Ignoring the limits of the instrument

Our work to re-discover and/or reimagine the music and playing techniques of an extinct oral tradition has a number of ingredients, and one of these is the physical instrument itself. I put a lot of value on the study of the museum harps, which we know were the actual instruments owned and played by tradition-bearers; for some of them we have the physical instrument, preserved since the death of the performer; we have a portrait of them sitting with the same instrument; and we have live transcriptions of their playing.

The instrument can sometimes be thought of as a kind of constraint on what you can and cannot do. However I am more and more aware of the idea that a performance tradition can ignore the constraints and especially the possibilities of an instrument. I think I first was explicitly aware of this from the Finnish jouhikko tradition, where Juho Villanen (1846-1927) played a three-string jouhikko for the collector A. O. Väisänen in 1912. Villanen had made the instrument himself many years ago. He tuned the three strings to g d’ a’. “Villanen did not use the G string at all, but played his pieces on a single [a’] string while the other [d’] was a drone” (Nieminen, Jouhikko, 2007, p.41).

I have posted here before about the Irish pipes, and how the tradition does not take full opportunity of the possibilities offered by the instrument. You could even say the same about traditional fiddle styles, where the player usually stays in first position and does not use the full length of the bow. I start to wonder if it is a classical idea that the instrument has to be used to its full potential, pushed to its limit, and I wonder if this is connected to the idea of “improving” the instrument to extend those limits. I also think there is a connection in Ireland in the 19th century, where the old Irish harp with wire strings was pushed aside and eventually supplanted by the Anglo-colonial-classical gut-strung pedal harp and lever harp.

Ellis Roberts (1819-1873) wrote a Manual or method of instruction for playing the Welsh harp originally written in 1858 but published in 1902. This seems to be a very interesting insider’s testimony from the Welsh tradition. He describes the tuning of the telyn deires (triple harp) on pages 3-4:

For some unexplained cause the Triple Harp is tuned in the Key of G…

He gives a cycle of 5ths and 4ths to set the diatonic notes of the outer rows including F♯ in the outer row. He starts to discuss the inner row of strings, starting with C♯ on the inner row “situated between C♮ and D♮” but then he gets side-tracked and says that

the C being raised half a note enables the performer to play in the key of D. Of course every C through the scale must be raised a semitone

– interesting to see that the outer rows of the harp are re-tuned for different keys, using the inner row only for “accidentals”.

Then he gives a cycle of 5ths for tuning the inner row of sharps and flats, and says “you will find the Harp scale Diatonic and Chromatic quite correct according to scientific principles”

On p.10 he writes “General Observations”. Again he says:

It being the custom (as before alluded to in Chapter I on Tuning) to tune the Welsh Harp in the key of G (though for what reason I have not been able to ascertain), the first Exercise will begin in that key. The more natural key would be C to open with.

He has what in Scotland might be referred to as “the cringe”, referring to his traditional practice as “inconsistent with modern usage” and quoting Boscha as an “eminent author” on harp technique. This is the same measuring ones own traditions by Anglo-classical norms, as I have seen in other Scottish music traditions in the early 20th century.

He continues:

The Chromatic Scale on the Welsh Harp is perfectly complete, though its notes are rarely used by Welsh Harpers, for the following reasons – First; Flats and Sharps are seldom used in Welsh Music except as accidentals. Secondly, being in the Middle they are difficult to play without causing a disagreeable jar upon the outside strings. And lastly, the art of Fingering the “Middle Strings” has been hitherto unknown to us. I never heard a chromatic passage played by any Welsh harper, though the whole Chromatic system is as perfect upon the Triple Harp as it is on the Piano-forte, but of course more difficult to execute. I see no reason why the Music of Corelli, Handel, Haydn, and Mozart should not be performed upon the Welsh harp.

It’s interesting that Ellis Roberts thought he was being radical and new suggesting Handel could be played on the Welsh triple harp; 100 years previously John Parry was in London playing Handel on a Welsh triple harp.

But more interesting still is that he learned in a tradition, and was strongly enough invested in that tradition, which used and required a fully chromatic triple harp but which made minimal use of the inner chromatic row of strings.

Does this parallel our testimony of Carolan playing his tunes with “no bass”, even though his harp had a full 2 octaves below na comhluighe g? Did Carolan have such a big harp with all those strings tuned to their allocated pitches because that was his tradition, even though he had no need or inclination to use them, just like Villanen had the 3rd string on his jouhikko?

Róis bheag dubh

I made a demonstration video of Róis bheag dubh, based on Edward Bunting’s live transcription from a tradition-bearer in the 1790s:

The transcription is at Queen’s University, Belfast, Special Collections MS4.29 p.62/58/067/f28v. This page from one of Bunting’s little pocket collecting pamphlets, is headed “Rosey Black or Rosh veg Dooy” and “very old”. The transcription seems nice and clear with notes and dots but no barlines. Underneath the transcription Bunting has written […m] McCracken Belfast” – I don’t know the significance of this. Then on the lower half of the same page, Bunting has made a neat edited version of the tune.

The transcription shows the tune written at pitch, either in E minor or A neutral. There is no F in the tune; there is an intrusive C at a few points (the notes of both E minor and A neutral pentatonic mode are E G A B D). Bunting has written a few + or tr marks, and a couple of wee grace-notes in the transcription. The strong notes and pulse of the transcription is ambiguous in places due to the lack of barlines, but we can perhaps use the neat copy to help us understand this, as long as we remember that the neat copy is Edward Bunting’s edition and has been changed in a number of ways from the transcription.

