WHEN I GLANCE OUT my window in the Outer Battery at dusk, I see, across the harbour, the lights of St. John’s. Acting as a mirror, the window simultaneously reflects the inside of my house. I am a trained musician, brought up with a typical North American classical music education. I have, in addition, learned and studied Newfoundland music though the oral tradition — "by ear," directly from older musicians — over the past 25 years. I believe this has given me an interesting perspective: my window reveals both inside and outside at once.

Since the cultural resurgence that began in St. John’s in the 1970s, Newfoundland dance music has been slowly emerging from the obscurity and neglect it endured for several decades. While this is to be applauded, it is regrettable that so much of this dance music has been lost from the oral tradition. The tunes that have survived in the public consciousness, such as "Mussels in the Corner" or "I’s the B’y" are truly the tip of the iceberg. Many lively, unusual, and worthy tunes exist, which most Newfoundlanders and Labradorians under the age of 70 have never heard.

The 1950s in Newfoundland and Labrador was a decade of political and cultural change. It brought fish plants, and thus the eventual abandonment of the production methods for saltfish on which many of our traditional social structures and roles had been based.[2] Union with Canada in 1949 brought us the baby bonus and, finally, the end of the truck system. It also brought ridicule and condescension from some Canadians, and this continues to this day (Conrad "Mistaken Identities?"). Many pro-confederation Newfoundlanders in the 1950s were only too happy to lose any possession, activity, or attitude that made them appear different from other Canadians, and thereby subject to derision (Taft 16). A government which valued modernization and failed to recognize merit in what already existed, aided and abetted the process of throwing babies out with bathwater. The abrupt change in domestic architecture, from traditional two-storey houses to bungalows is now our most visible reminder of that time. We threw out or gave away our material culture: we destroyed our handmade furniture to buy chrome sets from Eatons or Simpson Sears. Less obvious to us now, from our 21st-century vantage point, are the changes in our intangible culture. We embraced a newly imported style of music (country and western) and began composing songs in that genre. We began dancing after the fashion of contemporary North Americans. Centuries-long traditions of dance, and the music which accompanied those traditions, all but vanished in little over a decade. The government policy of resettlement in the 1960s, the largest enforced re-settlement in Canadian history, uprooted 30,000 people from their homes. This dislocation and dispersion of communities ensured that many local dance traditions, which might have rebounded from the effects of the 1950s, never recovered. The old square dance traditions still exist in some scattered locations, but only as performance art — a curiosity to be done for tourists or at a "come home year" event, not as a living cultural manifestation engaged in by the whole community.

In his seminal work defining his theory of "tune families," "Prolegomena to a Study of the Principal Melodic Families of British-American Folk Song," Dr. Samuel Bayard identified a repertoire of melodies which are common to the British Isles and the British diaspora. A tune family according to Bayard, is "a group of melodies showing basic interrelation by means of constant melodic correspondence, and presumably owing their mutual likeness to descent from a single air that has assumed multiple forms through processes of variation, imitation, and assimilation" (33). He identifies several manners in which melodies are apt to be altered through the influence of the oral tradition. Although he confines his study to song melodies, he refers throughout the article to dance tunes and tune collectors, and indicates that one of the processes of variation is a recasting of melodies to different uses, such as marching or dancing. Indeed, variants of a melody may be found occupying many functions, and there is no clear delineation between song and dance tunes. Whether or not one subscribes to the idea of a single "parent tune," Bayard’s enumeration of how and where similar tunes tend to diverge is both astute and useful.[3] In my experience, the Newfoundland example clearly indicates that not only is Newfoundland and Labrador heir to the "common repertory" but also that many of Bayard’s astute observations are applicable to dance music as well as to song melodies.

The repertoire of dance music in Newfoundland is a mélange: tunes attached to dances of various sorts that came over with settling peoples; melodies appropriated from songs or hymns and put to use for dancing; tunes which have evolved from other tunes by rhythmic alteration or other variation engendered by oral tradition; locally composed melodies; and tunes which have more recently entered the Newfoundland tradition by way of recordings, radio, and other media.

