The conceptualising function of Scottish Gaelic preposed adjectives

: This paper focusses on the conceptualising function of Scottish Gaelic preposed adjectives (i.e., AN vs. NA phrases). A combined analysis of a corpus study and interviews with native speakers was applied in the research which underlies the article. Preposed adjectives are often encountered with abstract concepts, verbal nouns, or with words with more complex semantics in general, while plain adjectives tend to qualify more tangible, countable nouns, such as people or objects, as well as pronouns. The plain adjective dona ‘bad’ often conveys criticism, and aosta/sean ‘old’ tend to refer to biological (or physical) age. The paper also addresses similarities with other languages.


Introduction
The aim of the research this article is based on was to investigate the difference between phrases containing preposed adjectives and phrases with plain adjectives (of the same meaning) (deagh-vs. math for 'good'; drochvs. dona for 'bad'; and sean(n)-vs sean or aosta for 'old'), as well as to identify some rules and factors which determine compoundhood in such phrases. This research is based on a corpus study carried out on a subcorpus of the Corpas na Gàidhlig (The Corpus of Scottish Gaelic), as well as on interviews with 10 native speakers to check and refine the observations arising from the corpus study. From the number of results that have emerged from the study, this particular paper focusses on the conceptualising function of Scottish Gaelic preposed adjectives as compared to their plain counterparts.
After the methods having been described in section 2, section 3 deals with the conceptualising function of the preposed adjectives deagh-ʻgood', droch-ʻbad', and seann-ʻold'. In section 4 the results of the interviews with native speakers are summarised. Finally, section 5 comments on the situation of preposed adjectives in the context of Celtic languages, and, 32 Veronika Csonka in section 6, a brief comparison is made between the discussed Scottish Gaelic adjectives and those in a couple of other languages.
In the discussion about adjectives, preposed adjectives (when referring to them separately) are marked with a hyphen to distinguish between the preposed and plain adjectival forms (i.e., deagh-, droch-, sean(n)-vs. math, dona, aosta/sean). In my discussion I apply the spelling for each type which occurs the most frequently in the sources for convenience.

Methods and materials
In order to investigate the difference between the two adjective types, a combined analysis of corpus study and interviews with native speakers was carried out. In the corpus study a great amount of data was analysed from a wide range of sources. However, the majority of the results have a speculative manner. By contrast, the interviews, although less representative due to the limited number of participants, have the overall advantage that they have provided a definite perspective of the few issues discussed through them. The advantages and disadvantages of both methods used are presented in the table below: Table 1 Advantages Disadvantages Corpus study great amount of data analysed results are speculative Interviews personal differences are better limited number of participants; reflected informants are more self-conscious, less natural 1

Corpus study
In the corpus study I wished to compare the use of the preposed and plain adjectives (A+N, N+A): deagh-/math for 'good', droch-/dona for 'bad', sean(n)-/aosta/sean for 'old'. For that purpose I collected all phrases containing these words occurring in a subcorpus of 74 texts from the 205 1 It has to be added that in this particular study, which focusses on the revitalisation of Scottish Gaelic, the self-conscious aspect of the interviews might prove even useful, if the informant lays emphasis on any potential differences in meaning, as it may help to retain the variations of the language.
where x i is the occurrence, i.e., number of tokens for each type and N the number of all occurrences of all types, i.e., the total number of tokens. The standard deviation (the square root of variance): which indicates the expected occurrence of a type in general, i.e., how far it may fall from the average. The sum of these two (mean + standard deviation) gives the threshold value over which the frequency of a type is salient compared to the average. Low occurrences may also be of importance if both types of adjectives are attested, as is the case with droch ghala(i)r and galar dona ʻbad illness' in section 3.2, for instance.
The Irish examples in section 5 are taken from the online corpus Corpas na Gaeilge (The Corpus of Irish).

