Corpus approaches to
sociolinguistic variation and
semantic change: parce que
bon...(because well…)
Kate Beeching, Reader, Linguistics and French
Head, International Corpus Linguistics Research Unit
University of the West of England, Bristol
Bon, quand même, quoi !
Yeah but all the same, like
A sociolinguistic survey of semantic change.
An investigation of the impact of daily interactional
activity on the meanings of words and how
sociolinguistic factors affect the way in which new
meanings are propagated.
« ….the real entities of language are utterances and
speakers ’ grammars. Language change occurs via
replication of these entities not through inherent change
of an abstract system » Croft, 2000:4
Overview of the talk 1
What do sociolinguists study?
The relationship between diastratic,
diaphasic and diachronic variation
What are discourse/pragmatic markers?
Are they a suitable case for (sociolinguistic)
PISC - politeness induced semantic change
Overview of the talk 2
Case-studies :
 quand même 1500-2000
 bon
 quoi
What is
The relationship between language and
Variationism: traditionally phonological
a linguistic variable such as /t/ may have
two variants /t/ and glottal stop: ‘butter’:
distributional frequencies vary across
populations (and indeed individuals)
Variation : The four
Diatopic variation
Diastratic variation (age, sex, social class)
Diaphasic variation
Diachronic variation
Studies in ‘real time’ and
‘apparent time’
Studies in ‘real time’ investigate differences
observed in the speech of comparable groups
of speakers separated by a significant period
of time
Studies in ‘apparent time’ investigate
differences observed in the speech of different
generations existing at the same time
Synchronic and diachronic
For a while, the start and end-point of the change co-exist in the form of
two different stylistic layers.... A change is, therefore, in the
beginning, a synchronic phenomenon.
Jakobson, 1952/1963:37 (KB translation).
There are no pure varieties of contemporary French, merely
quantitative differences in the distribution of key language variables.
Lodge, 1993 : 232.
Grammaticalization has to be conceived of as a panchronic process
that presents both a diachronic perspective, since it involves
change, and a synchronic perspective, since it implies variation that
can be described as a system without reference to time. Heine,
Claudi & Hünnemeyer, 1991:261
What are pragmatic
Like, sort of, kind of, well, y’know, I mean,
anyway, eh? Glasgow ‘but’
aber, ja, doch, eigentlich, eben, einmal, schon,
bon, enfin, hein, quand même, quoi
very frequent but tend not to appear in
dictionaries and grammars
Defining pragmatic particles
Brinton (1996: 33-35) highlights the following
characteristics of ‘pragmatic markers’:
 marginal forms, difficult to place in a word class
 little or no propositional meaning
 multifunctional, operating on several linguistic levels
 feature of oral, rather than written discourse,
associated with informality, often stigmatised
 appear with high frequency
 gender-specific? More typical of women’s speech?
Variationism and the
use of particles
Variationism is traditionally focused on
Recent studies (e.g. Fleischmann & Yaguello,
2004) suggest that certain DMs can be
identity markers, and can function like
phonological features
To sum up
Distributional frequencies of a small sub-set of
frequently occurring DMs in French
Investigating their correlation with demographic
factors such as the age, sex, educational
background and date of birth
Investigating their etymology and the extent to
which the hypothesis of PISC can be sustained.
Corpus Data
FRANTEXT literary corpus: contains 210 million
words in 3,737 texts from the 16th. to the 20th.
Orléans (ESLO) Corpus (1966-1970) : . 109 hours of
spoken French (902,755 words transcribed);
Beeching Corpus (1988-1990) : 17.5
hours of spoken French, (155,000 words
transcribed), 95 speakers.
Corpus de Référence du Français Parlé "CRFP"
(2002): 40 towns
in France, 400,000 words. See Véronis (2005).
