Being the first that has offered the public a grammatical account of Galic, it was recommended by several persons to frame a new alphabet, consisting of letters or combinations, to express all the sounds in the language, without any mute letter. This is impracticable; but though it could be effected, it would only render the etymology more perplexing. It was recommended to write
v instead of bh and mh, and y instead of dh and gh; which if I had done, the inflections of words beginning with b and m, &c. would be indistinguishable. Thus, it could not be known whether voladh was praise, from moladh, praising; or bholadh, the dative of boladh, smell. Rejecting, for these and other reasons, all remarkable changes, I have thrown away some useless consonants, retaining what are necessary to preserve the etymology and express the sounds. (Shaw, 1778, An Analysis of the Galic Language p. xv) 1
In 1778, the second edition of William Shaw’s
Analysis of the Galic Language was published in Edinburgh. In this monograph, Shaw provides a systematic treatment of the morphology and surface syntactic properties of Scots Gaelic, a Goidelic dialect very closely related to Ulster Irish, which is the focus of the present study. Shaw’s Analysis is interesting from the perspective of modern syntactic analysis, since it is — as far as we can tell — a purely descriptive work. Such an approach to grammatical description was relatively unusual during a period when most grammarians had more prescriptive purposes. Shaw himself seems to have been aware of this difference; he wrote (p. xiv):
In this treatise I have entirely confined myself to Scots Galic, and I think I have accounted for every phenomenon in its structure. To reduce to rule a language without books, and having no standard but the judgement of every speaker, is an undertaking perhaps adventurous; but finding the alphabet consisting of eighteen letters, in which it has hitherto been written, so well adapted, that, with a very few combinations, every sound in the language may be easily accommodated, it remained for me, after considering its genius, to raise this system on that foundation. If, nevertheless, it be found defective, it is altogether my own. I cannot, like other Grammarians, be called a compiler or transcriber; what I have delivered is the result of attentive observation.
Shaw’s work is of value both for its descriptive coverage as well as for his insightful — if perhaps inadvertent — indications of a deeper grammatical analysis of this language. For some current generative treatments of Scots Gaelic syntax, cf. Ramchand (1993), Adger (1994, forthcoming) and references therein.
The original ideas for this chapter were presented in various forms at the Second Student Conference in Linguistics at MIT in March 1990, at the 1990 Spring Meeting of the Linguistics Association of Great Britain, and on various occasions at USC. I am grateful to audiences at these meetings for their comments and suggestions. A earlier version of section 2.2 appears as ‘Movement and Mutation in Modern Irish’ in Sigal Uziel & Tom Green (eds.) MIT Working Papers in Linguistics #12.
Subsequently, in light of the work of Kayne (1993) and of Chomsky (1992/1993), a new analysis of VSOX word-order has been developed and integrated into this chapter. Parts of this revised analysis were presented during 1994 at WCCFL XIII (UC, San Diego), at a Linguistics Colloquium at the University of Massachusetts at Amherst (April 1994), and at the First International Conference on Language in Ireland (University of Ulster, June 1994).
Since the original ideas for were first presented, many people have contributed valuable comments and suggestions. I am particularly indebted to David Adger, Joseph Aoun, Abbas Benmamoun, Bob Borsley, Jim Gee, Osvaldo Jaeggli†, Richard Kayne, Jim McCloskey, Máire Ní Chiosáin, Dónall P. Óbaoill, Jennifer Ormston, Ian Roberts, Lisa Travis, Jean-Roger Vergnaud, Maria Luisa Zubizarreta, as well as to two anonymous
NLLT reviewers. As usual, responsibility for errors remains entirely mine.