Three different versions

There are many different notations and piano arrangements of Róis or Róisín dubh in Edward Bunting’s notebooks and publications, but we can group them together and see that the various copies derive from just three different performances, and so we can understand that Bunting has in all his many copies, three versions of the tune.

Two of the versions were printed as piano arrangements in 1840: No. 18, “Black Rose Bud” and no.19 “Second Set of Black Rose Bud”.

The hairdresser’s version

The first version is tune no. 18, on p.16 of the 1840 book, titled “Black rose bud”. In the index p.vii Bunting says it was collected from D. Black, Harper, in 1796. But I am questioning this attribution, because the melody of this piano arrangement is clearly derived from a transcription in Queen’s University Belfast, Special Collections, MS4.33.1 p.42. This transcription was made by Edward Bunting from a traditional singer in or around Westport, between Tue 6th and Thur 22nd July, 1802. Bunting had gone out to Co. Mayo to meet with Patrick Lynch, who had been collecting song texts. It seems that Lynch took Bunting to meet his informants so that Bunting could notate the tunes from them.

In Patrick Lynch’s journal QUB SC MS4.27 p.47 he lists the words of “Roisin Dubh” as being collected from P. Lynch Hairdresser in Castlebar, May 26th 1802. Was this P. Lynch Hairdresser also the source for Bunting’s transcription of the melody on MS4.33.1 p.42?

This index entry in Patrick Lynch’s Journal refers to the text of four quatrains written out neatly in Lynch’s song book, QUB SC MS4.7.091, no.58, titled “Roisin Dubh” and beginning “A Roisin na biodh bron ort far eighrigh dhuit”. Colette Moloney (Introduction and Catalogue p.209) says that these words are tagged “Castle Barr” at the top of the page, but I don’t see that on the online facsimile (the image may be cropped).

On the next two pages are 10 more quatrains, tagged at the top of the page “Drogheda”, numbered “59” and titled “Rois Bheag Dubh” and beginning “Beidh eiclips ar na speirigh is doirtear fuil”. This must be a second independent version of the song, inserted by Lynch for comparison. This manuscript is not a transcription notebook, I think it is a neat copy made by Lynch for presentation purposes. There is a second copy of the Castlebar words in QUB SC MS 4.11.

At the moment, I have not done any serious collation of the song airs transcribed by Bunting in QUB SC MS4.33.1 in early summer 1802, and the song words collected by Patrick Lynch on the same trip, presumably from the same informants – that’s a project for someone else. So, I would not say that for sure the lyric beginning “A Roisin na biodh bron ort far eighrigh dhuit” goes with Bunting’s first version of “Black rose bud”, and that the tune and words both come from the singing of P. Lynch, hairdresser at Castlebar, but it looks pretty likely.

The doctor’s version

The second version of our tune was printed on the very next page of the 1840 book (no.19, p.17), with the title “Second set of black rose bud”. In the 1840 index p.vii, Bunting tells us that this second version came from a peasant in Cushendall in 1804; in the introduction p.97 he says it was from the Lower Glens, co Antrim; in QUB SC MS4.27 he says it was from the Low Glens in 1803; and in QUB SC MS4.12 he labels it “Dr. McDonald’s set”. The tune appears in QUB SC MS4.29 p.34/34/043/f16v where it is titled “Doctors set bad good not very”. This notation in MS4.29 is clearly not a transcription but is copied from another written version; the other two tunes on this page are copied from a printed book. My guess is that Dr. James MacDonnell, originally from Cushendall in the Lower Glens, had notated it from a traditional singer in 1803/4, and had sent Bunting a manuscript copy of the tune, which Bunting subsequently copied into a blank space in his pocket notebook.

Bunting gives us more information about this version of the tune, in the 1840 introduction, p.97:

(No.19 in the collection) Roisin Bheag Dubh “Little Black Rose-bud” – Differs only slightly from the preceeding. It is here set according to the version preserved in the lower Glens of the county of Antrim. The cadence at the termination seems to lean so much more to E than A, that the Editor has adopted the former key-note as its tonic. This curious anomaly is frequently observed in these simple airs.

Black’s version?

Donald or Daniel Black is an interesting harper and we have some snippets of information about him. In the context of this tune, Bunting describes him singing with harp accompaniment – a rare eyewitness account of a harper singing to their own accompaniment.

(No. 18 in the Collection) Roisin Dubh. “Black Rose-bud” … It was sung for the Editor in 1792, by Daniel Black, the harper, who played chords in the Arpeggio style with excellent effect. The key-note at the end of the strain, accompanied by the fifth and eighth, without the third, has a wailing, melancholy expression, which imparts a very peculiar effect to the melody.

1840 introduction p.97

Bunting has put descending arpeggios in his 1840 piano arrangement of the Castlebar hairdresser’s vocal version of the tune, but I don’t think this piano bass is anything more than Bunting’s newly composed harmony, inspired by his memory of Black’s playing over 40 years previously.

So we have three conflicting pieces of information. we find the transcription source of Bunting’s first version of the tune in the Westport 1802 notebook; Bunting says that the tune was collected from Black in 1796; and he says that Black sung the song to him with harp accompaniment in 1792.

Perhaps all three are true. Perhaps Bunting prepared his 1840 first version from the Westport 1802 notebook. Perhaps he did collect a version from Black in 1796. And perhaps he did hear Black singing the song with harp accompaniment in 1792.