Interviews
In the interviews 10 informants were interviewed: 6 native speakers 2 from Lewis (1L, 3L, 5L, 6L, 11L, 12L), 1 from Harris (10H), and 3 from South Uist (2U, 4U, 7U). Concerning their age, 4 of them were between 25 and 60, and 6 were 60 or above. Their exact distribution among the age groups was as follows: Table 2 20-30: 1 (Lewis) 1L 30-40: 1 (South Uist) 4U 40-50: 1 (Lewis) 3L 50-60: 1 (Harris) 10H 60-70: 4 (1 from South Uist, 3 from Lewis) 2U; 6L, 11L, 12L 70-80: 2 (1 from South Uist, 1 from Lewis) 7U; 5L Each interview lasted for 30 or 40 minutes, and the test included seven exercises (referred to as sections ( §) in the discussion in section 3) altogether, three of these to explore the meaning and use of preposed and plain adjectives, the remaining four focussing on the use and degree of Scottish Gaelic intensifiers and other issues which are not pertinent to the subject of the present paper. The exercises which are relevant here included mainly translations, and a picture description. They were constructed to investigate conceptuality in preposed adjectives vs. tangibility in plain adjectives; the role of contrast in sentences containing both the preposed adjective seann-and the attributive plain adjective aosta/sean for ʻold'; etc. The productivity of the different types of adjectives was examined by non-sensible or loan words, and the conceptualising role of preposed adjectives was studied by unusual collocates.
According to my observations, plain adjectives qualify tangible nouns, while preposed adjectives convey conceptuality and abstractness. To test this observation, §1 contained tangible nouns: professions, animals, and vehicles. I gave two pictures of each to the informants with two adjectival phrases to be translated (I also used some other plain adjectives for distraction). In §1b the informants had to translate unusual phrases consisting of tangible or abstract entities and the adjective 'good', 'bad', or 'old' (e.g., good feather, old sadness).
In §2 the phrases to be translated were 'good day', 'bad day' with pictures reflecting weather, and 'good night, 'bad night' with pictures implying more complex/abstract meanings. Whether preposed adjective seannreflects traditionality, I aimed to examine with the pictures for the use of seann-/aosta/sean with people, clothes and dances, although this picture description did not really work out as planned, apart from a couple of examples of plain adjective aosta regarding the age of a person. In the corpus the choice for the adjective used with certain nouns appeared to be influenced by the number of the noun. In §3 I asked the plural of certain adjectival phrases to examine the preferences to the adjective in singular and plural.
In §4 the informants had to translate nonsense words and loan words qualified by 'good', 'bad' and 'old'. This section was supposed to identify the default adjective -the adjective used automatically, more productively by the speaker. Loan words may also relate to the default usage of adjectives with types of entities (e.g., object (yoyo), food (spagetti, sushi), abstract (déja vu), etc.).
The main purpose of §5 was to investigate the connection between pronunciation (i.e., stress) and orthography (i.e., hyphenation): to understand the use of hyphen in words such as droch-latha, droch-bhean, and occasionally the difference between the phrase with preposed and plain adjective (e.g., between deagh obair and obair mhath). The section was also meant to check more figurative or specific meanings, such as in drochshùil 'evil eye', droch-rud 'devil', and droch-dhaoine 'criminals/villains'. §6 and §7 served to investigate the use and meanings of intensifiers. Table 3 (overleaf) summarises the structure of the interviews.
The disadvantages of explicit questions and translation lists are obvious: informants tend to use prestigious forms without realising it. Another problem could be that they start seeing a pattern or will not concentrate on the actual collocate, which could influence their word choice -either using the same kind of adjective spontaneously, or (probably less usually) changing it for variation. In neither case do we gain a reliable picture of actual everyday speech. To minimise this problem the translations were mixed up and a couple of irrelevant examples were applied in the questionnaire as an attempt to distract the attention from preposed adjectives.
Due to limitation of time and of the length of the test, some apects of the interviews did not work out in the planned way and only a small number of the questions could be addressed from those emerging from the corpus study. Therefore the chapter on native speakers' judgements is not so high in proportion to the amount of data analysed in the corpus 36 Veronika Csonka study. On the other hand, this part of the research has clarified many of the questions which were addressed in the interviews, and in some cases even questions that I did not specifically raise. These include an insight to dialectal difference between Lewis and the southern islands, the difference between the attributive plain adjectives sean and aosta, the use of dona to express criticism, the use of deagh-in conceptual nouns and that of math in tangible ones. It has proved to be essential in the final distribution of preposed and plain adjectives.

Results of the corpus study
According to the results of the research, nouns qualified by preposed adjectives deagh-, droch-and seann-appear to refer to entities with more complex semantics, while plain adjectives math, dona and aosta/sean normally stand in more pronominal expressions or emphasise the quality in phrases (such as criticism with dona and biological or physical age-reference in the case of aosta/sean). In the following sections, I introduce evidence for the above statement in all three adjectives studied. 3.1 reflects upon the conceptualising function of deagh-in comparison to math, in 3.2 droch-is compared to dona, and in 3.3 the same is discussed for seann-vs. aosta (or sean).

Abstraction with deagh-
The occurrences of obair mhath in 12 out of 1066 examples (1.1%) and deagh(-)obair in 5 out of 908 examples (0.6%) may be close enough to consider obair as a word occurring in both constructions to a similar degree. Although the majority of examples for obair occur with math, in 6 out of 12 tokens obair mhath refers to placement, employment (1a), rather than the work itself, which makes it similar to a physical place. In (1b, c) obair mhath appears to denote a work which presumably has a tangible outcome.
With deagh-it refers to a more abstract concept (1d, e, f). a.
'He received good education; he found a good job; but he still wasn't satisfied.' b. Rinn e obair mhath an sin, gu sònraichte ann am mathematics… 'He carried out good work there, particularly in mathematics…' c. … a nis air faicinn na h-oibreach mhath, ùrail a tha chlann ri deanamh le Beurla… '… now that we have seen the fresh, good pieces of work that the children are doing in English…' d. …'s tha iad an diugh pòsda 's a' dèanamh deagh obair anns an t-saoghal.
'…and today they are married and do a good job in the world.' e. Ma rinn sinn deagh-obair anns na làithean a dh'fhalbh, molaidh an obair sin i fhéin… 'If we did a good job in days that have passed, that work will praise itself…' f. Cha rachainn-sa an urras ort fhèin nach tu a rinn e air son deagh-obair fhaighinn dhut fhèin! 'I wouldn't trust you not to have done it to get a good job/work for yourself!' In the last example deagh-seems to convey the same meaning as math; however, the hyphen may indicate that it belongs to another class for this speaker. Nevertheless, it still seems to be less specific than the previous examples with obair mhath.
As we have already seen in the above examples for obair 'work', deagh-may have a conceptualising function. (I give a further example with comhairle 'advice' in (2).)