Quand même: 1500-2000
The coalesced form quand même appears to
have started life as a strengthened form of
quand - ‘at the very moment when’
The conjunction acquired a concessive force
(cf. ‘while’) from at least 1500
In the 19th. Century, it appears as an adverb and begins to lose its strong adversative or
concessive sense
In 20th./21st. Century spoken French, it is
exclusively adverbial and may be either
adversative or expressive (hedging/boosting)
From concessive
Je prépare un discours qui la pourroit toucher
Quand mesme au lieu d’un coeur elle auroit un rocher.
(Du Ryer, Pierre, Les vendanges de Suresne, 1636, page 62, Acte 1, scène iv
I’m preparing a speech which should tear her apart
Even though she’d a stone where she should have a heart.
Conjunction to adverb
Et quand même nous ne réussirions pas, nos petites-filles
réussiront. (Marivaux, La Colonie, 1750, page 1851/Scène
And even though we might not succeed, our grand-daughters
Si je meurs, ce sera en t’adorant quand même, ainsi que j’ai
vécu! (STENDHAL La Chartreuse de Parme, II, XXIII).
If I die, I’ll go on loving you all the same, just as I did when I
was alive.
Grammatical and
semantic change
or contrastive
Table 1 Number and relative percentage rates of occurrence of Quand mesme/quand même in theatrical
works in FRANTEXT, used as conjunctions with a concessive vs. temporal function or adverbs with an
adversative vs. relational function, from 1500-2000
Spoken data: adverb>particle
ce n’est pas une ville qui bouge c’est une ville qui a quand
même un cinéma la saison estivale pendant la saison
estivale et deux boîtes de nuit deux discothèques
(Beeching Corpus, 4, 35-–36)
it’s not exactly leaping, as towns go, but it does have a
cinema in the summer season during the summer season
and two night clubs, two discotheques
ça a l’air d’être une famille quand même assez riche
(Beeching Corpus, 1, 647)
It seems to be quite a rich family really.
An excuse or apology
This mode has a familiar tone, more spoken than the first. Robert’s
definition is Il faut avouer, à vrai dire, on en conviendra. To that list,
one should probably add je ne devrais pas le dire mais... In speech it is
a tactical gambit which, by sketching an apparent attenuation of what
might be sensed as the impropriety of an affirmation, can enable the
reinforcement of the latter. … offers a justification for the statement
it accompanies, even a sort of excuse or apology for it. But thereby it
too has an adversative quality, faint and implicit, in that it hints at
contradicting an assumed objection. (Grieve, 1996: 417, my emphasis).
Semantic bleaching/pragmatic
Continuum M1 > M1/M2 [> M2]
Propositional(concession)explicit adversative
adversative Hedging/Boosting Expression
Conjunction 
Hansen (1998) « Acceptance »
Hansen (1998 : 253) claims that ‘the discourse marker
bon is, of course, derived from the corresponding
adjective’; she adds that it is clear that the adjective and
the DM are different: the DM is invariable (uninflected)
and behaves like an adverb.
Hansen suggests that adjectival bon indicates a positive
evaluation of something and that the DM also ‘marks
acceptance in a rather wide sense of the word’. She gives
examples (1998 : 253-254) of interjective and turn-initial
uses which can be interpreted in this way.
Jayez (2004) « mot de la
fin »
The utterance of bon by an agent a mediates
the following conventional implicature : a
believes or desires that a process in train is or
should be ended.
(Jayez, 2004: 4 – KB translates).
