Edward Lhuyd, travelling lexicographer
The major research tour which Lhuyd undertook, accompanied by his assistants Robert Wynne, William Jones, and David Parry, began in May 1697. By the thirteenth of May he was in Wales, at Chepstow in Monmouthshire. He toured Wales for more than two years. In late July or early August 1699, he crossed the Irish Sea to Dublin, and travelled north to Antrim. Thence he took the ferry to Scotland in late September or October—‘Twas too late by the time we came to Scotland to search for plants’—and travelled through parts of the Highlands and Lowlands, returning to northern Ireland by boat at the very end of January 1700. He then travelled through Ulster, Connaught, and Munster, reaching Killarney in the far south-west of Ireland by 22 July. By the last week of August he had arrived in Cornwall, probably direct from Ireland. In January 1701 he sailed to Brittany, returning in March of that year.1
The Welsh section of Lhuyd’s tour was the one in which he had least need to make wordlists from oral information: the everyday living language was his own, and the work that he did on Welsh in his travels—he was of course also busy with natural history, including some successful fossil-hunting, and doubtless with antiquities—was therefore chiefly directed towards the literary language.2 So, for instance, he obtained a manuscript of a Welsh-Latin dictionary by Thomas Wiliems (whose Latin-Welsh dictionary had been abridged to form the basis of the Latin-Welsh section of John Davies’ Dictionarium duplex), which he valued as an ‘Exposition of . . . obsolete Welsh words’; it was indeed a work in the learned Welsh tradition.3 He was thinking comparatively
- 1 An overview by R. T. Gunther is in Life and Letters of Edward Lhwyd 332; further details of the Scottish section are in Campbell and Thomson, Edward Lhuyd in the Scottish Highlands xvi-xxi (with a map ibid. xii; quotation from Lhuyd’s letter of 29 January 1700, probably to Thomas Molyneux, ibid. 6-8 at 6), and the Irish section is covered by Campbell, ‘Tour of Edward Lhuyd in Ireland’ (summary at 228), correcting Gunther on some points, and by R. Sharpe in O’Flaherty, Letters 96-108; the assistants are named by Ovenell, Ashmolean Museum 87.
- 2 For fossils, see Lhuyd, letter of 28 February 1698 to John Lloyd, in Life and Letters of Edward Lhwyd
- 3 For the Welsh-Latin dictionary, see Lhuyd’s letter of 25 May 1699 to an unknown addressee, in Life and Letters of Edward Lhwyd 417; it is identified in Burdett-Jones, ‘Dau eiriadur Henry Salesbury’ 249 n 17, and described in Report on Manuscripts in the Welsh Language 2: 336. For Wiliems in the learned tradition, see C. Davies, ‘The Dictionarium duplex’ 150, and for him and John Davies, see ibid. 151-3 and 156, and E. D. Jones, ‘Brogyntyn Welsh manuscripts’ 316-25.
Small Dictionaries and Curiosity. First edition. John Considine.
© John Considine 2017. First published 2017 by Oxford University Press.
about Welsh at this stage: in October 1697, he sent a message via the antiquary Thomas Tanner to find out whether a mutual friend ‘has any Cornish MSS. and whether he has heard of any manuscript Dictionary in that language’, and he returned to the possibility of a Cornish dictionary in a later letter in which he expressed the hope that the ‘diligent collector of Cornish antiquities’ John Keigwin might have compiled one. In both of these letters, he presumably had in mind a Cornish dictionary which would be wholly or at least partly based on written records rather than on fieldwork or lived experience. By July 1698, Lhuyd was trying to contact the Breton antiquary Paul-Yves Pezron, who had published on ancient chronology and was supposed to have ‘compos’d (but that I presume is in Manuscript) a Celtic Dictionary, though he expressed mild scepticism at the greater project in which he knew Pezron to be engaged: ‘This gentleman, so the Dr [Lister] informs me, labours to prove all Europe and the Greek language originally Celtic; for he acquainted Dr Lister he had 800 Greek words that were manifestly Celtic’ (Leibniz would likewise see ‘a little too much mythology’ in Pezron’s arguments a year later). The word Celtic had been used of some of the inhabitants of Gaul and their language by ancient authors, and thence of the Gaulish language by writers in French and English since the sixteenth century. Pezron’s list of eight hundred words was not, however, of Greek words which resembled words directly attested from ancient Gaulish, for which the evidence was scanty (about ninety Gaulish words, for instance, had been gathered by Boxhorn, and these included personal names, place-name elements, and some doubtful forms). Rather, it was of Greek words which resembled words in modern Breton, which was taken to be a faithful reflection of Gaulish. We shall return to this wordlist.