Veronika Csonka
(2) a Mhàthair na deadh chomhairle 'oh, Mother of good advice' (in a prayer) If we consider the frequent occurrence of preposed deagh-with words referring to emotions, mental concepts and morality (e.g., deagh(-)dhùrachd ʻgood wish', deagh dhòchas ʻgood hope', deagh eòlas ʻgood knowledge', deagh aobhar ʻgood reason', deagh(-)ghean goodwill, deagh-nàdar ʻgood nature, good temper'), it can easily be understood why it is appropriate for this function. In some cases it is associated with respect (deagh charaid 'good friend', deagh mhaighistir 'good master'), and frequently occurs with verbal nouns as well (deagh ghabhail 'good let', deagh phàigheadh 'good payment', and see deagh oibreachadh 'good working' below). Having an abstract sense is not surprising in the case of deagh dhòchas 'good hope' and deagh chomhairle 'good advice', which do belong in this category, and occur mostly with deagh-. What really is of interest here, is the abstraction present in examples with deagh obair, but usually absent from those with obair mhath. The ability of obair to have a more abstract meaning as well as a more factual one, therefore, may be a good reason for its more frequent occurrence in both combinations.
'In selling condition (lit. 'at the selling stage'), the shoes were "worked" (i.e., fashioned) well and were very attractive.' The word choice may be influenced by semantics in astar as well, where astar math usually refers to distance or size of an area, and deagh astar to speed (note that we perceive distance and size as more concrete compared to speed). The four examples of deagh astar originate from three different sources. One of them (4h) does not fit this theory, referring to distance. However, it is encountered in a poem, from the same writer as one of the other examples, and as such, does not necessarily follow the general rules of the language. There also might be an exception among the examples with math (4i), but it may be ambiguous as well. 'This road took us a long way out from the town.' e. … a tha a' ruith troimh dùthchannan anns am faighear daoimein agus astar math de ùrlar a' chuain aig bèul nan aibhnichean sin… '… that runs through countries in which diamond and a good piece of sea floor can be found at the mouth of those rivers…' f. Bha iad a nise a' dèanamh deagh astar… 'Now they travelled at great speed…' g. Le gaoth bho 'n ear dheas bha an deagh astar aice… 'With the wind from southeast it travelled at great speed…' (i.e., the ship) h. … Bha i deagh astar uap'. (in a poem) '…it was a long way from them.' i. Tha na ròidean anns a' chuid mhóir de'n Fhraing farsaing agus dìreach, agus dhèanadh càr astar math 's gun mhóran coileid orra… 'The roads in a great part of France are wide and straight, and the car would travel at speed (a good distance) and without much stir on them…' By contrast, àm math and deagh àm seem to show different meanings without exception: àm math stands for 'appropriate time' (e.g., Tha sinn an dochas gu'n toir gach ni a tha am Freasdal a' toirt mu'n cuairt 'n a am math fhein, luathachadh latha mor na sithe. 'We hope that everything that the Goodness (i.e., Heaven/God) brings about, will bring the big day of peace closer in its own good time'; … chan e àm math a th' ann dha duine, tha cadal gad iarraidh. 'it's not a good time for a man, sleep wants you'), while (ann) an deagh àm means 'in time, in its time' (Thill sinn an deagh àm air son dìnneir… 'We returned in good time for dinner.'; … rainig iad an Druim-ghlas an deagh ám. 'they reached Druim-ghlas in good time.'; Tha thu dìreach ann an deagh am…'You were just in (good) time.'; Cluinnidh sibh sin an deagh am. 'You will hear that in (good) time/when its time comes.'), or Nuair a tha sinn a' feitheamh an latha anns am bi barrachd smachd againn air ar dòigh-beatha fhéin ann an Albainn, 'se deagh àm a th' ann beachdachadh air an inbhe 's air an obair tha gu bhith aig a' Ghàidhlig 's na tha fuaighte rithe ann am beatha ar dùthcha. 'When we are waiting for the day when we'll have more control over our own lifestyle in Scotland, it's a good time to consider the status and work that Gaelic is going to have and which is attached to it in our country's life.' This latter example occurs in the only formal text among the relevant tokens. (See also deagh dhuine meaning 'the right man/person' as opposed to duine math for 'good/religious man/person' below.) In several cases both combinations deagh dhuine and duine math occur in the same text(s), which can be really useful. In Na Klondykers, for 40 Veronika Csonka example, the two occurrences of duine math appear in neutral, descriptive sentences, meaning 'a good man' … cha do thionndaidh duine math eile an-àirde. '… no one else good turned up.'; 'S e duine math a bh'ann. Iain. Duine snog. 'He was a good man. Iain. A nice man.' In contrast, deaghcan be observed in the sense 'the right man', as in the paragraph below: (5) Agus an uair sin, dh'fhaighnich iad dha Iain an deigheadh e ann. Bha aon àite eile air a' bhàta agus bha Iain aon uair anns an Oilthigh, agus mar sin 's e deagh dhuine a bhiodh ann air an sgioba bheag aca. 'And then they asked Iain if he would go (there). There was one more space on the ship and Iain was once at University, and therefore he would be a good man in their small crew.' (Note that the default collocation for duine is with math, i.e., that occurs in most constructions (28 tokens vs. 2 for deagh dhuine), due to its (concrete) reference to a person.) In the case of cuimhne, which word has a high occurrence both with deagh-and math, I was not able to differentiate between the word combinations -neither do dialects or register/style of writing show any preference to one adjective over the other, not even in the same work (cf. examples from Hiort in (6)). On the other hand, in most tokens of cuimhne the two words are distributed among the sources so that both do not occur in the same work, which suggests there must be personal preference for one adjective or the other. There is only one source (Suathadh ri Iomadh Rubha, from Lewis) where both types occur and a change in meaning can be observed, although only three tokens occur in this source. Here, cuimhne mhath carries the most frequent meaning 'to remember well', whereas deagh chuimhne shares its meaning 'good memory' with cuimhneachan math 'good memories'. Note that the same pattern of pluralisation can be observed in this example as in the case of deagh rùn ʻgood intention' -rùintean math ʻgood intentions', which can be found in the same source. (Furthermore, air never precedes any collocations with math in the corpus.) a. In general, the examples with cuimhne appear to show a random distribution among the writers, indicating their individual preference, rather than referring to dialect or register. This arbitrary usage is rather frequent throughout my data and creates exceptions and uncertainty in most cases. However, some patterns may be discerned, even in cases where exceptions are rather rare.

