Achtergronden van een tweetalig gesprek in Amsterdam en Friesland

Door Reitze J. Jonkman

In de Neerlandistiek van 7 mei dook in het artikel ‘Afrikaans in Amsterdam’ een interessant sociaal-psychologisch verschijnsel op dat ik mijn vorige leven als taalsocioloog uitgebreid heb bestudeerd, namelijk het tweetalig gesprek. In het onderhavige geval ging het om een Afrikaans/Nederlandse dialoog, in mijn geval ging het om een Fries/Nederlandse tweegesprek. In mijn doctoraalscriptie Friese taal- en letterkunde (Skaadwizers, 1985) over dit onderwerp haalde ik ter inleiding de volgende woordenwisseling in een winkel in de Frans- en Engelssprekende stad Montreal aan (oorspronkelijk in The Montreal Star 26-1-1978).

The other day I walked into a department store and had a conversation which made me feel foolish. It was also frustrating… It’s the kind of conversation I have an awful lot of nowadays… The conversation always goes something like this: I walk up to the counter, intent on buying some socks. “Bonjour,” says the woman behind the counter, smiling. “Est-ce que je peux vous aider?” “Oui,” I smile back. “Je voudrais acheter des bas come ca.” I point to some socks on display in the showcase. “En beige, s’il vous plait.” “Yes, of course, Madame,” she responds in English. “What size?” “er…” I pause, “nine and a half, please.” Our transaction continues smoothly and I thank her and leave the store. But inwardly, the whole time this pleasant bilingual woman is fishing my socks out of the showcase and putting them in a bag and taking the money, I am cursing. Dammit, I want to say. Dammit, lady, why do you always switch to English?… Does my French sound so terrible that you’d rather not converse in it with me?.. Do you recognize an anglophone… and presume I’d prefer to use my own language? … Could it even be that … you’re telling me…that you’re a federalist? …

In Montreal verliep het gesprek met het Frans dus niet zo gemakkelijk als met het Afrikaans in Amsterdam. Het leek mij passend wat enige aanvulling over de theoretische achtergronden van dergelijke gesprekken te geven. Het gaat om de aspecten die een rol spelen bij het gebruik van Fries en Nederlands in dezelfde conversatie, een vaak voorkomend verschijnsel in Friesland. Op grond van de Speech Accommodation Theory (SAT) van Street en Giles in 1982 noem ik dit een ‘Non-Convergent Discourse’. Onderstaande bijdrage van mij maakte deel uit van een paper die ik in 1996 op een sociolinguistisch symposium in Wales heb gepresenteerd.

Non-Convergent Discourse in Friesland 


Haugen refers in his article on ‘Semicommunication’ in 1972 to the fact that communication does not require the participants to talk the same language in a discourse. Despite the growing loss of efficiency in the communication process as language codes deviate, differences in codes can be overcome by speakers if they wish to understand each other. In the bilingual province of Fryslân/Friesland in the Netherlands two related – West-Germanic – languages are spoken: Frisian and Dutch. About 90 percent of the Dutch speakers (standard and dialect) in Friesland claim to understand the minority language Frisian and all Frisian speakers understand (and speak) the state language Dutch. There is a wish to understand each other and, in general, there are no great difficulties for most of the inhabitants in understanding each other’s languages. All the same most discourses are in one code; the bilingual members of the Frisian language group most often converge to monolingual members of the Dutch language group (‘convergent’ language behaviour in the SAT). But a category of Frisian speakers opt for what I call a ‘Non-Convergent Discourse’ (NCD) with non-Frisian speaking persons, so two linguistic varieties are consistently used in one conversation by members of different language groups.

Results from empirical research in Friesland

I start with the size of the NCD phenomenon to indicate whether the phenomenon is negligible or not. In a representative language survey from 1980 non-Frisian speakers (n=295) were asked whether it happened that people started speaking Frisian to them and continued to speak Frisian. One third (35%) said that it happened that Frisians continued to speak Frisian to them. Frisian speakers (n=752) answered the question as to whether they tend to speak Frisian to their neighbours who usually speak Dutch or dialect. About 30 percent claimed to do so (quite) often. Recent figures show that 22 percent of the Frisian speakers (n=747) claim to speak Frisian in return when addressed in Dutch by a salesperson in a shop in an urban setting. An urban shop is chosen because it is associated with non-Frisianess.  

A second aspect is that there are two reasons/motivations for NCD speakers: 1. The speaker is accustomed to speaking Frisian in its own surroundings and expresses himself better in that language 2. The speaker stands up for the right of the use of the Frisian language in Friesland. The first is a reason at the micro-level and the second at the macro-level. These two reasons can occur in combination in different proportions, but for a number of people only one is valid. In Auer (Bilingual Conversation, 1984) also mentioned these two reasons. He called this type of switching preference-related. He starts with the assumption that there is a preference for same language talk in discourse, but he recognizes that this preference is nothing like a universal. The reasons for such a preference are an altogether different issue. By preference-related switching, a speaker may simply want to avoid the language in which he or she feels insecure and speak the one in which she has better competence. But preference-related switching may also be due to a deliberate decision based on political considerations.

The third aspect is the attitude towards non-convergent language behaviour. To find out I taped – among others – two discourses of two persons (matched-guises of the same couple) who do not know each other. The first speaker asks the second person to show him the way to a certain place, the second showing the way. The first discourse is a monolingual dialogue (Frisian/Frisian), a CD, and the second is bilingual (Dutch/Frisian), an NCD. In both cases the person who asks is overtly showing that the explanation is clear to him and saying thanks to the guide. Students of teacher training colleges were asked to make a speaker evaluation of the two ‘guised’ guides with fillers in between. Actually without knowing it they made a comparison between convergent (CD) and non-convergent language behaviour (NCD). The affective evaluations were given on 15 bipolar seven-point rating scales with prestige and solidarity variables. The outcome on the variable ‘decent’ (‘fatsoenlijk’) is illustrative for the other variables how NCD is perceived in relation to CD. The non-Frisian speakers (L1) born in Friesland, and also the Frisian speakers (L1), evaluated the NCD more negatively than the CD. The non-Frisian speakers did so on many more traits than the Frisian speakers. The conclusion is that NCD in these conditions (two people who do not know each other) is disapproved of by non-Frisian observers on certain traits. The Frisian observers do not reject such behaviour as strongly as the other group. It is possible that an NCD between two persons who know each other will have a more positive speaker evaluation, but this has not been tested yet.


Uit de gepresenteerde achtergronden van het tweetalig gesprek in Friesland komt naar voren dat zo’n gesprek op zichzelf wel als neutraal kan worden ervaren, maar dat er wel duidelijk voorwaarden aan verbonden zijn. Een van de belangrijkste is natuurlijk de onderlinge verstaanbaarheid van de beide betrokken talen voor de gespreksdeelnemers, maar het elkaar persoonlijk kennen is wel zo belangrijk. Bij het voorbeeld van het genoemde tweetalig gesprek van Afrikaans/Nederlands in Amsterdam werd wel aan de eerste voorwaarde voldaan, maar aan de tweede niet. Daarom was deze niet-convergerende dialoog een zeer opvallend incident en daarmee het rapporteren op Neerlandistiek waard.