In the following month, Lhuyd was still thinking about Pezron’s historical arguments, and was still trying to develop his own:
His notion of the Greek, Roman and Celtic languages being of one common origin, agrees exactly with my observations. But I have not advanced so far as to discover the Celtic to be the
Mother-tongue, tho’ perhaps he may not want good grounds, or at least plausible arguments, for such an assertion. The Irish comes in with us [i.e. shows some affinity to Welsh], and is a dialect of the Old Latin, as the British is of the Greek, but the Gothick or Teutonick, tho’ it has also much affinity with us, must needs make a Band apart.
Discovering ‘the Celtic to be the Mother-tongue’ meant claiming that Latin and Greek were descended from Gaulish, and this was clearly eccentric by the highest standards of the time: the similar claim made for Flemish by Goropius Becanus had long since been laughed out of court. Lhuyd’s own suggestion was that Irish and Welsh (his ‘British’) were descended from Latin and Greek respectively, and that the Germanic languages were likewise related in some way to all the others he mentioned, but were not close to any of them. So far, then, his picture of the origins of the languages of the British Isles was moving slowly; if anything, he was separating Irish and Welsh further than the language of the ‘Design’ of 1695 suggests.
Lhuyd’s understanding of the relationships between Irish, Welsh, Latin, and Greek would change as he began to make wordlists of living languages. Between the late summer of 1699 and late July or August 1700, he was in Ireland and Scotland, and for part of this time he had access to speakers of Gaelic. By no means did this prevent him from continuing his study of antiquities and, as far as the season allowed it, natural history: he visited the great prehistoric monument at Newgrange in Meath and the Giant’s Causeway in Antrim, located and when possible bought Gaelic manuscripts, botanized, studied traditional amulets, and noted inscriptions.
All these activities are reported in his letters, but another, which would have momentous consequences, was not. We know about it from one of Lhuyd’s surviving manuscripts, a little vellum-covered notebook, with a flap at the fore-edge of the front cover for extra protection when the book was carried in a pocket. The names of people who might sell him manuscripts are scribbled sideways on the first written page: for instance, ‘Mary Doyle a Carpenters wife at Kiss-arse lane formerly sister in law to Ardel Mac-mahon has or lately had physic MSS in Irish to sel.’ They are a reminder of the ways in which the old literate culture of Ireland and Scotland was falling apart by the time Lhuyd was inquiring after it, the manuscripts which would have been handed from one learned person to another turning into cheap commodities. After this page, the book begins with a wordlist. This is an Irish-English classed dictionary, compiled by obtaining equivalents for the entries in Ray’s Dictionariolum trilingue from a bilingual informant or informants. The Dictionariolum was an excellent choice for this purpose, because as we saw above, it was based on Ray and Willughby’s innovative prompt-list for lexicographical fieldwork. Before the end of August 1701, Lhuyd was to acquire a copy of a partial translation of the Dictionariolum into Scottish Gaelic, made before 1692 by a Scottish minister called Robert Kirk who died (or, according to a later tradition, was carried off by the fairies, whom he had annoyed by his study of popular beliefs about them) in that year. We shall return to his eventual publication of this wordlist; for now, we should note that it would be quite a coincidence if Kirk and Lhuyd had both chosen independently to translate the Dictionariolum into Gaelic, and that it is therefore not unlikely that Lhuyd had already seen and been inspired by Kirk’s Scottish Gaelic translation of the Dictionariolum before he made his own Irish one.