Abstraction with droch-
The feature of conceptualisation can be observed in the case of other preposed adjectives as well. In this section some more abstract meanings are introduced in the cases of duine ʻperson, man', rud ʻthing', àite ʻplace', bean ʻwoman, wife', boireannach ʻwoman', and galair ʻillness, suffering'.
The only word which is common with both droch-[19] 3 and dona [11] is duine. Both of these can mean 'bad man; bad person', droch dhuine seems to show the meaning 'non-Christian' or 'pagan' (as opposed to duine math -see example (20)), and duine dona has the connotation of a 'grumpy, unsatisfied person'. Regarding droch-in combination with duine, a distinction can be observed between singular and plural: droch dhaoine refers to non-Christians, or bad men but with a religious connotation in all but one source (which refers to pirates ('people who are out of the law'), thus it arguably may carry the same meaning after all); whereas droch dhuine mainly means 'bad man' (whose behaviour or intentions are not acceptable). The various meanings are listed in examples (8)-(10): '-[…] I enjoyed riding that girl. I have never enjoyed a game so much. I had to make loads of shielding, you know? -Bad man.' "Oh, an dìol-déirce truagh!", ars ise. "duine dona! na bithibh a' toir feairt air, car son tha sibh a' dol a dh'éirigh gus am bi e faisg air an latha? A' cosg soluis!" ' "Oh, the miserable wretch!", she said. "bad man! don't pay him any attention, why are you going to get up before it is near daytime (i.e., why are you getting up before daylight)? Wasting (the) light!" ' The first two examples in (8) are from the same source. Compare the second and the third examples, in which duine dona reflects the speaker's criticism of someone (both are dialogues). Dona is common in vocatives, non-verbal statements, or criticism: Àite dona! 'A bad place!'; Bean dhona, cha n-fhiù i,/Cuir g' a dùthaich i dhachaigh! 'A bad woman/wife, she's not worth it, send her home to her country (i.e., place)!'; also: Tapadh leat, a dhuine dona! 'Thank you, bad man/non-Christian!' in the religious poem with the title Fàilte an diabhail do'n droch dhuine 'The devil's welcome to the bad man/non-Christian'. (There may be a similar distinction between droch bhoireannach and bean dhona, discussed below -see (21) and (22).) (9) occurs in a narrative about Paul Jones, spùinneadair-mara oillteil 'a dreadful pirate' (lit. 'sea-robber'). It refers to people outside the law; however, its meaning is arguably the same as in (8). (I have not encountered this specific meaning for duine dona.) (9) 'criminal, villain': Tharruing e mu 'thiomchioll sgioba de dhroch dhaoine mar bha e féin agus ghoid iad air falbh leis an t-soitheach. 'He gathered a crew of villains as he was himself and they plundered with the vessel.' (10a) evidently refers to pagans as it is explained in the text itself, the other example is not so specific but still occur in a religious context.
'A bad man (i.e., not acceptable by religion, without faith) … He himself and the Finch, God's malediction is on them.' Both combinations may occur in more recent texts, although usually with rather similar meanings.
'It is not that the good day wouldn't come, but that the bad person wouldn't wait for it.' This special connotation of dona is present in a couple of texts from around the 70s, mainly proverbs. Droch-may have a similar meaning, attested in a text from 1983; although, the meanings 'grumpy' and 'bad man' may overlap in this example (see (12) below).
(12) Droch dhuine a bh' ann am Paddy Manson -fear à Liverpool a bha cho buaireant', greannach ri cat air lìon-beag! 'Paddy Manson was a bad person -a man from Liverpool who was as annoying, bad-tempered as a cat on a fishing line!' As the various connotations of droch dhuine/duine dona are very closely related (religion regards a bad person as non-Christian, just as an unsatisfied, ever-complaining person can be annoying in other people's eyes, and therefore considered 'bad'), these different meanings can all overlap, and in a number of cases it is hard to distinguish between them: a.
'It is not that the good day wouldn't/doesn't come, but/it is that the bad person wouldn't/doesn't wait for it.' , on the other hand, refers to something more specific -a 'bad issue', a 'crime'; or to the 'devil' (religious connotation), which senses are very similar to those of droch dhuine. In some cases (example (17)), droch rud has a similar meaning as rud dona: 'a bad thing' (connected to the words iomadh, fìor, sam bith; but also to idir). Examples (14a) and (14b) are from the same source; example c appears in the same source as (15a), which thus represents a clear distinction between rud dona as 'something bad' and an droch rud meaning 'the devil' (note the article; in (15a-b) 'devil' (or 'demon') is referred to with two different words). '… that tax/fine, was in proportion to how bad/serious the offence/crime was that he committed.' Example (17), droch rud, is understood as 'a bad thing', but in a more tangible sense, similar to rud dona 'something bad, anything bad' -if droch rud (for other, grammatical reasons) is meant to refer to 'anything bad', it tends to combine with sam bith (see Examples (16a) and (17d)). In example (17a), idir may serve as evidence for droch rud referring to a more specific thing than rud dona, since idir mainly accompanies dona in predicative or adverbial phrases (see (18)), whereas rud in rud dona is used in a more general, pronominal sense. Also consider the proverb Cha d' fhuair droch-ràmhaiche ràmh math riamh 'A bad rower (has) never found a good oar' for the possible distinction between the two adjectives. There are two words meaning 'woman' in Gaelic: the masculine noun boireannach and the feminine word bean, the latter used both for ʻwoman' and ʻwife'. In the case of bean a shift in meaning can be observed between its form with droch-and that of dona: droch-bhean [9] is quite common in proverbs or riddles, probably with the meaning 'bad wife' (see (21a)), while the two occurrences of bean dhona appear to be very similar to duine dona in that they describe the quality of the person they are referring to -the first example in (21b) is from a waulking song, in which bean dhona expresses the speaker's opinion -just as in the similar examples for duine dona above (in (8)); the second example here occurs in a list of qualifying features (also a description). Interestingly, the example of boireannach dona [1] shows religious connotation (although it may qualify/criticise the Christian values of the woman), while droch bhoireannach [2] means something like bean dhona, with the difference that it may convey a more specific, more integrated sense -referring to something more beyond the compositional meaning 'bad woman', to someone who teases and plays with men. Note the use of fìor before dona (which is unusual according to our observations so far) -this may indicate that the abstraction firmly determines a kind of meaning shift in this word (i.e., galar).