In common with southern Ireland, where our Irish immigrants originated, Newfoundland has quite a high proportion of tunes in 2/4 time. Called "polkas" in the British Isles, they are referred to as "singles" in Irish areas of Newfoundland,[4] or, if you live on the Great Northern Peninsula, "straight" tunes. However, outside the Codroy Valley and the Port au Port/St. George’s area which were settled in the mid-nineteenth century by Scots and Acadians from Cape Breton, we have comparatively few reels.[5] (A reel is a tune in which each beat is divisible by four, properly notated in sixteenth notes with a 4/4 time signature, or eighth notes in 2/2.) This may be explained by the fact that the popularity of reels was just beginning in England and Ireland in the late eighteenth century, and it may be that not very many reels had reached southern Ireland or the English West-Country from Scotland before the main emigrations to Newfoundland.[6] Perhaps the repertoire of tunes brought by earlier immigrants, which would have contained no reels at all, reduced the ratio. Square dances in other parts of North America were often done to 4/4 tunes. This was rarely the case in Newfoundland where tunes in 2/4 and 6/8 predominated. For whatever reason, we seem to have fewer 4/4 tunes than other cultures with a similar ethnic mix. Jigs, or 6/8 tunes, which are termed "doubles" in most parts of Newfoundland, we have in great quantities. We also have some horn-pipes, which, along with the reels, we call "step-tunes," except in Irish areas, where they are called "triples" or "tribles." Confusingly, in some areas such as Bonavista Bay, a hornpipe or a reel can also be referred to as a "double," that being the name attached to a particular step-dance.[7]

The naming of tunes, like the "Naming of Cats,"[8] is a difficult matter. Some tunes have no name associated with them; others have three different names or more. Tune names often change from community to community, and sometimes from one musician to the next. The least inscrutable are those tunes we hold in common with other English-speaking cultures, some of which have names which never belong to more than one tune: "Smash the Windows," "Saddle the Pony," "Speed the Plough," and "Haste to the Wedding" are all in this category. These tunes crossed the Atlantic attached to a dance of the same name, carried in the heads and feet of immigrants. Other tunes might still bear the name they arrived with, but are not associated with any particular dance — "Tatter Jack Walsh," "Larry O’Gaff," "Stack of Barley," and many more. Slightly less scrutable are tunes which are recognizable as common and familiar, but which have acquired a sensible everyday name. Tunes in this category are widespread. "Doggie Doggie Bark at the Cat" (or "Bugger the Cat," depending on your location in the province and your present company) is known in other places that are heir to Bayard’s "common repertory" as "Off She Goes." Our "Auntie Mary" (alternatively "Chase Me Charlie") is otherwise known as "Cock of the North." Frequently, these are tunes to which a rhyme has been made to fit the rhythm of the melody, and a few words or a line of the rhyme suffice as the name of the tune. The MacEdward Leach and the Songs of Atlantic Canada website (2005), created by the Centre for Music, Media, and Place at Memorial University, illustrates this perfectly. Three performers give us a rendition of the hornpipe widely known as "Stack of Barley." One is unnamed, another is called "The Devil in the Garden" ("Did you ever see the devil in the garden digging praties"), and the third is named "Coming down from Jerry’s with a Bucketful of Berries." Yet another rhyme for this tune was given to me in Happy Valley, Labrador, by Aden Clark: "Did you ever see the devil with his wooden spade and shovel, did you ever see the devil with his ears sticking up" (Smith MD 15/09/04).

More examples of local names include tunes that have become associated with a dance, or part of a dance, in the repertoire of one community or region. Several different tunes in Newfoundland are called "Round the House" or "Form a Line" after parts of the square dance. Rufus Guinchard played a tune known elsewhere as "The Muckin’ of Geordie’s Byre," but his name for it was "Strip the Willow," as that was the name of the dance for which it was played. Jack Fitzpatrick of Lord’s Cove had several 6/8-time tunes which he named for dances, such as "The American Eight," or "The Six-handed Reel." In common English parlance, the word "reel" has come to refer exclusively to the musical form: a strophic, usually bipartite form in 4/4 time with symmetrical eight-bar strains and a sixteenth-note subdivision of the beat. An older usage survives in Newfoundland. Here, "reels" can be singles (in 2/4) or doubles (in 6/8), because our title refers to the dance itself, and not to the musical form. The tune for "Kitty Jones Reel," which is a double, is another example.[9] Tunes could also be named for the fiddler from whom they were learned (a common practice of Rufus Guinchard’s), the person whose favourite tune it was, or the person locally famous for dancing to it.[10]

Bayard’s study of song and ballad melodies led him to the belief that, with a few notable exceptions, "a two strain, or one strain tune, with eight bars to a strain, is normal" (20). In The Melodic Tradition of Ireland, James Cowdery illustrates clearly what is considered "normal" for dance music, not only in the Irish tradition, but in all the places which are heir to Bayard’s "common repertory": "A basic binary [two part] form is built into the dance music in which each ‘strain’ (or ‘part’ or ‘line’) is usually played twice. In addition, these strains are almost always clearly divisible into two or four equal phrases" (15). As we shall see, much of Newfoundland dance music does not accord with this definition of "normal."