The neat and even handwriting of Lhuyd’s Irish wordlist—in two alternating hands, the first being his own and the second that of an amanuensis who enjoyed imitating Gaelic letter-forms—suggests that it was written up from rough field notes, which had perhaps been taken in an interleaved copy of the Dictionariolum or on loose papers with headings copied from that dictionary. Lhuyd found, as Ray and Willughby had, that his informants could not provide equivalents for every word; a section of names of plants has the English words with no Irish equivalents, evidently written in the hope of finding a good Irish botanist who could supply the latter, and there are many other blanks and occasional signs of hesitation, such as ‘Myk-vara is probably A porpoise’ (muc mhara ‘pig of the sea’ is indeed a porpoise; and since the sound represented by mh is [v], Lhuyd’s transcription was good). However, the dictionary is a good beginning as far as it goes. This manuscript dictionary must belong to Lhuyd’s first Irish journey, before his visit to Scotland, because at the back of the little vellum-covered pocket- book in which it is written, he noted some useful phrases in Irish—‘Is drink sold here? . . . How many pence the pint? . . . What shall I give for this book? . . . Is this the right road?’—and only subsequently added their equivalents in Scottish Gaelic, so the order of writing was surely (i) Lhuyd writes in the front of the book, (ii) Lhuyd adds Irish material at the back of the book, (iii) Lhuyd goes to Scotland and adds Scottish Gaelic to his Irish material.
Another page of notes at the back of the same book shows Lhuyd applying and systematizing the knowledge which he had gained as he compiled the Irish dictionary. It presents three columns of equivalent words: in the first, Lhuyd believes that he sees Welsh [g] in regular correspondence with Irish [f] as in Welsh gwyn and modern Irish fionn ‘white’, in the second, Welsh [p] with Irish [k] as in Welsh pump and Irish rnig
‘five’ (the terms P-Celtic and Q-Celtic are still used to divide the Celtic languages into two groups, one including Welsh with forms in [p] corresponding to the forms in [k] in the other group, which includes Irish), and in the third, Welsh [h] with Irish [s] and [f] and English [s] as in Welsh hen and Irish sen ‘old’ or Welsh had ‘seed’ and English seed (the point about Irish [f] was an afterthought, and was Lhuyd’s only systematic mistake in the tables of equivalences). When he wrote this page, he understood as nobody before him had done that there was an affinity between Welsh and Irish which could be demonstrated by pointing to regular correspondences of sounds. Claims that given sounds were, as a general rule, interchangeable were nothing new, but this was a precise, language-specific observation. Likewise, as we have seen, the possibility of a relationship between Irish and Welsh had been a matter for quite frequent speculation since the sixteenth century, but Lhuyd was demonstrating, not speculating.
The approximate date of the breakthrough can be ascertained. The page written immediately before the page of equivalences is a dialogue in Irish between Lhuyd and the ferryman taking him to Scotland in late September or October 1699, so the page of equivalences belongs to the last part of 1699 or to January 1700 but no later: Scottish Gaelic material gathered before the end of January was added to the bottom of one of the columns after the rest of the page had been written. It may be possible to narrow the date down further, for it is striking that the one liminary poem in Glossography which refers to Lhuyd’s discovery of the regular sound correspondences between Welsh and Gaelic is the work of Colin Campbell of Achnaba, the polymath minister of an Argyllshire parish, to whom Lhuyd talked about his work at Inverary, around November 1699; could it have been that conversation in which Lhuyd explained his philological insight for the first time? To recapitulate, the Irish wordlist was compiled between August 1699 and the crossing to Scotland in September or October. The page of equivalences was written after the crossing to Scotland, but before late January 1700, and very possibly in November 1699. So, making the wordlist of the living language— and not, as Toland claimed, reflecting on the fragments of ancient Gaulish—led Lhuyd to his great philological discovery, and it led him thither without delay.