In which case, is the QUB SC MS4.29 p.62 transcription from Black in 1796? Bunting tells us (1840 intro p.76) that Black’s “chief resort, when in Antrim, was Mr. Heyland’s seat at Glendaragh, near Antrim, where the Editor saw him shortly before his death, in 1796. He sung to the harp very sweetly”. (see my Bunting Collecting Trips post)

It is also possible that some of this information printed in 1840 is wrong, and that we are being led wildly astray by Bunting’s disorganised waywardness… Bunting was telling us about the events of 44 years ago; we can see looking at his manuscripts that his notes are scattered, partial, and full of context-less notes.

Bunting did jot down a tune-list, in QUB SC MS4.29 p.178:

{Peggen a Leaven Daniel Black
+{Brough ne Shannon ditto——–
{Garran Buoy yellow horse Ditto —-
{Collin Fin ——— Ditto ——–
{Black bird & thrush Ditto ——–
{Little hour before Day Ditto ——
{Castle Moon Ditto ——-
{Huar ma fian Ditto ———-

We can see from my Old Irish Harp Transcription Project Tune List spreadsheet, that a lot of these tunes appear in QUB SC MS4.29 between pages 93 and 106. We also have a tune, tagged Black, on p.64 (possibly related to the p.62 Róis bheag dubh). It is not clear to me at the moment how the different gatherings of ms4.29 have been shuffled and re-assembled before their first binding (see my pdf), and whether they may originally have been related.

If the p.62 transcription was written from Black’s performance at Glendaragh House in the summer of ’96, was he singing it? There is no indication in the transcription of “chords in the Arpeggio style” which made such an impression on Bunting. And the structure of the melody as transcribed seems very harp-like. Perhaps Bunting asked him to play the tune on the harp so that he could more easily make a transcription. We really know so little about the details of this process.

Caitlín Triall, and the Friar and Nun

If we check my Old Irish Harp Transcription Project tune list spreadsheet then we can see that the section of Queen’s University Belfast, Special Collections, MS4.29 from p. 44 to p.61 seems to be mostly transcriptions notated from Denis O’Hampsey in Magilligan.

However there are two transcriptions in this section which seem to me to be written in a different style, and which I think might be noted from Hugh Higgins in 1792.

Caitlín Triall

The first of these is Kitty Tyrrell. Bunting wrote a live transcription on Queen’s University Belfast, Special Collections, MS4.29 p.47/43/52/f21r. Bunting has titled his transcription “Cathleen Treeall”. It is a lovely traditional air with many variants, still well-known today in the living tradition. Here is my tentative demonstration:

Caitlín Triall is one of the tunes tagged “Higgins” in the annotated copy of Bunting’s published piano arrangement (1797 no.8). I’m not convinced that his arrangement is really based on the transcription; it might derive from an earlier printed edition.

At first sight , the p.47 notation looks very clear and simple. But, because Bunting does not indicate bar-lines, this transcription does not show what the stressed or strong notes are, which is vital for reconstructing the flow and structure of the tune. Some parts are clear enough, and to help me make sense of it, I have listened to other traditional versions, including Teresa McCormack’s 1930 harp recording, where the tune is titled “Eamonn na Cnuic”, and Máire Ní Arbhasaigh’s sung fragment from 1931.

Bunting has noted “B” at certain points in the tune, which is usually interpreted as an indication of a note played an octave lower, in the bass. I have also chosen to play the low D-E passages in the bass, though that is not indicated in the transcription. This video is perhaps less a “demonstration” of the transcription, more an “interpretation” of it.

The Friar and Nun

Turning over to the next page, QUB SC MS4.29 p.48 shows another tune in the same transcription style, without bar-lines. Here is my demonstration of it:

As well as this transcription, Bunting gives us a fair amount of commentary on this tune. I don’t have copies of the manuscript pages, but they are cited by Donal O’Sullivan in his Bunting (1983, p.199), and Colette Moloney in her Introduction and Catalogue (2000, p.229). There is a little English song text:

A Lovely Nun to a Friar Came
to Confess on a morning early
In what, my dear, are you to blame
come tell me most sincerely
Alas, my guilt I dare not name
but my lad he loved me dearly

There are also three different descriptions by Bunting, all within a few pages in QUB SC MS4.12 which kind of repeat the same information but with different emphasis or detail. Two of the descriptions say

This air is given on Hempson’s authority as being an Irish tune; he had learned it from his first master…

but the third does not mention O’Hampsey. We are told that

Many other harpers had it and played it with the same variation of the Octave to imitate the soft effeminate acute voice of the young lass and the masculine deep voice of the friar at confession

In the video, I am playing the sections marked “B 8ths” in the transcriptions using bass octaves. These sections in the harp transcription don’t match the male and female voices in Bunting’s song text. There is no mention of the harpers singing any words to it.

Bunting also tells us that

This was a favourite tune with the old harpers, and listened to by every auditor with great delight

Bunting published a piano arrangement in 1840 (no. 142), and in the introduction p.100 he writes:

This is the only air admitted into the collection which is not of unquestionable Irish origin; but the Editor has adopted it as Irish, on the authority of all the old Irish harpers with whom he has conversed; it was at all times a favourite tune of theirs… the higher and lower octaves aptly coincide with the alternations of the male and female voices in the song

Bunting’s piano arrangement is very enthusiastically classical piano style; it doesn’t follow the transcription very closely. He includes a division-type variation, and at the top of the page he says “The var[itation] by Lyons in 1698”, but I have no idea where this information comes from and I don’t see it in the transcription manuscripts.

There is some interesting sociological stuff going on here; compared to the beautiful and sophisticated Irish song airs played by many of the harpers, this is a sniggering schoolboy ditty of a song. How are we to understand the harpers and their listeners taking “great delight” in this “favourite”?


