3b are independent developments. It has long been disputed whether P-Celtic vs. Q-Celtic (there are a bundle of other sound changes that go with this one, it’s just the one chosen as a label) is the deep split, or whether the changes happened independently on the Continent and the British islands. The four Celtic groups may represent a linkage, making it impossible to reconstruct a simple tree.

    And of course you’re right that Dollo’s law applies to large assemblages, not to individual changes. French will never become Latin again, even if it should happen to lose vowel nasality or fricatives other than /s/.

  31. Curse words = AmEng; Swear words = BrEng ??

  32. David Marjanović says:

    See Mallory-Adams (1997) for more details:

    Oh, nice! This explains why the *kʷ isn’t there in Slavic.

    From there*:
    “This association is further reinforced by the Germanic tradition that Þórr strikes his primary foes, the giants, when they hide under the oak tree, one of the most frequent trees struck by lightning in a forest, but he cannot hit them when they hide under a beech, a tree that is very rarely struck by lightning.”

    German has a whole list of proverbs about where to supposedly go in a thunderstorm:
    Von den Eichen sollst du weichen “from the oaks you should recede”
    Die Buchen sollst du suchen “the beeches you should seek”

    * Actually, I wasn’t shown the page. So I changed the address to another top-level domain. The next page, though, remains invisible when I change the TLD (or the language).

  33. David Marjanović says:

    It seems to me that irreversibility in such cases concerns much more complex structures

    Yes. There isn’t actually a law here – it’s just a matter of probability.

  34. @David Marjanović: It’s remarkable how many deep results (such as the Second Law of Thermodynamics) are ultimately instances of the fact that really unlikely things effectively never happen.

  35. David Marjanović says:

    Well, yes.

  36. unlikely things effectively never happen

    Or happen just once 😉

Speak Your Mind