Marks the stages in a narrative:
j'avais perdu mon père à douze ans + et je ne connaissais
pas tellement la fabrication + ma mère + a fait tout ce
qu'elle a pu mais eh eh + elle était pas du métier bon +
alors /j'ai cherché, je cherchais/ + à ayant deux frè- deux
frères et une soeur + à leur laisser la place pour t- + avoir
une profession + CRFP PRI-AMI-3
Marks a reformulation:
frère aîné qui avait quatre ans de plus que moi était très gâté
parce qu'il passait de de fille en fille vous comprenez +
tandis que moi j'ai été élevé de bon il a servi de brouillon
pour moi or j'étais apparemment mieux réussi que que
Bon restriction/concession
and hedge
Marks a restriction or concession (bon….mais):
prend ses fleurs en Hollande mais nous c'est que des fleurs
de France + on (n') en prend pas en Hollande + bon il y en
a qui viennent de Hollande mais c'est un fournisseur + en
particulier qui fait des cultivations euh + en France (CRFP,
Hesitatory or hedging
Oui alors bon oui je bon ma fille a bon elle a pas poursuivi
ses études pour la bonne raison c’est qu’on l’a foutue
dehors à l’âge de seize ans
(from Jayez, 2004)
Brémond (2004)
With respect to dialogic situations, Brémond
(2004:7) notes that (KB translates):
The (very frequent) use of the little mark bon in spoken
exchanges rarely indicates agreement, or, at least, it
never indicates agreement without indicating at the same
time traces of disagreement… the use of this mark
seems rather, perhaps by giving the surface appearance
of agreement, to indicate the management of
intersubjective heterogeneity; the use of the little mark
seems to indicate an ongoing negotiation
. [bon] …
might be seen as playing a role in the cooperative
management of the exchange.
Bon – a (surface)
agreement marker
masking what is actually
a disagreement
y’a des haricots verts dans votre plat ↑
ben c’est-à-dire que si on met [pas les haricots
verts on peut mettre de la laitue
t’en veux pas d’haricots verts ↑ ]
Non ↓ j(e ) veux pas d’haricots verts ↓
Bon – a (surface)
agreement marker
masking what is actually
a disagreement
Ca are there green beans in your dish↑
Ca bon (OK)
Mar well that’s to say that if you don’t [have green
beans you can have lettuce
you don’t want green beans ↑ ]
Ca No ↓ I don’t want green beans↓
Linguistic change in
 In spontaneous spoken contexts, the
‘acceptance ’ and `mot de la fin` usages are
gradually being superseded by a new sense
which includes concession or what Brémond
2004 calls ‘traces de désaccord /négociation`
 Would support Traugott ’s (1982) thesis that
semantic change follows a unidirectional path:
Propositional >Textual>Expressive
As it becomes semantically bleached, it can be
used in more contexts and can thus increase in
As Haspelmath (1999: 1062) remarks:
Semantic generalization or bleaching is usually a
prerequisite for use in a basic discourse function,
that is, for the increase in frequency that triggers
the other changes.
Quantitative survey real time
Quantitative survey apparent time
Example 1: a 92 year-old
L1 en contact avec euh l'Ecole normale + alors tant sur le plan
de l'art + tout seul que sur le plan du langage
L2 hum hum
L1 bon et je me suis toujours + attachée à ce que les enfants
parlent + bien + juste + construisent une phrase et réfléchissent
+ réfléchissent bon + vous voyez la formation de l'esprit à
l'école maternelle c'est important + il y a la la formation du
langage + il y a la lecture c'est évident + bon il y a un
minimum de calcul c'est bien évident + mais l'art + euh fait à
mon avis beaucoup l'art et la musique hein la peinture et la
musique c’est ça
Example 2: a 20 year-old
je suis rentrée dans cette entreprise pour un mois + /donc, bon/ c'était court +
mais bon euh ça m'a permis de voir un peu ce que c'était + et euh donc j'ai
travaillé en collaboration avec le D.R.H. et ça ça m'a plu + on a on a fait un tas de
trucs et euh j'ai je me suis occupée de formation informatique pour euh les salariés
+ je me suis occupée des des détachés qui étaient à l'étranger euh avec les
missions les ordres de euh + des exportés etc. donc euh donc ça ça m'a permis et
puis bon j'étais euh j'étais assez autonome + dans le sens où il y avait personne à
côté de moi pour me dire tu fais ci tu fais ça j'avais ça à faire je le savais et il
fallait que je me débrouille + mais bon euh sa- sachant toujours que si j'avais un
problème euh il y avait quelqu'un tu peux m'aider euh oui donc ça posait pas + ça
posait pas de problème puis il y a eu une bonne ambiance + bon c'était dur
forcément c'était la première fois que je travaillais donc euh + mais bon ça ça m'a
vachement plu + et euh + et voilà …..