All but two of the tunes in QUB SC MS4.29 p.44 to p.61 are tagged Denis O’Hampsey in later piano arrangements. The two exceptions are Kitty Tyrrell on p.47, and A ghadaidhe ghoid mo shláinte Uaim (The Jointure or the Golden Star) on p.53, which are both tagged “Higgins” in the annotated 1797 print. The Jointure is not a transcription, it is a neat edited copy, but we could be missing the facing page (indicated in my ms29 text pdf) which may have contained the transcription which this neat edited copy is based on.

I suspect that the Friar and Nun was also transcribed from Higgins in 1792; the notation style of both it and Caitlín Triall on p.47 of the manuscript looks different from the O’Hampsey transcriptions, but reminds me of Tá mé mo chodhladh on p.28.

Did Bunting carry these Higgins ’92 transcription sheets with him when he visited O’Hampsey in Magilligan in ’96? Did he scribble O’Hampsey’s tunes into spare spaces in the ’92 collecting booklets? Did he ask O’Hampsey’s opinion of the tunes he had collected from Higgins?

Thoughts on the Old Irish Harp Transcriptions Project

It struck me that these two raise questions about what my current work is; the manuscript transcriptions can offer different and interesting sets of tunes from what we get from classical arrangements, or from present day living tradition. It depends really what you want to do. I understand more and more that for me, my motivation is to understand how to play the old Irish harp, what is its natural style and idiom, how does a tune become adapted by the harper to sit easily and freely on the instrument.

This Old Irish Harp Transcriptions Project developed out of the realisation that the transcriptions “froze” the performing style of the last of the old harpers, and so a detailed study of those transcriptions could reveal subtle aspects of old Irish harp style and idiom that would not be available from any other source.

However, it quickly became clear that the study needed to firstly discriminate, so as to filter out all of the manuscript notations that are not transcriptions at all, but were copied from printed books and other written sources, or are new piano arrangements; then to filter out all the transcriptions that are not from harpers, but are from singers or other instrumentalists; and now the third level of filtering, to recognise transcriptions that are genuinely notated live from the playing of an old Irish harp tradition-bearer, but which are deficient in some way so as to not give us what we need to know, about how the tune was actually played.

A chailíní, an bhfaca sibh Seoirse?

I made a demonstration video of A chailíní, an bhfaca sibh Seoirse, played (with some editorial adjustments) from Edward Bunting’s live transcription from old Irish harp performance in the 1790s.

This interesting tune was published twice by Edward Bunting; he had transcribed it twice, from two different harpers on two different occasions.

The published piano arrangements are no.6 in his 1797 book and no.11 in his 1840 book. Both of these piano arrangements are derived from the same live transcription, made probably in the summer of 1796 from the harper Denis O’Hampsey in Magilligan, County Derry. The transcription of the instrumental tune with variations is spread across pages 46, 48, 49 and 50 of Queen’s University, Belfast, Special Collections, MS4.29, in a section of the manuscript (p.44 to about p.61) which is mostly transcriptions from O’Hampsey in 1796.

However, that is not what I am playing. I am working from the other, independent transcription of the tune which Bunting made, on QUB SC MS4.29 p.32/32/41/f15v. This transcription which I am working from is in the second, “difficult” section of the manuscript, which mostly seems to have been written in the summer of 1792 (see my tune list spreadsheet). Previous tune transcriptions on p.26 and 28 are tagged as being from Hugh Higgins in 1792; I am wondering if the transcriptions on p.30 and this one on p.32 are also from Higgins in ’92.

Bunting has written other information around the notation of our tune on p.32. At the top of the page in a cartouche, he writes a note to himself: “get Molly bawn the first tune learned on the harp”. The only transcription we have of Maíli Bhán is from Patrick Quin, probably between 1802-1806, in QUB SC ms33.1 p.62/71/f31v. Bunting also writes two other tune titles, “Rosh veg Doo or Rosey Black” at the top of the page, and “Hugar M’ Fean” at the bottom. These might not be relevant or contemporary notes. Both these tunes were later transcribed in QUB SC MS4.29. On p.86, in the other main group of Higgins ’92 transcriptions, we find a transcription of Thugamar féin an samhradh linn; but the copies of Róisín Dubh in the manuscript appear to date from 1796 (p.62, maybe from Black) or from a manuscript copy of 1803-4. (p.34)

Then Bunting writes information about our tune. He writes: “Shalena <Callena> vacca sheo shorse or / Chooheeir a Chur heava / McCabe’s verse’s on Carolan to the same tune”. This information relates to different titles and/or song texts connected to the tune.

Song texts and airs

Bunting’s phonetical titles refer to song texts, even though both of the ms4.29 transcriptions are clearly harp instrumental settings.

A chailíní, a’ bhfaca sibh Seóirse

We have a song text collected by Patrick Lynch from Denis O’Hampsey in 1802/3, and preserved in Lynch’s manuscript QUB MS4.26 p12/2j. Under the page header “Fragments from Denis Hempson”, Lynch has written the title “A chailinigh bhfaca sibh seorse” (Girls, have you seen George?) and four lines of text beginning “A chailinigh chailinigh bhfaca sibh Seoirse”. This text must relate to our tune, both the harp version on QUBSC MS4.29 p.46-50 transcribed from Denis O’Hampsey, and the harp version I am playing transcribed perhaps from Higgins. Both of these harp instrumental versions are titled with phonetic spellings of the title “A chailíní, a’ bhfaca sibh Seóirse”.