Abstraction with seann-
In the case of the adjectives for ʻold' certain cases show a distinction between the preposed adjective for expressions with more complex semantics and the plain adjective for the more tangible meaning of age. Seann-, with its great number of tokens, is highly productive, appearing with all sorts of words, while aosta (appearing much less frequently) does not occur with any noun in significant numbers which cannot be found also with seann-. The plain adjective aosta (and sean) normally refers to the age of a person, animal, or -sometimes -object (e.g., aodach aosta 'old clothes'), while seann-is commonly encountered in fixed expressions, it may convey traditionality (e.g., seann-taigh 'a (traditional) black-house' vs. taigh 50 Veronika Csonka aosta/sean 'an old house' (physical age of the building)), or refer to former types (e.g., seann téipichean céire 'old wax tapes') and roles (e.g., seann saighdear 'old soldier, veteran', seann leannan 'old sweetheart'). Consider also the possible distinction between caraid aosta/sean 'old friend' (describing the age of the person) and seann-charaid, referring to a long-existing friendship. In respect of plain adjectives, sean is commoner in South Uist, while aosta is preferred in Lewis. When both are used, aosta is regarded as more polite, entitles respect, and it shows a somewhat intensified sense (i.e., 'really old') compared with sean ʻold'.
Aosta is the typical qualifier of pronominal words like cuid 'some' and dithis 'two persons'; however, tè 'one (female)' and feadhainn 'ones' can be encountered with seann-in a number of tokens (discussed below). The few verbal nouns in the corpus are all qualified with seann-, and the only loan word with aosta is baidsealair 'bachelor' (which itself has a much more common synonym (fleasgach), which usually stands with seann-), whereas seann-qualifies a great number of loan words otherwise. This suggests that here aosta may highlight the old age of the bachelor.
The three most common nouns both with seann-and aosta are duine 'person/man', bean 'woman/wife' and boireannach 'woman'. All of these show similar patterns. The distinction is not very clear in either case, since both adjectives are present in most sources, with subtle differences in meaning. The collocate with seann-seems to be a neutral compound expression (e.g., 'S ann thachair sean bhean thruagh orm … 'That was when I came across a wretched old woman'), whereas aosta may be used in cases where the quality of being old is important from the speaker's point of view. In certain cases seann bhean 'old woman' may refer to a particular person (… nach ann a chaidh Coinneach a shealltainn air seann bhean a bha air an leabaidh. '… wasn't that that Kenneth went to see an old woman who was on the bed.'), as opposed to statements like Sgreadail mhnathan aosd' agus ghruagach 'Screaming of old women/wives and maids'.
Combinations of the two adjectives occur twice -one with daoine, the other with mnaoi (dat. sg of bean): seann-daoine aosda chaithte shàraich 'weary worn aged old people'; air an t-seann mhnaoi aosd 'on the aged old-woman'. The redundant use of aosta may indicate that seann daoine 'old people' and seann mhnaoi 'old woman/hag (dat.)' are treated as compounds, although both tokens occur in poetry, thus it may only serve as a device for emphasis. (See also Irish examples, such as seanóir aosta ʻold elder' and sean-nós aosta ʻold traditional custom'.) Seann taighean [34] refers to 'traditional houses' or 'black-houses' (see (24a) below). Alternatively, in some poems, it may mean a house where somebody used to live (see (24b)). The only token with plain adjective an tigh aosda ud (from Lewis) may literally mean '(that) old house' (where the quality of oldness is important). There are three more cases encountered with both types of adjectives which could be of interest here, the first of these is a time expression, the other two are the pronominal expressions tè 'one (fem.)' and feadhainn 'ones'. Preposed seann-is the adjective used with words referring to time (like tìm/aimsir and uair). In the case of làithean 'days', most tokens (24) follow this rule and have a very similar meaning. Nevertheless, 3 tokens stand with aosda (all three in poetry). These may refer to a person's age, and/or are connected with cuimhne 'memory'.
Tè and feadhainn, usually exhibiting a pronominal sense, would be expected with aosta, which, however, is not attested in many cases. In the corpus, I have encountered only 1 té aosd beside nine tokens for seann té (although three times in the same poem and further two in two other poems from the same source). Seann té appears to be related to the more informal language of the storytelling register (three tokens appearing in narratives, autobiographies). Another possible explanation for the choice for seann-is related to dialects, as the source of poetry containing five tokens of seann tè originates from South Uist. Uist dialect(s) seem to show a preference to use the preposed adjective seann-over the plain adjective aosta. Most importantly, however, seann té happened to serve as a reference to an inanimate feminine noun in only one example; although most examples of seann tè meaning cailleach 'old woman/female', come from South Uist (the one from Lewis is encountered in an autobiography), whereas the only example of té aosd is from Lewis (where the plain adjective is more commonly used).
Similarly, in the case of feadhainn (18 with seann-, five with aosta), most tokens mean 'people'. However, there are some among those with seann-, which only function as a back-reference to something (like taigheandubha 'black-houses', brògan 'shoes'), i.e., it represents a rather pronominal sense. On the other hand, seann-very often occurs in general statements 52 Veronika Csonka ('the old ones/old people'). These statements mostly refer to old customs or lifestyle, which represents a very similar aspect to compounds like seann òran 'folksong', seann sgeulachd 'traditional story', sean-fhacal 'proverb', etc. (being associated with traditions), or are related to old times (the 'old ones' may have been young then; cf. sean shaighdear 'veteran' below). Interestingly neither do the examples with aosta show a pronominal sense, all referring to people. However, they appear to have a more qualifying function (as opposed to its more lexicalised usage in 'old ones'), or may refer to a particular situation, rather than a general statement. Feadhainn aosd(a), is more of an adjectival phrase (where the quality of age is more important and highlighted). Again, there is only a coordinative example with the plain adjective aosta from South Uist, whereas the rest are from Lewis. Seann-is more evenly distributed among the sources. A good example for the usage of aosta here is from Lewis: feadhainn aosd-aosd (pl) and fear aosd-aosd (sg) occur in the same dialogue. Both feadhainn and fear refer to people; however, their old age is even more emphasised by the repetition of the adjective. (Similarly, fàileadh dorch aost 'an old dark smell': highlights the aging quality of the smell (probably very uncomfortable as if food had been left somewhere for a very long time).)