Lhuyd soon hoped to build on the achievement of the Irish dictionary, writing to Campbell of Achnaba to ask for information about ancient monuments, amulets, ‘peculiar Games and customes’ and similar topics, and, before all these requests, stating that ‘An Interpretation of y Nouns in Mr Ray’s Dictionario[lum] Trilingue; with y
Addition of y Verbs and Adjectives in ye vulgar Nomenclatura, into your Western Ersh would be very acceptable’. The specification of ‘your Western Ersh’ suggests an awareness that there were different regional varieties of Gaelic to be surveyed, even within Scotland: the Gaelic of Argyllshire is not western from the point of view of Ireland, but from that of more easterly parts of Scotland, so Lhuyd was contrasting it with other kinds of Scottish Gaelic, not with Irish. Again, this may suggest that Lhuyd already had Kirk’s wordlist, which is indeed from a location to the east of Argyllshire. Lhuyd spent most of January 1700 on the Kintyre peninsula, waiting for favourable winds so that he could return to Ireland, and complained to Lister that he had had nothing to do apart from adding shells to his, or the Ashmolean Museum’s, collection. In fact, his time there had not been quite as empty as he suggested, for by 29 January 1700, he could write to an Irish correspondent that ‘I translated Mr Rays Dictionariolum trilingue into the highland Tongue, and find about two thirds of it to be a manifest Dialect of the British I intend to paralell all these words with those used in the three provinces of Ulster Connaught and Munster’ (in other words, to make a union wordlist which would bring together his existing Ulster and Argyllshire vocabularies with further material from Connaught and Munster, through which Lhuyd planned to travel).
This translation is preserved in a vellum-covered notebook similar to the one in which Lhuyd wrote his Irish vocabulary, many pages of which are stained from its use to press botanical specimens. The book starts with a few bibliographical notes, some being transcribed out of Robert Sibbald’s Scotia illustrata, a wide-ranging natural history of the sort which Lhuyd himself intended to write. These notes soon come to an end, to be followed by a page of Highland place-names and the translation of the Dictionariolum trilingue. This is written as an English-Gaelic wordlist, and therefore has the English prompts running down the left-hand edge of every page, clearly written before the Gaelic was added: its Irish predecessor had had Irish on the left and English on the right, but this meant that when an Irish equivalent was not known, it was hard to decide how far from the left-hand edge of the page the English word should be written. (This improvement of technique is further evidence that the Irish wordlist was written first, before the tour of Scotland.) The dialect features of the Scottish Gaelic wordlist suggest interviews with informants at several locations on the Kintyre peninsula. It runs to 1665 entries, a good proportion of the 2732 on Lhuyd’s prompt-list;
even the botanical section which had defeated his Irish informants had equivalents for 94 of 212 plant-names.
In 1700, after Lhuyd’s second transit of Ireland, he was feeling the need of a historically oriented Irish dictionary. He had acquired a number of manuscripts,
but the ignorance of their criticks is such, that tho’ I consulted the chiefest of them, as O Flaherty (author of the Ogygia) and several others, they could scarce interpret one page of all my manuscripts; and this is occasioned by the want of a Dictionary, which it seems none of their Nation ever took the trouble to compose.
Any texts in Irish, he added, would be of use ‘to any that would compose a Dictionary of their language, a hint that he was entertaining the idea of making such a dictionary himself. What he said about Irish learning was a distortion of the truth. The historian Roderick O’Flaherty (Ruaidhri Ua Flaithbheartaigh) certainly could read and understand medieval Irish manuscripts, and although Lhuyd may not have seen an Irish dictionary in O’Flaherty’s hands, another Irish antiquary had recently shown him a manuscript copy of the monolingual hard-word dictionary Focloir no sanasan nua, compiled by the learned historian Micheal О Cleirigh (Michael O’Clery), which had been published at Louvain in 1643 (Lhuyd admitted ungratefully that he had been told, ‘but how truly I know not’, that an Irish dictionary had been ‘lately printed’ at Louvain). Lhuyd’s words are better evidence for his own ignorance than for that of the Irish antiquaries whom he had encountered. In 1700, he was unaware of the ancient tradition of Irish lexicography as well as of its modern heirs: he actually owned a fourteenth-century Irish manuscript which included collections of glosses, but had not studied it carefully enough to find them (reading medieval Irish script is a specialized skill). He was unaware that there was a copy of О Cleirigh’s dictionary in the Bodleian. He had not yet heard of the unpublished Latin-Irish Vocabularium Latinum et Hibernum/Focloir Lainne agus Gaoidheilge compiled by the friar Richard Plunkett (Risteard Pluinceid). In the next three years, he started to engage with medieval Irish lexicography, found the Bodleian copy of О Cleirigh, and borrowed a manuscript of Plunkett.