Above the transcription of Denis O’Hampsey’s harp instrumental set of the tune, (QUB SC MS4.29 p46), Bunting has titled the tune “Callena Vacca Sheo Shurse / with English words”. It may be that O’Hampsey gave Bunting a metrical English version of the song, like he did for Burns’s March (QUB SC MS4.29 p.51). But if he did, it does not seem to be in the manuscripts any more. Patrick Lynch made an English translation; on QUB SC MS4.36.48 f144v, a page headed “Fragments from Dennis Hempson”, he has written four English lines, starting “Girls O girls have you seen George”. This would have been done in 1802/3, so it seems less likely that the note in MS4.29 p.46 refers to Lynch’s translation.

Conchubhar Mhac Coiréibhe

James Cody’s song air and words: facsimile of QUB SC MS4.5 p.20 with the text transcribed from ms4.6 p.63. Published by Charlotte Milligan Fox in the Journal of the Irish Folk Song Society, vol VI, 1908, p.26

We have a song-air version of the tune in James Cody’s manuscripts, written between 1805 and 1810 (QUB SC MS4.5 p.20, and QUB SC MS4.6 p.53/62). Cody’s title for the tune is “Connchúbhar Mhac Coirhéibe”. Cody also wrote down the words of the song in QUB SC ms4/6, p.63. As well as Charlotte Milligan Fox’s version, shown above, Donal O’Sullivan printed this text in his Bunting (1983 p16); I haven’t seen this manuscript page. O’Sullivan’s text begins: “A chnochúir, a chnochúir, a chnochúir ‘ac coiréibhe, buin a’ chluig ins a’ ghoirt agus déanamuíd réidhteach”.

On the preceeding p.62 is the title “[Conchubár] mhc Coirhéibe le seumus mhac párlaín air eilínóra iní ruarc”. The song text on that page begins: “Tá gach glaoighe dá mongaibh, is a loinne mar ghlan pheúrla”; Collette Moloney in her Introduction and catalogue (ITMA 2000, p.201) lists this as the song of Conchubhar Mhac Coiréibhe, but I think it must belong to Cody’s previous tune on p.53, Siobhán Bhán Ní Dhónaill. The Irish Song Project at Queen’s just adds to the confusion by transcribing the Siobhán Bhán Ní Dhónaill text from p.62, but offering as a “translation” Donal O’Sullivan’s translation of the Conchubhar Mhac Coiréibhe text from p.63, and stating incorrectly that Bunting “significantly ammended” Cody’s vocal melody “for publication in 1840”.

In general I think that Cody’s manuscript QUB MS4/6 with song lyrics and their matching tunes written on adjacent pages has been very much neglected; Bunting does not seem to have used it very much in his piano arrangements, and I am not aware of any attempts to re-combine these lyrics and melodies. This is a potential PhD or Masters project for someone!

We also have a version of the words of Conchubhar Mhac Coiréibhe written down by Patrick Lynch in 1802 (QUB SC MS4.17 f52v-53r, as listed in Moloney 2000 p.280). I haven’t see these pages. Thomas Hughes made a copy of the translation in QUB SC MS4.14.222.

McCabe’s verses on Carolan

Bunting writes above the p.32 transcription (perhaps from Higgins), “McCabe’s verse’s on Carolan to the same tune”. It is not entirely clear to me what this tag refers to. Charles McCabe was a very good friend of Carolan, a poet and a harper; Donal O’Sullivan, in his Carolan (1958 vol 1 p.67 on) gives us information about McCabe.

There is a story about McCabe playing a practical joke on Carolan, pretending to bring news of his own death; Carolan took the news very seriously and composed a lament for his friend McCabe, which begins “Ní Cathaoir mar chathaoir an Cathaoir fó ‘gcaoiniom” (QUB MS4.7 no127/191). but Donal O’Sullivan suggests (Bunting 1982 p.17) that Carolan’s lament for MacCabe should be sung to the tune Sgarúint na gCompánach / have you seen my Valentine (DOSB 25/57). He says that is a variant of our tune (which I don’t think is true) and suggests that Bunting’s tag is the wrong way round and should say “Carolan’s verses on McCabe”. This is a whole tangle in itself and deserves its own blog post.

There is another story of a practical joke (Donal O’Sullivan, Carolan vol 1 p.72-81) where, after a drinking match, Carolan tied McCabe up in a sack, and so the two poets exchanged satirical verses against each other. But there is no information about whether these improvised flyting verses were sung.

McCabe also composed a fine elegy after Carolan died, which begins “Nach í so an chuairt easbhach a laguidh mé réis mo shiubhail” (Donal O’Sullivan, Carolan vol 1 p.106). Is this what the tag “McCabe’s verse’s on Carolan to the same tune” refers to?

Carolan’s Elegy on the death of his wife Mary Maguire

Carolan’s wife, Mary Maguire, died in 1733, and he composed a lament for her, which begins “Intleacht na h-Éireann na Gréige ‘s na Róimhe” (Tomás Ó Máille, Amhráin Chearbhalláin p. 161). We are told by Joseph C. Walker (Irish Bards 1786, Appendix p.93) that the lament is sung “to the Irish Air of Concovar Mac Curely”

Interpreting the transcription and reconstructing the performance

The transcription of the tune of A chailíní, a’ bhfaca sibh Seóirse, perhaps from Hugh Higgins, on p.32 appears at first sight to be very clear, but there are problem bits which need addressed. The notation is at pitch, and the tune is in E minor, with a prominent and intrusive C (the notes in an E minor pentatonic mode are E, G, A, B, D). The tune is in 6-time, but Bunting has written the first three bars with only four beats. We can imagine that the E-G quavers at the start of these first three bars should be crotchets. Also, the bar count is off; Bunting has 9 bars in the first part of the tune. I think that Bunting’s bars 4 & 5 should be just one bar; his barline seems to be inserted.