Native speakers' judgements
I intended to test the conceptualising function of preposed adjectives with unusual phrases (such as ʻgood feather' or ʻold happiness'). However, the use of adjectives deagh-and math did not show any difference between tangible and conceptual nouns. One possibility is that it might have worked as one of the factors in the past that determine the present distribution of deagh-and math (consider words like 'good reason' and 'good intention', the Gaelic for which were deagh aobhar (occasionally reusan) and deagh rùn for most speakers), but synchronically it has less influence on the word choice. According to my results, deagh-seems to substitute math in the south (probably losing the distinction), and spread in the north as well (giving more opportunity for variation). 4 The eldest informant from South Uist (7U) still uses math with many words: consider each math 'good horse', ite mhath 'good feather', gloinne mhath 'good glass', as well as the loan words spagetti math 'good spagetti', yoyo math 'good yoyo'. What is remarkable here is that all of these words are tangible -referring to either an object or an animal -as opposed to expressions like deagh 54 Veronika Csonka or animal (three informants from Lewis (one of them was 12L) translated 'old horse' as each aosta (it might have been influenced by the picture, which shows a particularly old horse)). 6

Figure 1
Some of my informants managed to give me subtle differences in meaning between phrases with preposed adjectives and those with their plain counterparts. 1L translated deagh obair as 'a job well done', while obair mhath as 'a good kind of work'. 2U would say deagh obair when 'you're doing a good job' or 'a job is good', but obair mhath only in the second meaning. This means that deagh obair refers to a more abstract concept for these speakers, although in a somewhat different sense.
A similar distinction can be observed in 7U's word choice for 'good memory' (if there is any distinction at all). 7U translated 'good memory' as deagh chuimhne; however, could not think of a plural for this word at first: probably considering 'memory' as a mental skill. Trying to say 'memory' as a countable noun (i.e., a picture in your mind) the informant said Bha cuimhne mhath aca… 'They had a good memory' Bha deagh chuimhneachan aca. 'They had good memories.' If my interpretation is right, this speaker prefers deagh chuimhne for 'good memory' as a mental skill and cuimhne mhath for 'memory' as a a picture in your mind (which is a more tangible, countable noun), but again deagh chuimhneachan in the plural (note that the first meaning does not have a plural). Although consider that this informant does not usually use the latter variant (cuimhne mhath) of this expression.
The informants gave me the following translations for droch rud: 'a bad thing/happening'; 'a thing or sg you're talking about, a news is bad'. Only 3L differentiated between two meanings, although with regard to the hyphenation even this informant was uncertain. According to the pronunciation (and to the abstractness of the meanings), he suggested "droch rud 'a bad thing' (literal) and "droch-rud 'full of badness' (more abstract). Answering my question the speaker added that the latter could refer to 'devil', but only in an abstract sense.
Speakers 1L and 4U helped differentiating between droch bhean 'a bad wife' (failing at what a wife traditionally means) and bean d(h)ona: 'a wife that is a bad person as well'. Here the plain adjective dona may convey criticism: droch-refers to the semantics (meaning) of bean 'wife', while dona evaluates the person herself. (6L, 7U and 10H would not use dona with bean at all.) Droch dhaoine refers to 'bad people' in general, reflecting on evilness. I asked only half of my informants (5) about the pictures which showed villains and criminals, but all of these speakers found that the phrase droch dhaoine matched the pictures: We have already seen that the plain adjective aosta (and sean) normally refers to the age of a person, animal, or -sometimes -object (e.g., aodach aosta 'old clothes'). This tendency is confirmed by the distinction between seann-taigh and taigh sean for 4U: in the sentence Tha taigh sean agam. 'I've got an old house.', taigh sean refers to a house in which 'everything is old', whereas seann-taigh denotes a previous house (e.g., the family's old home). I gave two sentences containing 'old friend' to the informants, separated with other sentences: one with reference to 'a long-existing friendship' (I've got an old friend from primary school.) and one referring to age (I've got young friends and old friends.). In the translation for the first sentence every informant (apart from 12L) used the preposed adjective sean(n)-. In the other sentence, in which 'old' was in coordination with the opposite adjective 'young', four informants chose a plain adjective: caraidean sean (1L, 7U) or caraidean aost(a) (10H, 11L) (five with 12L). Nevertheless, I doubt it was influenced by the presence of the other plain adjective òg (as we usually see the opposite in coordinations with math and droch-(see (20) in section 3.2), which means that this word choice marks the age reference. Two informants from South Uist (2U, 4U) and 6L applied stress to make the distinction more obvious: "seann-charaidean ∼ "seann-"charaidean (but seann-"charaid in the first sentence).
Two informants commented on the distinction between aosta and sean. 10H (Harris) felt that aosta, when used in relation to people, is more polite and milder than sean. For 2U (South Uist) aosta is stronger than sean, sean meaning 'old' and aosta 'really old'. At first sight these two interpretations seem rather contradictory. However, 2U also adds that aosta refers to the older generation, which may eventually mean that aosta entitles respect, thus it may be felt more appropriate in connection with people. (This also may be the cause for the decreased use of sean these days.) In my opinion, this lofty connotation may explain its use in more literary expressions, and perhaps also with abstract concepts such as brònach(d) aosta 'old sadness' (10H), déjà vu aosta 'old déjà vu' and toileachas aosta 'old happiness' (6L) (although 6L does not always distinguish between preposed and plain adjectives).
Regarding the adjective ʻold', it is worth noting that the meaning [being around for a long time] and actual age are very close meanings, which may account for the high productivity of seann-. It is difficult to differentiate for example between a more tangible, physical age of a building ('it's old so it's falling apart') or the concept of having existed for a long while. Seann-being the stereotypical adjective for the sense 'old', phrases with aosta are idiosyncratic, unique in a sense (cf. seann eòlas vs. eòlas aosta for 'old knowledge'). Their meanings can be illustrated on a scale: animate: aosta/sean inanimate (tangible -objects) abstract (concepts): seann-

Figure 3
These meanings tend to merge (they are not clearly distinctive), which leaves seann-accepted in all meanings.

