The interplay of language norms and usage patterns. Comparing the history of Dutch, English, French and German

Historical sociolinguistics has come a long way. Inspired by the empirical approach to spoken language in modern sociolinguistic research, authors such as Romaine (1982) and Milroy (1992) felt the need to apply sociolinguistic research methods to language history. It became important to create reliable databases comprising source materials that were suitable for sociolinguistic investigation. In the absence of spoken language data, the most oral-like written sources were sought for, an enterprise that was theoretically underpinned by the work of authors such as Biber (1988) and Koch & Oesterreicher (1985), who criticized too rigid a distinction of the spoken and written code. The call for written language data reflecting the spoken language as much as possible led to the compilation of various corpora of what are now often called ego-documents, a cover term referring to genres that are considered to have been important in people's private lives and personal experi- ences, such as letters, diaries and travelogues. All too often, however, the traces of the past are exclusively linked to the upper ranks of society. Sources from the middle and lower ranks are notoriously difficult to come by. As these groups made up the large majority of past societies, historical sociolinguistics set itself the task of not only compiling corpora with relatively 'oral' language, but moreover with data from writers who are often neglected in traditional language histories such as women and members from other social ranks than the elite. The approach to lan- guage history that focusses on such sources and that aims to repair the social and gender bias found in many language histories, has come to be known as language history 'from below' (e.g. Elspaß 2005).

In many historical-sociolinguistic studies, traditional language histories are criticized for being one-sided, partial, biased, founded on a limited collection of sources linked to upper-class men, focused on literary language, on print language, and so on. In addition, they are criticized for creating a view of linguistic history that typically runs from medieval variation to present-day uniformity (Watts 2012; cf. the pluricentric approach to standards of English presented in Hickey 2012). Indeed, the history of European languages in post-medieval times is often cast as or limited to the history of standardization. However, it has also been noted (Fairman 2007) that purposely and explicitly leaving aside the more 'standard'-like textual sources found in print, in literature, in elite documents, and setting aside the possible influence of supraregional writing conventions, language norms and prescriptions, may run the risk of presenting another one-sided view of language history. In many of the recent historical-sociolinguistic studies based on large cor- pora of ego-documents, language norms and standardization are hardly referred to, or in any case mainly used to demonstrate the value of the sources 'from below' as far as they show other conventions and non-'standard' writing norms (e.g. Elspaß 2005; Elspaß et al. 2007; Elspaß & Vandenbussche 2007; Hickey 2010b; Dossena & Del Lungo Camiciotti 2012; Nobels 2013; Simons 2013; van der Wal & Rutten 2013; Rutten & van der Wal 2014). Based on the considerable research tradition in historical sociolinguistics that has come into existence over the past few decades, the time has now come to integrate both perspectives, and to reassess the impor- tance of language norms, standardization and prescription on the basis of sound empirical studies of large corpora of texts.