In the second and the third section of the tune, Bunting has crossed out bars. I think he has deleted one bar in each section; I assume this is to do with the repetitive nature of the division-style music in these sections.

An interesting thing about this transcription is that it seems to represent a different class of transcriptions from many those I have discussed previously. A tune like Planxty Drury or Diarmaid Ó Dúda has the dots written first and then a neat copy on the facing page or lower down on the same page. But this transcription has the neat copy written over the top of the dots, almost entirely obliterating them. I think this can be more difficult for us, because we cannot see all of the changes that Bunting makes from his initial quick dots reaction to the performance, when he begins to think and make a neat edited version.

In the first two bars, the 4-note semiquavers a-g-e-d have a two-note quaver motif e-d written directly beneath. I wondered at first if this is a bass motif transposed up an octave, but it doesn’t seem to make sense like that. I feel that it is more likely to be an alternative way of playing that bar, if the a-g quick notes are a kind of optional extra. It is also possible that the e-d quavers represent how the tune might be sung – but that presumes that the rest of the notation is vocal, which I doubt. The transcription from O’Hampsey has this same feature.

The second section of the tune has double bars with repeat dots at the beginning and end, as if the tune were to be played as a four-part tune, 1 – 2 – 2 – 3.

The last section of the tune has a word written above the first bar. I cannot properly make this word out. I wondered at first if it might read “Sym” but there is a second tail, as if other letters have been inserted over the top. This final section is like a division variation, and can be compared to Denis O’Hampsey’s version which has a full set of division and other variations.

The same process of reconstruction can be done with O’Hampsey’s version, from Bunting’s transcription on QUB SC ms4.29 p.46-50. I did some work on that version 10 years ago but I wouldn’t now agree with a lot of my conclusions from back then! The O’Hampsey transcription has a lot of bass notes in the transcription, and is an important witness to old Irish harp right-hand (bass) practice. But the Higgins (perhaps!) transcription I am working from has no bass notes at all, and so I am playing it with no bass for this demonstration.

Other instrumental versions

As far as I can tell, this melody is not current in the living tradition of Irish music, but it does appear in one other old source. There is a baroque fiddle version in John and William Neal’s “Colection of the most Celebrated Irish Tunes proper for the violin, German Flute or Hautboy”, printed in 1724, p.18-19, where it is titled “Challeeny vacca shu sheorshe”.

facsimile edition by Nicholas Carolan, Irish Traditional Music Archive 2010, p.72-73

It seems to me that there is some kind of parallel between the sequence of variations in the Neal print and the sequence of variations transcribed by Bunting from Denis O’Hampsey, and indeed the single division variation transcribed by Bunting from Hugh Higgins (I think), which I am playing in this demonstration video.

Cornelius Lyons

On p.98 of the introduction to his 1840 volume, Bunting writes:

(No. 11 in the Collection) Chonchobhar Mac Areibhe. “Connor Mac Areavy”, known also by the name of Calleena bhacha su Seorse, “Girls, have you seen George?” – The melody is extremely ancient, and the variations by Lyons (Lord Antrim’s harper) are excellent. The modern musician will be surprised to find such an admirable arrangement by a person ignorant (as it is presumed all the Irish harpers at the beginning of the eighteenth century were) of modern musical science.

Cornelius Lyons (usually given as c1670-1740) was a contemporary of Carolan. He was harper to the Earl of Antrim. I think our main information about him comes from Arthur O’Neill’s Memoirs. O’Neill says “Cornelius Lyons was the other Great performer and a very fanciful composer especially in his Variations to the Tunes of Ellen a Roon, Calleena a Voch a thoo Shoarsha (Girls did you see George) Green Sleeves, the Cooleen, and several others. He was a County Kerry man” (QUB SC MS4.14 p.25). O’Neill also recounts some anecdotes of Lyons writing down a tune from Carolan’s live performance (the very first Old Irish Harp Transcription?) and travelling to London in the entourage of his patron the Earl of Antrim.

O’Neill tells us the names of two students that Lyons taught the harp, Hugh Quinn and Echlin O’Kane. Echlin went to Scotland and some of his repertory made its way into classical keyboard or pedal harp arrangements, including versions of some of Lyons’s compositions or variation sets.

It is not clear to me how we should understand Lyons. He was sighted, literate, and moved in cosmopolitan Anglo circles, including going to London. The works attributed to him in Bunting’s notes are mostly in the form of variation-sets, supposedly on traditional airs, but there seems some confusion about this even in our sources. Lyons’s variation set The Lady of the Desert is said by Bunting to be based on the traditional song air The Coolin, but there are clear differences and it is not immediately certain in my opinion whether this is true. Perhaps there is a parallel, in that A chailíní, a’ bhfaca sibh Seóirse might be Lyons’s instrumental variation set, which he composed based loosely on the traditional song air Conchubhar mhac Coiréibhe.

It is not clear to me how much this music is consciously based by Lyons on older traditional airs, and how much it is new composition with a traditional form and style. There is a lot of “classical” harmonic structure and content in these variation sets as well, especially the baroque “division” variations. It is not impossible that Lyons could have got some formal classical European music training in London as well as learning the old Irish harp traditions back home.