Traditional Newfoundland tunes are in many ways straightforward, plain clothes kinds of tunes, highly suitable for step-dancing. They are played with few ornaments and with a strong pulse which is equally divided; eighth notes and sixteenths notes are played with no "lilt" or "swing." For all the plainness and simplicity of Newfoundland tunes, though, they have one quite startling and almost mystifying attribute. A significant proportion of them do not exhibit "normal" eight-bar symmetrical strains. As in some of the music of Québec and Acadian New Brunswick, many of our tunes just do not "come out right." This can be explained to the non-musician in terms of verse. Most "normal" tunes scan exactly like a simple rhyme. For example,

To \market to market to \buy a fat pig, \Home again home again \jiggety jig

To \market to market to \buy a fat hog,\Home again home again \jiggety jog.

Here we have one strain of a tune, divided into two symmetrical phrases of eight beats, or four musical bars (Each beat is equivalent to one trochaic foot. The beats are indicated by bold type and the bar divisions by \.) The phrases themselves are symmetrically divided into four-beat (two-bar) units.

In Newfoundland, however, this tune might just as commonly sound like,

Oh, to market to market to market to buy a fat pig, home again jig

Oh, to market to market to buy a fat home again jiggety jog jog.

The effect of this oddity on the listener can be quite disturbing. To those raised to believe that traditional dance tunes originating in the British Isles come in neat four-or eight-bar packages, it sounds as if the tune has gone severely wrong: the musician is drunk, incompetent, or has commited a grievous musical sin. I have encountered people who consider that the crookedness of such tunes is proof of their debasement, and that they should be abandoned in favour of the "correct," inevitably these days meaning Irish, versions of the tunes, which might be found in books, if only one could discover their deep and inscrutable singular name.[11]

A visit to Thesaurus.com might lead one to believe that asymmetry is not a highly valued ideal in English-speaking cultures. The following words are listed as synonymous:

aberration, abnormality, asymmetry, buckle, contortion, corruption, crookedness, damage, defacement, defect, depravity, disfigurement, evil, grossness, hideousness, impairment, injury, irregularity, knot, malconformation, malformation, misproportion, misshape, misshapenness, repulsiveness, ugliness, unattractiveness, unsightliness, warp.[12]

Few of these words have any positive connotations. In spite of this, there seems to be a burgeoning appreciation for these unusual melodies. An internet search gives 296 hits for "crooked tunes" including, among other things, information on CDs of crooked tunes, a radio program entitled "crooked tunes," and discussions of crooked tunes on a tune-player’s bulletin board.[13]

Observation shows that our "defective, evil, hideous, ugly" tunes come in two different varieties: those whose extra beats occur at the beginning or end of a phrase or strain, and those whose metrical "disfigurements, depravities, and malformations" are an internal, more intrinsic anomaly. In tunes of the first category, an extra beat has been inserted either before or after a regular four- or eight-bar strain, and functions as a kind of punctuation between repeats of the tune. Those beats appearing at the end of the phrase are cadential appendices or addenda. Bayard covered changes in these locations under his discussion of "alternating cadential formulae" (29). For convenience I have taken to calling anomalies such as this one a "jog," as in "Home again home again jiggety jog, jog." In rarer cases, an extra beat occurs in the middle of the strain, at the end of a phrase or halfway through the A or B part ("to buy a fat pig, pig," or "jiggety jig, jig"). Rather than call this a "jig," which might easily be confused with the established dance music term, I refer to it as a "median jog." A jog is invariably the same pitch as the note it follows. Two relatively long notes of the same pitch are a very common cadential formula in Western musical cultures, the most obvious example being "Amen." It is interesting to note that all hornpipes end with a couple of unison notes in a similar manner. Readers of a certain age may remember the "Sailor’s Hornpipe," which was the theme music for the "Popeye" cartoons. Sullivan took advantage of this defining hornpipe characteristic when composing Sir Joseph’s song in the operetta HMS Pinafore. A clear example is provided by the following:

Bayard documented changes effected by "the influence of other melodies con-temporarily current in tradition" (18). It seems likely that in addition to being influenced by their contemporaries, melodies can also be influenced by other genres of tunes. In our tradition, musicians seem to have altered the cadences of singles and doubles to conform to the repeated-longer-note hornpipe cadence formula.

