There exists as yet no definitive description of the present-day phonology of the Northern Isles. A number of young scholars, however, are currently researching topics such as the Shetland vowel system, aspects of quantity in Orkney and Shetland speech, and dialect levelling in young speakers. The final results from this research, which tends to focus on realizations of Standard (Scottish) English rather than traditional dialect, are unfortunately not yet available at the time of writing this text.

The only existing full-length work on Orkney dialect as spoken in the 20^th century is Marwick’s The Orkney Norn (1929). Confusingly, Marwick uses the term Norn both for the all-Scandinavian language once spoken on the islands and for contemporary Orkney dialect. His work is mainly a dictionary of the dialect but with a brief introduction to grammar and phonology and with phonetic transcriptions of all headwords. As the title suggests, it has a marked Scandinavian and historical bias, particularly apparent in the phonology, which takes the Old Norse sound system as its starting-point, simply listing its modern reflections in Orkney. Although contemporary evidence suggests that Marwick’s data are characterized by a touch of “Nornomania” and that he had preconceived notions of “correct” answers from his informants, his work is clearly of great importance for the present study. As a phonetician he seems very competent, and fairly narrow distinctions, such as [o] vs.  have been noted in individual entries.

Shetland dialect as spoken at the end of the 19^th century was carefully documented in the Faroese scholar Jakob Jakobsen’s monumental An Etymological Dictionary of the Norn Language in Shetland (1928–1932). As the title suggests, it, too, has a clear Scandinavian bias but provides information about the language variety as a whole, including phonology (pre-structural, naturally). Jakobsen, who was a trained philologist in the German school, notes very fine distinctions indeed, to the degree that he has been accused of practising “phonetics run riot” (cf. Waugh 1996: 6). Some of his headwords have up to twenty-five different realizations, but there is no indication of a systematic account of vowels and consonants. This does not mean that he should be ignored in a study of Shetland phonology.

The phonological section of The Linguistic Survey of Scotland (LSS) (cf. Mather and Speitel 1986), which above all was designed to elicit vowel systems, included a number of localities in the Northern Isles (thirteen in Orkney, ten in Shetland). John C. Catford, who was instrumental in setting up the survey, took the view that Shetland phonology was unique among Scottish accents in its rich vowel system, palatalization of final /d/, /n/, and /l/, certain consonant mergers and characteristic syllable structure. Before the actual launching of the LSS, Catford found it necessary to do some pilot fieldwork in Shetland, “a phonological reconnaissance”, which resulted in a special Shetland section in the questionnaire, e.g. eliciting Scandinavian-based words expected to be realized with [ø], such as brööl ‘moo’. There was no similar highlighting of Orkney. Catford (1957: 75) assesses Shetland dialect in general as having a “somewhat archaic character”, suggesting that its vowel system may be similar to Scots as spoken in the metropolitan area of Scotland in the 16^th–17^th centuries. Interestingly, aspects of Shetland verbal usage can also be characterized as archaic (cf. Melchers, other volume).

A recent excellent study of Insular Scots, i.e. Orcadian and ‘Shetlandic’, based on data from LSS and considering Catford’s preliminary analyses of vowel systems, can be extracted from Paul Johnston’s chapter on regional variation for the Edinburgh History of the Scots Language (Johnston 1997).

The only existing account of a particular Insular Scots accent is Elise Orten’s The Kirkwall Accent (Orten 1991), an M.A. thesis submitted at the University of Bergen, claiming to be “based on the London School of phonology”, but not making use of the Wells lexical sets.

An interesting source of information is John Tait’s article on Shetland vowels (Tait 2000). Tait, a native Shetlander, first began taking an interest in Shetland phonology for the purpose of creating a workable writing system. He takes a critical view of the LSS material and introduces the concept of “soft mutation”, i.e. the raising of certain vowels before certain consonants, which “provides, along with vowel length, a framework for looking at Shetlandic vowel phonology as a whole”.

With the help of instrumental analysis, van Leyden (2002) has investigated vowel and consonant duration in Orkney and Shetland dialects, taking Catford’s impressionistic observations as her starting-point. Whereas her Shetland data suggested a Scandinavian-like pattern, Orkney showed more affinity with Standard Scottish English.

In addition to the research described above, this presentation draws on material collected for a project entitled The Scandinavian Element in Shetland Dialect, directed by the present writer. The material consists of tape-recordings eliciting phonological as well as lexical and attitudinal aspects. In addition, a great deal of material recorded for the purpose of oral history has been placed at my disposal by the Orkney and Shetland Archives. This is particularly useful since the interviewers are mostly dialect speakers themselves, which means that the informants do not tend to adapt their language.

For the purpose of this publication, recordings were made in Shetland and Orkney during the summer of 2002. Regrettably, however, the presentation will still have a marked “Shetland bias”, since considerably more data and information is available on the most northerly part of the Insular Scots region.