Lhuyd was able to interview fluent speakers of Irish and of Scottish Gaelic for his wordlists of those languages. Cornish, the third language which he investigated in order to understand its relationship with Welsh, was a different matter. As Ray had observed, the langage was becoming moribund in the second half of the seventeenth century. There were, therefore, three kinds of source available for its study: a dwindling community of uneducated native speakers; a small group of educated persons, not all of them native speakers (for instance Nicholas Boson, a leading member of the group, ‘could neither speak nor understand it’ in his childhood), who sought to preserve the language in speech, writing, or both; and a small body of medieval and early modern manuscripts in Cornish. Lhuyd drew on sources in all three categories during his time in Cornwall:
The way that I took to get some knowledge of the Cornish Language, was, partly by writing some down from the mouths of the people in the West of Cornwall, in particular in the parish of St. Just [in Penwith]; and partly, by the like help of some Gentlemen [including John Keigwin
and Nicholas Boson], who wrote out for me many Cornish words____But I got the best part of
my learning from three Manuscript Cornish books, put into my hands by . . . Sir Jonathan Trelawney, Bishop of Exeter; and . . . John Anstis, Esq and the aforesaid Mr. Keigwyn.
These three manuscripts were, respectively, of the medieval dramas called the Ordinalia (Trelawney, who was the head of a great Cornish family as well as being a bishop, had invited Lhuyd to call on him at his Cornish seat to see ‘some Cornish books’, but the others must have been in Latin), the fourteenth-century ‘Pascon agan Arluth’ (‘The Passion of our Lord’), and the sixteenth-century play ‘Gwreans an bys’/‘The creacion of the world’. Lhuyd had the benefit of translations by John Keigwin to help him with all of these manuscripts, so he got off to a much better start studying Cornish than he had when he embarked on Irish. He made a Cornish dictionary, ‘Geirlyer Kyrnweig, which is preserved in a notebook like those he had used for Irish and Scottish Gaelic, but it is not a translation of Ray’s Dictionariolum: no native informant could have given Lhuyd an adequate set of responses to a prompt-list, and so he had to collect his vocabulary ad hoc from written texts and spoken usage.
A hint of his disappointment as he sought for Cornish words is given by a record on one of the endpapers of his copy of Ray’s Collection of English Words. The notes on etymologies which he had sent to Ray around the end of 1690 were based on marginalia in this book. He took it to Cornwall with him, and began a wordlist on the endpapers with the heading ‘Geiriau Kernweg’ (‘Cornish words’). But then he crossed out the second word, and replaced it, so that the heading became ‘Geiriau Kernweg saesneg yngherniw’ (‘Cornish English words of Cornwall’). The list is indeed of English regionalisms, its fullest entry bringing together Lhuyd’s researches in Cornwall and in Wales: ‘In Cornwal, pembr. & Devon. they | for to milk say to milky, for to | squint to squinny; this, thicky | & after most verbs ending with con-| sonants they clap a y. but more | particularly common y lower part of pembroksh’. Lhuyd also added English words from Cornwall to the margins of the book, localizing a couple of them, bettaxe ‘pick-axe’ and easier ‘wide-meshed sieve, to St Just in Penwith, where he had found the best opportunities to encounter spoken Cornish. St Just is a mining town, and the pick-axe and sieve would have been miners’ tools.
From Cornwall, Lhuyd and his companions moved on to Brittany. Some of their work on Breton was done from printed sources. Davies’ Dictionarium duplex includes some Breton words, and Lhuyd was able to obtain copies of two seventeenth-century Breton-French dictionaries, those in Le dictionaire et colloques frangois-breton of Guillaume Quiquer, and in the Sacre college de Jesus of Julien Maunoir, which also includes a grammar. To his annoyance, the latter was ‘so scarce that ’twas my Fortune to meet but with only two Copies, and those in Convents’, so that he could only obtain it by exchange for his copy of Davies, which he thought was a much better dictionary than Maunoir’s. He does not appear to have known the first printed dictionary of Breton, the Breton-French-Latin Catholicon of Jehan Lagadeuc, compiled in 1464 when Brittany was an independent duchy, and printed in 1499; it was a source for Davies, and would have interested Lhuyd as evidence for the state of Breton in the fifteenth century, but it was very hard to come by. As well as using printed sources, Lhuyd also spoke with one or more oral informants, as we know because there are Breton words in his published Glossography which were not available in any dictionary. We can guess the identity of one of these informants: there are striking similarities between the material which can be attributed to them and the material in a later Breton-French dictionary which is explicitly attributed there to the priest and Breton scholar Guillaume Roussel, with whom Lhuyd corresponded. Lhuyd’s work with material which he had written down in interviews can be glimpsed in a couple of his letters. He writes in one of these that he was at Saint-Pol-de-Leon in Lower (i.e. western and Breton-speaking) Brittany, three weeks after his arrival in the province, when a messenger arrived to arrest him on suspicion of espionage:
The messenger found me busy in adding the Armoric words to Mr Rays Dictionariolum Trilingue, with a great many letters and small manuscripts about the table, which he immediately secured____All these papers he tyd up in a napkin.