Compound Forms
Rate of occurrence of bon ben, mais bon and parce
que bon in the ESLO Corpus (1968), the Beeching
Corpus (1988) and the CRFP (2002)
Raw number of
ESLO Beeching CRFP
1968 1988
Rate per 10,000 words
ESLO Beeching CRFP
1968 1988
Bon ben, mais bon,
parce que bon
bon ben rate
mais bon rate
parce que bon
Taux moyen d'occurrences sur 10.000 mots
18-30 ans
30-65 ans
Tranche d'âge
66 + ans
Educational background
Change in the sense of bon
Most occurrences of bon can be classified as
« mots de la fin » as textual, structuring usages
Textual « bon » is often associated metonymically
with contexts to do with restriction or as a
hesitation marker
The « acceptance» sense > « acceptance up to a
point», demurral
Far from being a « mot de la fin », bon opens the
door to co-construction of meaning and
C’est superbe quoi!
Examples from spoken corpora
je suis de nationalité française mais je suis très contente d’être
bretonne je suis fière d’être bretonne quand même quoi (Beeching
Corpus, 77, 211-12).
I’m of French nationality but I am very happy to be Breton I am really
kind of proud to be Breton
ah oui moi je, j'ai un travail qui me plaît beaucoup quoi.(Beeching
Corpus, 16, 312)
Ah yes I I have a job which I kind of love.
c’est superbe quoi! (Beeching Corpus, 5, 126-128)
It’s kind of fantastic!
Politeness markers
bon, quand même and quoi, in their different
syntactic positions, work together to oil the
wheels of social interaction
These usages are associated with spoken,
informal contexts; quoi, in particular, is highly
demotic (stigmatised)
Wheeler (1994) suggests that speakers adopt
a casual style in order to implement Positive
Social payoff in being informal > Positive
feedback loop
“Apparent time” data:
intergenerational usage
in the CRFP
F o rm a lité
6 0 .0 0
D é fé re n c e
C a m a ra d e r ie
ca d r
5 0 .0 0
s v vr
qm r
Taux moyen
h e in r
4 0 .0 0
e n fin r
q u o ir
3 0 .0 0
2 0 .0 0
1 0 .0 0
0 .0 0
1 .0 0
2 .0 0
T r a n c h e d 'â g e
3 .0 0
Language and identity
Linguistic identity appears to conform more to
generational norms than to class or sex
Older speakers tend to make a restrained use of
particles and adopt a formal mode of speech
All speakers use ‘deferent’ markers, this is a
default position and is stable across time
Younger speakers tend to use ‘camaraderie’
markers such as quoi, enfin and bon
Diaphasic and Diachronic
Through strategic use or non-use of particular
particles, speakers can adapt to circumstance,
and their role in the conversation, appearing
more formal (expert), young/old, deferent or
Younger people appear to be moving towards a
less formal mode of politeness - a type of
democratisation: their speech is solidary yet
deferent, warm yet hedged, characterised in
particular by a plethora of PPs
___________►____________► Grammaticalisation _________►_________►
___________►____________► Semantic bleaching ________►_________►
Diachrony {lexical item ► conjunction
{ propositional ► textual
► adverb
► expressive
pragmatic particle
► identity marker
diaphasic frozen/formal ____________►____________►____________►informal
diastratic (age) older
si vous voulez
quand même
The question remains...
To what can we attribute the increased distributional
frequency of bon and quoi?
 Metonymic concomitance?
 Semantic bleaching?
 A change in society?
 Wheeler’s ‘positive feedback loop’? (‘Yesterday’s
informal is today’s formal’.)
 “Face redress is a powerful functional pressure on
any linguistic system.” B & L 1987: 255.
Beeching, Kate. 2005 Politeness-induced semantic change: The case of quand
même. Language Variation and Change, 17, 155-180.