O’Hampsey played a number of Lyons variation sets, including Lady of the Desert, A chailíní, a’ bhfaca sibh Seóirse, and Eibhlín a Rún. Other harpers also had Lyons’s music; Hugh Higgins is listed as the source for Bunting’s transcription of Sliabh Gaillean, and also as one of the multiple sources for Lyons’s tune Miss Hamilton.

Untangling strands

As you can see there are a lot of tangled strands to this tune. One of the things I am trying to do here is to untangle things, to try and be more specific and to follow each thread and look to each specific fragment of the old tradition at a time on its own terms, to make genuine connections and to look sideways to make comparisons.

For this tune, Donal O’Sullivan has been clear and helpful, printing both of Bunting’s ms4.29 harp transcriptions alongside Denis O’Hampsey’s words of A chailiní, a’ bhfaca sibh seoirse (DOSB 6 and 6a); and printing Cody’s words and tune together of Conchubhar mhac Coiréibhe (MOSB 11).

Bunting himself has been unhelpful; he printed two different piano arrangements, both derived from the Denis O’Hampsey harp instrumental transcription, but under different titles in 1797 and 1840. Ann and Charlie Heymann were perhaps influenced by Bunting’s 1840 title when they sang Cody’s words of Conchubhar mhac Coiréibhe, to a synthetic combination of the two different harp instrumental versions of A chailiní, a’ bhfaca sibh seoirse, on their CD Cruit go nÓr (2006).

In this blog post I have tried to lay out all the different strands, the three different notations of Lyons’s version of the tune with variations; the song air from Cody, and the various sets of lyrics that are said to be related to different variants of the tune. I think that we can always learn new insights into the old tradition if we carefully separate out different threads, different versions passed down through different lineages and shown to us by different witnesses. And we can learn a lot also, if we can identify and then contrast versions that come through harp tradition, song tradition and fiddle tradition, because I think that each different instrumental or vocal tradition will take the material in its own idiomatic direction. I even think that different harpers had different regional styles, or different styles inherited from their teachers, so that if we have multiple variants transcribed from different harpers we might gain insights into how different individuals’ performance styles differed or were similar.

But in my playing in the demonstration video I am following my principle in this Old Irish Harp Transcription Project, in trying to make a realisation of one particular transcription of one particular version as notated live from one particular tradition-bearer at one particular point in the summer of 1792. I think we still have a long way to go to understand this tradition and this music before we can dive in and start creating new variants based on a deep insider knowledge of the old Irish harp tradition. But, I also feel that we are getting closer…

Tá Mé Mo Chodladh, from Hugh Higgins in 1792

I made a demonstration video of Tá Mé Mo Chodladh, from Edward Bunting’s live field transcription of the playing of old Irish harper Hugh Higgins in 1792.

Edward Bunting wrote his live field transcription into a little collecting pamphlet, which is now bound up with a lot of other collecting pamphlets into Queen’s University Belfast, Special Collections, MS4.29 page 28/28/37/f13v.

We can check my transcript of the text in ms29 to see that, above the notation, Bunting has written the title of the tune:

Ta me ma halla / Im[..] asleep and
na russkin me / dont waken me

Pádraigín Ní Uallacháin suggested to me that this can be understood as a phonetic attempt at “Tá mé mo chodladh, nár dhúisce mé” (I am asleep, may I not awaken).

Below the notation, Bunting has written

from Hugh Higgins
in 1792

It’s very unusual for Bunting to give us provenance or date information in his transcriptions. Normally we have to collate provenance information from later piano arrangements or indexes, which can be contradictory or unreliable. Of course, we can’t be sure that this title and attribution were written onto the transcription page at the time – nothing is certain in Bunting’s manuscripts!

Bunting’s notation on p.28 is unusually detailed and clear for a 1792 transcription. He has not barred the tune, but he has written in note stems and beams to indicate the rhythm. Unfortunately he seems to switch his note durations, sometimes writing crotchets and quavers, and sometimes writing quavers and semiquavers. He clarifies his thinking on the facing page 29/29/38/f14r. This edited copy is perhaps a little closer than usual to the transcription, but he has still made edits, omitting some of the notes, and expanding some of his “tr” or “+” marks into little notated note clusters, which may reflect Higgins’s harp practice, or may alternatively reflect Bunting’s piano imagination.

In my demonstration, I have used two different fingerings for these trills or ornaments. The trills or ornaments on E (bars 7, 15, & 31) can easily be fingered 2-1-2, which is cúl aithris in the table of “graces” (1840 intro p.25). However, I found this less convenient for the trills or ornaments on B (bars 4 & 12), and there I have used 3-2-4 which is not in the table of graces. I rationalised it to myself as a variety of barrlúdh, although the bárrludh gestures seem to rise up to finish on the upper note. Interestingly, the very next page on ms4.29 shows a live transcription of Burns’s March, which also includes a similarly shaped ornament (system 2, notated f-a-f) which is in a position where 2-1-2 fingering (cúl aithris) does not work very well. We have usually assumed this notation of Burns’s March is transcribed from Denis O’Hampsey, but I am now thinking that it is not in an O’Hampsey group, and it is noted one note up, whereas all of O’Hampsey’s transcriptions are noted at pitch. I have wondered before if this transcription of Burns’s March might be from Higgins; and so it is notable that this unusual ornament matches.

Bunting made a little note to himself on p.29, writing “major” above bar 17 where the flat 7th first appears. I am starting to seem more of Bunting’s naivety about this music in 1792 transcriptions, and a greater confidence in the 1796 transcriptions. (see Bunting’s collecting trips for more about these different tours).

Bunting did not take this transcription any further; as far as I can see he did not develop this version of the tune into a piano arrangement in any of his manuscripts or books.