The ‘letters and small manuscripts’ were surely connected with his work on the dictionary. ‘[M]any of them were writ in Welsh and some in Cornish’, suggesting that he may have been including comparative material in some of his entries; but some of the others must have been fieldwork notes. These small manuscripts are lost, as is the Dictionariolum wordlist with Breton equivalents on which Lhuyd was working when he was arrested. However, parts of two alphabetically ordered Latin wordlists which include Breton, Cornish, and in one case Welsh equivalents are preserved in a collection of Lhuyd’s papers, and they give us some hint of the sort of material which he handled as he examined Breton and its insular relatives.
To conclude the story of Lhuyd as travelling lexicographer, he also had a Manx wordlist in the form of a fifty-five-page notebook, copied fair by his assistant William Jones and perhaps based on fieldwork by Jones himself, or on an interview with a Manx-speaking informant in Oxford, since Lhuyd appears not to have visited the Isle of Man. Like all of Lhuyd’s major fieldwork wordlists, this one was based on Ray’s Dictionariolum. The botanical section was ‘very poorly answered’.
-  Lhuyd, letters of 20 October 1697 and ?1698 to Tanner, in Life and Letters of Edward Lhwyd 348 and369-70 at 370.
-  Lhuyd, letter of 25 July 1698 to Richard Mostyn, in Life and Letters of Edward Lhwyd 378-379; Lhuydhad heard of his work as early as 1691 (see his letter of 12 February 1691 to John Aubrey, ibid. 133-4).Leibniz, letter of 25 August 1699 to Johan Gabriel Sparwenfeld, in his Samtliche Schriften 1.17: 445-8 at446, ‘j’ay peur qu’il n’y ait un peu trop de mythologie dans l’un [Olof Rudbeck, who identified Old Uppsalaas the site of Atlantis] et dans l’autre [Pezron].’ Cf. Roderick O’Flaherty’s letters to Lhuyd of 6 November1702 and 2 July 1706, in his Letters 211-20 at 215 and 299-307 at 305.
-  The first English instance seems to be in a translation: Aneau, Alektor (1590) 82: ‘the title of Frank,which in the Celtique language signifieth liberall and hardie’ (main text in black letter, Frank in Roman,Celtique in italic), from Alector (1560) fo. 52v, ‘tiltre de Franc, qui en la[n]gage Celtique est a dire Liberal,et Hardi.’ Franc/Frank is in fact Germanic, but had since Carolingian times been supposed to come fromthe name of a Trojan hero, Francus (see Borst, Turmbau von Babel 485 and ad indicem s.v. Francus). FrenchCeltique meant ‘Gaulish’, as in Lemaire de Belges, Illustrations de Gaule sig. a3v, ‘Jupiter Celte .ixL Roy deGaule . . . don[n]a le nom a la Gaule Celtique’; for the ancient word, see Collis, ‘George Buchanan and theCelts in Britain’ 92.
-  Boxhorn, Originum Gallicarum liber 6-45; for his predecessors, see Van Hal, ‘Alauda to Zythus’236-48.
-  Lhuyd, letter of August 1698 to Martin Lister, in Life and Letters of Edward Lhwyd 399-400 at 400.
-  See Frederickx and Van Hal, Johannes Goropius Becanus 191-205.
-  See, for instance, his letters of 15 December 1699 to Tancred Robinson, in Life and Letters of EdwardLhwyd 421-3 (Newgrange, Giant’s Causeway, manuscripts, botany), and of 29 January 1700, probably toThomas Molyneux, in Campbell and Thomson, Edward Lhuyd in the Scottish Highlands 6-8 (botany,amulets).