Preposed adjectives in other Celtic languages
There is a great number of preposed adjectives in Brittonic languages (i.e., in Welsh, Breton and Cornish The same phenomenon can be observed in Romance languages: cf. Spanish niño pequeño 'little boy' vs. pequeño niño 'baby boy/son', French fille petite 'small girl' (i.e., a girl who is small) vs. petite fille 'little/young girl' and petite-fille 'granddaughter', where the normal N+A order indicates a simple grammatical phrase, whereas the idiosyncratic A+N order refers to a more abstract construction, which may be considered a compound.
Regarding Brittonic examples, A+N constructions are very often loose. However, there is a considerable number of examples for strict compounds as well: 7 W. "gau "broffwyd 'false prophet', "prif "ddinas 'chief town' vs. "prifford 'highway', "hen ddyn 'old man' (Morris-Jones 1921, 80-84;Thorne 1993, 840-844). 7 In proper compounds the first element qualifies the second giving a specifier-generic constituent order. Proper compounds can be further divided into strict and loose compounds according to their stress patterns: strict compounds show regular word stress indicating their word status, loose compounds have stress on both elements, or secondary stress on the initial element (the specifier) (Morris-Jones 1921, 18-20, 80-84;Thorne 1993, 840-844;Hardie 1948, 55-56) 58 Veronika Csonka In Irish, the other major Goidelic language among Celtic languages, the distinction between the preposed adjectives dea-, droch-, sean-and the plain adjectives maith, dona, aosta is rather similar to the observations in Scottish Gaelic, although proper research has not been carried out in this language yet. Although all phrases may not follow the same patterns as in Scottish Gaelic, certain tendencies show obvious similarity. These include preposed adjectives relating to abstract or figurative/more integrated meanings; tangible words and enumerators 8 more commonly qualified by maith; the rarity of phrases with dona in general; and the high number of both adjectives seann-and aosta with nouns denotating people. Consider also dea-bhéasa ʻ(good manners;) etiquette' vs. béasa maithe ʻgood manners' and dea-scéal ʻgood news' vs. scéal maith ʻa good story'. Most abstract nouns qualified by droch-do not occur with dona (neither olc ʻbad, evil') in the corpus. Further figurative examples include: drochfhéachaint ʻevil look' (ScG droch-shùil), droch-chroí ʻill will' (lit. 'bad heart'), drochobair ʻmischief' (lit. 'bad work'). Among the nouns frequent with droch-, scéal and rud are also found relatively frequently with dona (scéal dona ʻbad news, bad state of affairs', rud dona ʻa bad thing'). Duine ʻman' occurs in quite a high number with both adjectives meaning ʻbad', although higher with preposed adjective, which may indicate a kind of compound usage, just as in One apparent difference between Irish and Scottish Gaelic preposed adjectives is that fixed structures in Irish are more common with maith, even though they refer to abstract, mental concepts: aithne mhaith/eolas maith/fios maith + ag + ar ʻto know someone/something well' (lit. 'to have good knowledge of someone/something'), cuimhne mhaith + ag + ar ʻto remember something well' (lit. 'to have good memory/remembrance of something'), which is not always the case in Scottish Gaelic. Furthermore, in Irish the preposed adjective corr-ʻodd, occasional' also has a postnominal counterpart: cf. "corr"dhuine ∼ "corrdhuine 'oddball' vs. duine corr 'an odd person', which does not apply to Scottish Gaelic.

Preposed adjectives in other languages
In this section certain suggestions emerging from the study on Scottish Gaelic adjectives are discussed in a wider context. Firstly, the animate/inanimate contrast is a part of universal semantics in the world's languages, so it is not surprising to exist in relation to the adjectives referring to age. Secondly, in section 6.2 conceptualisation and other distributions of meaning are revised regarding adjectives in a couple of other languages.

Opposites (quality, age)
First of all, we find the opposites of quality ('good' vs. 'bad') and age ('new/young' vs. 'old') in every language, these concepts representing a part of human cognition. In Gaelic we have even seen the role of contrast in the different frequency patterns for the use of deagh-and math as opposed to droch-and dona. To recall the discussion on this, I here repeat sentences from (20) What is not universal in the world's languages, is the distinction between animate and inanimate usage of both 'new/young' and 'old', although the concept is present in most languages in some way or other: it often occurs in at least one of them, just like in English, which shows a distribution for the meanings 'young' (animate) and 'new' (inanimate). Examples from other languages are referred to in Table 4. Even in English the word elder can only be used for people and not for objects or concepts. In light of this it is worth taking a look at the other member of this opposition.

Animate/inanimate symmetry (age)
In the case of Scottish Gaelic three variations of 'new' can be encountered. These are as follows: ùr meaning 'new' or 'fresh', òg usually in the sense 'young', and nuadh, which may be a dialectal variant for ùr, in the inanimate sense 'new' (the Irish word for ʻnew' is nua or úr), but it also occurs in 62 Veronika Csonka innovative compounds with the meaning 'modern' (e.g., nuadh-eachdraidh for 'modern history'). Considering these varieties, it is obvious that Gaelic does show the animate -inanimate distinction for both 'old' and 'new', the only difference being that there is no distinction of order in the case of 'new/young' (although note the A+N compounds of nuadh-), whereas the various meanings of 'old' may also be distinguished by word order. Interestingly enough, however, there are proper compounds to be encountered with òg as the specifier, referring to an early time, rather than to the age of a person or other living creature: óg-mhadainn 'early morning, dawning', Ógmhios 'June' (lit. 'early/young month'). The opposition of 'young'/'little' and 'old'/'big' has also proved to be a useful device in languages to express relationships: they form part of conventional compounds like grandma (German Grossmutter, Hungarian nagymama, etc., all of which literally come from 'big mother' (cf. Irish mo mháthair mhór 'my big mother', i.e., 'grandmother'), similarly shima soni in Navajo and seanmhair in Gaelic (literally meaning 'old(er) mother'). In some languages we even encounter parallel expressions with 'little', as in Navajo shima yazhi 'little mother' for 'aunt' (the -now derogative -Hungarian kisanyám 'my little mother' might originate in the same -or similar -meaning). Chinese xiăo 'little, small, young' + SURNAME and lăo 'old, experienced' + SURNAME have special meanings in addressing each other, and show hierarchy at a workplace.