Beeching, Kate 2007a A politeness-theoretic approach to pragmatico-semantic
change. Journal of Historical Pragmatics, 8/1 (2007): 69-108.
Beeching, Kate 2007b Social identity, salience and language change. In
Ayres-Bennett, Wendy & Jones, Mari (eds.) The French Language and
Questions of Identity. Oxford: Legenda.
Beeching, Kate (2007c) La co-variation des marqueurs discursifs bon, c'est-àdire, enfin, hein, quand même, quoi et si vous voulez : une question
d'identité ? In Gaétane Dostie & Claus Pusch (eds.) Special edition of
Langue Française « Marqueurs discursifs, sens et variation » 154/2 : 7187.
Brémond, Capucine 2004 La petite marque bon, l’indice d’un accord en cours
de négociations. Travaux de Linguistique 48 : 7-19.
Brinton, Laurel 1996 Pragmatic Markers in English. Grammaticalisation and
Discourse Functions. Berlin: Mouton de Gruyter.
Brown, Penelope, and Stephen Levinson. (1987 [1978]). Politeness. Cambridge:
Cambridge University Press.
Croft, William 2000 Explaining language change. An evolutionary approach.
Harlow: Longman.
Fleischmann, Suzanne, and Marina Yaguello. 2004. Discourse markers across
languages? Evidence from English and French. In: Carol Lynn Moder and
Aida Martinovic-Zik. (eds.) Discourse Across Languages and Cultures.
Amsterdam: John Benjamins, 129-147.
Grieve, James. (1996). Dictionary of contemporary French connectors.
London: Routledge.
Haspelmath, Martin. (1999). Why is grammaticalization irreversible?
Linguistics 37(6):1043–1068.
Heine, Bernd, Claudi, Ulrike, & Hünnemeyer, Friederieke. (1991).
Grammaticalization: A conceptual framework. Chicago: University of
Chicago Press.
Hansen, Maj-Britt Mosegaard 1998 The semantic status of discourse markers.
Lingua 104: 235-260.
Jakobson, Roman 1952/1963 Essais de linguistique générale. Paris: Éditions
de Minuit.
Jayez, Jacques 2004 Bon : Le mot de la fin. Université de Genève.
Lakoff, Robin 1975 Language and Woman’s Place, New York, Harper &
Lodge, R. Anthony. 1993 From Dialect to Standard, London, Routledge.
Mendoza-Denton, Norma 2002 ‘Language and Identity’. In Chambers, J.K.,
Trudgill, Peter & Schilling-Estes, Nathalie, The Handbook of Language
Variation and Change. Oxford: Blackwell.
Moeschler, Jacques, & de Spengler, Nina 1981 Quand même: De la
concession à la réfutation. Cahiers de Linguistique Française 2:93–112.
Saint-Pierre, Madeline & Vadnais, Marguerite 1992 Du modalisateur au
marqueur de ponctuation des actions: le cas de bon. Revue Québécoise de
Linguistique 22: 241-259.
Traugott, Elizabeth Closs 1982 From propositional to textual and expressive
meanings: Some semantic-pragmatic aspects of grammaticalization. In:
W.P. Lehmann and Y. Malkiel (eds.). Perspectives on Historical
Linguistics. Amsterdam Studies in the Theory and History of Linguistic
Science IV. Current Issues in Linguistic Theory. Volume 24. Amsterdam:
John Benjamins, 245-272.
Véronis, Jean 2005 « Présentation du Corpus de référence du français parlé »,
Recherches sur le français parlé, 18, pp. 11–42.
Waltereit, Richard 2001 Modal particles and their functional equivalents: A
speech-act-theoretic approach. Journal of Pragmatics, 33, 1391-1417.
Wheeler, Max. 1994. Politeness, sociolinguistic theory and language change.
Folia Linguistica Historica, 15 : 149-174.
Winther, André 1985 Bon (bien, très bien): ponctuation discursive et
punctuation métadiscursive. Langue Française 65 : 80-91.

Pragmaticalisation, the gender paradox and semantic change