-  Now Trinity College Dublin, MS 1368 (formerly H. 4. 27).
-  Trinity College Dublin, MS 1368, p 4; cf. Lhuyd’s letter of 15 December 1699 to Tancred Robinson,cited in note 10, 423, and Campbell and Thomson, Edward Lhuyd in the Scottish Highlands xxiii-xxiv, 22, 77.
-  Maclean, ‘Life and literary labours’ 331-2, 353-6.
-  The wordlist is pp 5-83. Pp 4-39 are in Lhuyd’s hand, and pp 40-59 in that of the amanuensis; at thebottom of p 54, Irish equivalents in Lhuyd’s hand start to be added to the English wordlist written by theamanuensis; then at pp 60-3, Lhuyd writes both Irish and English; at pp 63-7, the amanuensis writes both(with a few Irish words by Lhuyd at the bottom of p 67); at pp 68-83 Lhuyd writes both.
-  Trinity College Dublin, MS 1368, p 34.
-  Trinity College Dublin, MS 1368, pp 175-74 (working inwards from the back cover of the book);Campbell and Thomson, Edward Lhuyd in the Scottish Highlands xxx, 228.
-  Trinity College Dublin, MS 1368, p 169.
-  See Cram, ‘Edward Lhuyd and the doctrine of the permutation of letters’ 326-31. For interchangeability in general, see Metcalf, On Language Diversity 45 (discussing Abraham Mylius in 1612; see alsoMorpurgo Davies, Nineteenth-Century Linguistics 48) and for language-specific statements like Lhuyd’s,ibid. 21 (with examples from Meric Casaubon in 1650 and Stephen Skinner in 1671); for Lhuyd’s owndistinction between the two, see Glossography 34-5 (first sequence of pagination).
-  Trinity College Dublin, MS 1368, p 170; Lhuyd was working inwards from the back cover of the book,so p 170 was written before p 169.
-  The poem is in Lhuyd, Glossography sig. c2v; for Campbell of Achnaba, see D. Evans and Roberts,‘Introduction’ 26 and their note in Lhuyd, Archaeologia Britannica: Texts and Translations 77-8; for theconversation at Inverary, see Lhuyd, letter of 20 December 1699 to Campbell of Achnaba, in Campbell andThomson, Edward Lhuyd in the Scottish Highlands 4-5 at 4.
-  Lhuyd, letter of 20 December 1699 to Campbell of Achnaba, cited above, at note 10.
-  Lhuyd, letter of 12 March 1700 to Martin Lister, in Life and Letters of Edward Lhwyd 426-7 at 426. Hehad committed himself to make monthly contributions of coins, antiquities, or ‘natural bodies’ to themuseum during extended absences: see Ovenell, Ashmolean Museum 96.
-  Lhuyd, letter of 29 January 1700, probably to Thomas Molyneux, in Campbell and Thomson, EdwardLhuyd in the Scottish Highlands 6-8 at 7; a similar account is in his letter of 12 March 1700 to HenryRowlands, in Life and Letters of Edward Lhwyd 427-31 at 428.
-  Now Trinity College Dublin, MS 1369; overview at Campbell and Thomson, Edward Lhuyd in theScottish Highlands xxxi.
-  Trinity College Dublin, MS 1369, fo. 2r, from Sibbald, Scotia illustrata, 10 item 4 and 11 item 4.
-  Campbell and Thomson, Edward Lhuyd in the Scottish Highlands 91-100 (introduction; localizationof dialect at 92 and 95 n 1, and entry counts at 95), 101-218 (edition and commentary; note on plant namesat 114).
-  Lhuyd, letter of 25 August 1700 to Tancred Robinson, in Life and Letters of Edward Lhwyd 431-3at 431.
-  See the discussion by R. Sharpe in O’Flaherty, Letters 104-7.
-  For the Irish manuscripts used by O’Flaherty, not all of them medieval, see his Letters 388-404 (andfor the spelling of his name, ibid. 1 n 1); for the copy of О Cleirigh’s dictionary seen by Lhuyd, see ArthurBrownlow, letter of 10 June 1704 to Lhuyd, in O’Flaherty, Letters 114 n 357; ‘how truly I know not’ is fromthe letter of 25 August 1700 to Tancred Robinson.