Meaning varieties and meaning change
According to the observations in the previous sections, preposed adjectives show a certain conceptualising function, for which we have seen various, colourful examples, often triggering change in meaning. In the case of droch latha and latha dona 'bad day' the difference is grammatical: the first being more subjunctive (with the future or conditional tenses), the latter more factual and certain (in present or past). Deagh obair has proved to be associated with job in a more abstract sense ('good job'), while obair mhath referred to an individual work ('good work'). Other examples for meaning change included astar math for 'good distance' and deagh astar for 'good speed', àm math for 'appropriate time' and ann an deagh àm meaning 'in time', galar dona for a 'bad illness' and droch ghalair for 'great suffering' (i.e., 'bad illness' in an abstract sense). The difference is even more salient when the more abstract phrase becomes a compound with a figurative sense such as droch-àite for 'hell' as opposed to àite dona referring to a dirty, unsuitable place. Similar compounds are an droch-shùil 'the evil 64 Veronika Csonka is jiù. However, we can express the sentence 'This book is old.' in two different senses: if we are referring to the paper the book was printed on, we can use the word jiù without any problems. On the other hand, if we mean to refer to the content of the book (an abstract concept), we have to say lăo instead of jiù.

Conclusion
The feature of abstraction appears to lie in the vocabulary -it forms part of the semantics of a word. This is a "historic" aspect: some words are inherently used with deagh-, others with math (e.g., deagh aobhar 'good reason', deagh rùn 'good intention', deagh smuaintean 'good thoughts', etc. vs. each math 'good horse', ite mhath 'good feather', gloinne mhath 'good glass', etc.). With regard to abstract vs. tangible phrases, preposed adjectives tend to relate to more complex semantic structures, leaving the more obvious, more simple meanings to the plain adjectives. Therefore, plain adjectives tend to refer to the most tangible entities (people, animals, objects) and pronominal expressions. For instance, deagh-stands with professions or with a person with the appropriate qualities for a job, etc. (deagh dhuine as 'the right man' and also maybe deagh oidhche iasgaich 'the right night for fishing'), whereas math qualifies vehicles and animals. Similarly, drochrefers to the complex semantics of droch dhelicatessen 'bad delicatessen' (it may refer to the products, staff, etc., i.e., all parts of the whole institute -not only to the condition of the building itself), droch àite 'bad place' likewise (consider its occurrence in contexts with clear references to 'hell'), whereas dona can describe the qualities of food. Rud dona (which is encountered in more recent sources of the corpus) expresses a pronominal sense similar to rudeigin dona 'something bad', whereas droch rud 'a bad thing' is probably meant more as a concept, let alone the abstract reference to someone's character or behaviour as droch-rud 'full of badness'. Spiorad 'spirit' is the only religious word in the corpus the tokens of which are each accompanied by the preposed adjective droch-, as opposed to references to people or anthropomorphic entities all qualified by dona.
Seann-also carries more abstract meanings compared to more pronominal aosta (cuid 'some', dithis 'a couple (of people)'; feadhainn 'ones', tè 'one' (fem.)), although the latter two, mostly being references to people, also occur with seann-in the corpus (in sources from South Uist). Seann-, just as the other preposed adjectives, may easily convey more integrated meanings, such as seann bhean ʻold woman', seann-duine ʻold man', seann bhoireannach ʻold woman', etc. (also consider seann tè, which is similar in meaning to cailleach ʻold woman, hag' vs. tè aosd ʻold female') or traditionality (see seann-taigh 'a (traditional) black-house' vs. taigh aosta/sean yet. Regarding more universal aspects, the animate/inanimate distinction between adjectives of age can be found in various languages (such as English young/new and Hungarian öreg/régi). The adjectives of age serve as a useful device to indicate relationships in the world's languages (cf. Navajo shima soni 'big mother' as ʻgrandmother' vs. shima yazhi 'little mother' as ʻaunt').
As regards the relationship between word order and abstraction concerning age, certain similarities and differences can be observed between the Scottish Gaelic and Spanish adjectives meaning ʻold': (1) former roles are indicated by a preposed adjective in both languages (un antiguo colega ʻa former colleague' -seann-leannan ʻold sweetheart'); (2) old institutes/families etc. ('previous' or 'having existed for a long time') can be expressed by using a plain adjective (un pueblo antiguo ʻan ancient people' -'na shloinn seanchaidh aosta ʻin his old storyteller family'), although not necessarily (see ScG Sean-Ghàidhlig ʻOld Gaelic/Irish' and Irish Sean-Ghall ʻAnglo-Norman, Old English'); (3) when qualifying certain words, the preposed adjective may refer to the length of the relationship (una vieja amiga -seann charaid ʻan old friend'), whereas the plain adjective to the age of a person (una mujer vieja ʻan old woman'-duine aosta ʻan old man'; see also partitative cnàmhan aosta ʻold bones'). On the other hand, seann-is a common adjective concerning people in every dialect of Gaelic (as well as in Irish), which means that seann-duine ʻold man' is at least as frequently used as duine aosta, if not more. Although all adjectival phrases in Mandarin are preposed A+N, the two Mandarin adjectives for ʻold', lăo and jiù, besides the animate/inanimate opposition, show an abstract/tangible distinction as well, similar to the distinction between seann-taigh ʻprevious house' and taigh sean/aosta ʻan old house': the adjective for animate nouns is used to express abstract concepts, while inanimate jiù refers to the physical age of an object.