Hugh Higgins

According to Bunting (1840 intro p.64), Hugh Higgins was 55 years old in the summer of 1792, so he must have been born c.1737. He is said to have died in 1796.

Higgins was later described by Arthur O’Neill, in his Memoirs: the first draft is QUB SC MS4.46 p.16, and the neat revision is QUB SC MS4.14 p.20. The two texts are slightly different, and O’Neill seems to be talking a bit sideways, but he tells us that Higgins was from a respectable family in Tirawley, Co. Mayo. His mother’s name was Burke; he was blinded early on and his parents sent him to learn to play the harp. He did extremely well at his studies and became an excellent harper: “one of the best I ever heard” says O’Neill in ms4.14. Higgins had “genteel manners”, and travelled in the style of a gentleman, with a servant boy; he “spared no expense” on his clothes, but because he was blind he had to trust his boy to choose the colours.

The young William Carr said in 1807 that Higgins “played very well and had an elegant harp”. (ed. Angela Byrne, A Scientific, antiquarian and picturesque tour – John (Fiott) Lee in Ireland, England and Wales, 1806-7 Routlege 2018)

I have not yet properly investigated Hugh Higgins. From looking at my ms29 text transcript and my tune list spreadsheet I notice that there are groups of tunes transcribed live from Higgins by Edward Bunting in 1792. (Higgins died in 1796, so presumably missed Bunting’s ’96 collecting tour). There is this possible group around Tá mé mo chodhladh on p.28, including Toby Peyton on p.26, and perhaps also Burns’s March on p.30 and A chailíní, an bhfaca sibh Seoirse on p.32. There is another group from p.78-90: Mild Mable Kelly, Sliabh Gaillean (Slieve Gallen), Cupán Uí Eaghra (O’Hara’s cup), Caitlín Ní Uallacháin (Kitty Nowlan), Thugamar féin an samhradh linn (with variation), Rois Dilloun / Young Lady Dillon, and Dr. Hart.

This group also includes a tune list on QUB SC MS4.29 p79, titled “From Hugh Higgins”:
John Jones Carolan ill
Grah ga miste very old
Cathleen ne Oullahan
Anthony [Mo] [Sonav-]
Slumber Maggenis Carolan

Perhaps most intriguing is the notation of Sliabh Gaillean on page 80-81. Page 81 has the unusual variation with a melody played in the bass register. Underneath this, Bunting has written the scale of a harp, and marked out the “sisters”, and the “right hand” in the bass and the “left hand” in the treble. Does this give us the gamut of Higgins’s harp, and did Bunting note down this scale as part of noting down Higgins’s bass variation to Sliabh Gaillean?

Other versions

Tá Mé Mo Chodladh is a lovely traditional song air. There are variants known in Scotland, and there are other related tunes and songs. For our purposes it might be worth noting two other variants of this tune.

As well as his live transcription from Hugh Higgins in 1792, Bunting also made a live transcription, apparently from the playing of Denis O’Hampsey, possibly in 1796. This transcription is on Queen’s University Belfast, Special Collections, MS4.29 page 171/169/178/f84r. It is titled “Tame ma Collad na Duss”, which Pádraigín suggests represents “Tá mé mo chodhladh ná dúis…” (I am asleep, let me not…)

You can see that it is done in a very different writing style, which suggests to me Bunting’s 1796 collecting trip(s). It is also in section 3 of MS4.29, but I don’t know that the sections of the manuscript are organised chronologically at all. This transcription has a lot of notes marked “B” which represents bass notes. In 1798, Bunting developed this p.171 transcription into a piano arrangement (QUB SC MS4.33.3 p56-7) and he writes there, “This setting is exactly set <copied> from Hempson – both Bass & Treble”. Of course it isn’t, as it is a piano arrangement in the key of f major, but it does have spare bass octaves in exactly the places where he has “B” marked in the transcription. There is a more developed piano arrangement frim the 1830s in MS4.27, where it is tagged Dennis a Hempson, Magilligan, 1796; and the published piano arrangement in 1840 is labelled in the index Hempson, Magilligan, 1792. (for all the fame of Bunting’s published books of piano arrangements, there are very few recordings of his arrangements being played on the piano, and I can’t find one at all of Tá Mé Mo Chodladh, no.100 in 1840)

I also was interested to consider a traditional recording of Tá Mé Mo Chodladh as an instrumental slow air. This is Willy Clancy in 1969:

Charlotte Milligan Fox published notation of a similar version, collected by her from “Piper Kelly, an old musician who wanders round the North of Ireland… he said this lament was given to him by “ould Jimmy Joyce, a Galway piper” who had been to the Belfast Convention of Harpers in 1792″.

Journal of the Irish Folk Song Society, volume 1, [1904], p.16

I would say that Higgins and O’Hampsey are much closer to each other, but rather different from Clancy & Kelly, who are much closer to traditional song versions. But I think that O’Hampsey is also “squarer” than Higgins, that Higgins has potentially more of the flavour of Clancy & Kelly. The main difference I am hearing between the harp versions, and the pipe versions, is that the pipers have a sharp 7th whereas the harpers have a flat 7th. I don’t know if this is because of the pipe scale, or if the pipes are following singers, or what the reason is.

Again, I wonder what the connection is between the performance style of our harpers a bit over 100 years ago, and this piper, a bit over 50 years ago. If Bunting tried to make a transcription of Willie Clancy’s playing, and then worked up his transcription into a piano arrangement, what would he end up with? Should we try to understand and interpret the 1792 transcription from the playing of Higgins, through an understanding of Clancy’s recording?