-  The manuscript is described by Abbott and Gwyn, Catalogue 152-4, with a note of Lhuyd’s inscription stating that he had bought the manuscript from Eoin Agniu (Eoin О Gnimh); it must therefore havebeen one of the purchases described in Lhuyd’s letter of 15 December 1699 to Tancred Robinson, in Lifeand Letters of Edward Lhwyd 421-3 at 423. For Lhuyd’s other Irish lexicographical manuscripts, seeO’Sullivan and O’Sullivan, ‘Edward Lhuyd’s collection of Irish manuscripts’ 68-70, and for his study ofmedieval Irish abbreviations, see ibid. 71 and Glossography 304.
-  For Lhuyd’s use of medieval wordlists and О Cleirigh, see his letter of 8 February 1703 to HumfreyWanley, in O’Flaherty, Letters 250 n 122; for his use of О Cleirigh and Plunkett see his Glossography 311-12(translated in Lhuyd, ‘Irish preface’ 196-8); for his borrowing of a copy of Plunkett, see R. Sharpe inO’Flaherty, Letters 115-21 and Lhuyd, Glossography sig. c1v; for the copy he borrowed and the transcriptof it made for his use, ibid. 436 and Abbott and Gwyn, Catalogue 116.
-  For Boson, see his ‘Few words about Cornish’ in Cornish Writings of the Bosun Family 24-31 at 26.
-  Lhuyd, ‘To the courteous and noble inhabitants’ sig. [A]2v, translating Glossography 222; see alsoHawke, ‘Rediscovered Cornish-English vocabulary’ 94-5.
-  Hawke, ‘Rediscovered Cornish-English vocabulary’ 95; Trelawney’s invitation is in a letter by hiscanon Thomas Newey of 3 September 1700 to Lhuyd in Oxford, Bodleian Library MS Ashmole 1816,fo. 452; Lhuyd was in Looe, four miles from Trelawney’s house at Pelynt, in November (see his letter of29 November to Thomas Tonkin, in Life and Letters of Edward Lhwyd 435-7 at 435).
-  Lhuyd, Glossography 253; the manuscript is National Library of Wales, MS Llanstephan 84.
-  Bodleian Library, Ashmole A1, recto of third front endpaper.
-  Bodleian Library, Ashmole A1, recto of third front endpaper; see Oxford English Dictionary s.v. -y,suffix2 (specifying south-west England but not Pembrokeshire) and thilk, adj. and pron., etymological note(which does specify Pembrokeshire).
-  See English Dialect Dictionary s.v. beat-axe and casar; the latter is from a Middle Cornish kazher‘sieve’ and Lhuyd did note ‘Ridara cazher’ as meaning ‘sieve’ in his Cornish vocabulary, National Library ofWales MS Llanstephan 84, p 125.
-  Lhuyd, Glossography sig. b2v and Le Bris, Etudes linguistiques’ 178-9. For John Davies’ Breton, seehis Dictionarium duplex sig. *4v.
-  For its ‘rarete extreme5, see Le Men, Preface, sig. n4v; in 1732, the Breton lexicographer Gregoire deRostrenen was only able to procure a fragmentary copy (Dictionnaire frangois-celtique, sig. a3r).
-  Le Bris, ‘fitudes linguistiques’ 184-5, 189-91, 192-3.
-  Lhuyd, letter of 10 March 1701 to Henry Rowlands, in Life and Letters of Edward Lhwyd 439-42 at 439(long quotation), 440 (Welsh and Cornish), 441 (Breton books); see also his letter of 26 April 1701 toRichard Mostyn, ibid. 443-5. For the boundary of the Breton-speaking area in Lhuyd’s day, see Le Bris,‘fitudes linguistiques’ 187.
-  The wordlist which includes Welsh, from ob to zona, is now Trinity College Dublin MS 1392/3,fos. 1r-27r; the other, from ostrea to zythum, is ibid. fos. 30r-38v.
-  Ifans and Thomson, ‘Edward Lhuyd’s “Geirieu Manaweg” ’ 129-31 (introduction; quotation at 130)and 131-53 (text).