An extra beat can also occur at the beginning of a phrase; some tunes display a "pick-up" beat in front of the tune and include it on each repetition of the tune. The verse equivalent is putting "oh-" before selected lines of the rhyme. "Oh, to market to market, to buy a fat pig, Oh, home again ...," etc. Periodically a tune is played with both an "oh" and a "jog." Mrs. Clara Belle Fennelly’s version of "Doggie Doggie Bark at the Cat" is a case in point. Example 2 (Smith MD 16/07/02) starts with an "oh," continues with the requisite eight-bar/sixteen-beat strain (divided into symmetrical eight-beat phrases), and ends with a "jog" for a total of ten bars/eighteen beats in the first strain. Similar examples may be heard via the Internet, on the MacEdward Leach website.[14]

Occasionally a tune will exhibit the opposite characteristic: a cadential beat, which would have made the tune symmetrical, is lacking. Amateur folksingers routinely do this. A classically trained musician will hear it as a mistake: the performer has obviously neglected to fill out the bar with the appropriate number of beats between phrases. The following example was collected in St. Mary’s Bay by Wilf Wareham. It is a version of a tune known in Ireland as "Maggie Pickens." The second cadential beat is missing from the end of both strains.

The second category of tunes display rhythmical anomalies in the middle of the phrases. It’s not as easy with these tunes to pinpoint any changes by which they might have arrived in their present condition. In many cases, it is impossible to discern whether there was originally any four- or eight-bar phrase structure, and, if so, what it might have been. Again, the manner of singing songs might play some part. Traditionally, singers sang unaccompanied and had a certain freedom in stretching the phrases to accommodate dramatic events (or extra metrical feet) in the lyrics. In some of these tunes, the phrases may be the requisite eight bars, but they will not divide evenly or symmetrically into four-bar multiples: an eight-bar phrase may divide into five and three bars, or some other grouping instead. This is the first part of a tune from John Joe Pidgeon, which he calls "Mickey Clarke’s Tune" (Smith MD 21/07/04). Others may have phrase lengths of any number of bars/beats.

Example 5 is the second tune Mrs. Fennelly played for me that she learned from the "bankers" — fishermen who had come ashore from the banking schooners to buy bait in Aquaforte. The first part has two phrases of nine beats each, with a pickup, for a total of nineteen beats in the strain.

My first exposure to these rhythmic characteristics was the playing of Rufus Guinchard, whose repertoire contains a high percentage of these "crooked" tunes. Guinchard was the musician of choice for dances up and down the Great Northern Peninsula from about 1913 to the early 1970s, when, in an effort to encourage the people to resettle, Parks Canada bought and destroyed Roberts Lounge and Gas Station in Sally’s Cove (Brookes, pers. commun. 2005). The lounge was, according to Kelly Russell, in Rufus Guinchard: The Man and His Music, probably the very last venue in Newfoundland that hosted a regular square dance. In a poignant story recorded on the Red Ochre film, Rufus!, Guinchard relates his experience of having somebody turn on the jukebox while he was playing at Maynard’s Lounge in Hawkes Bay.[15] He never went back to play there, but maintained his repertoire by having a "little play after supper." Not counting waltzes, which became popular in Newfoundland in the 1940s and were never used for square dancing, 48 percent of Guinchard’s tunes which are published in Kelly Russell’s Collection Vol. 1 (2000) display asymmetrical strains. Of these, 25 tunes display category 2 asymmetry, and 18 category 1. Three have both.

I initially thought this quality of phrase-length eccentricity in Guinchard’s repertoire must have something to do with the French influence on our west coast. As the French shore rights ended only in 1904, and Guinchard was born in 1899, there is little doubt that those people he learned from had at least a few French tunes.[16] At least a dozen Breton fishermen were both fishing and living along that coast in the second half of the nineteenth century, and some lived into the twentieth.[17] One was Guinchard’s grandfather. Anyone who has heard any quantity of the music of Québec, which is unarguably French-influenced, must needs come to the conclusion that a significant amount of the music of that province also displays this (to some people, disturbing) quality of not staying within the confines of a perfectly good regular phrase structure.[18] Contemporary Québecois musicians refer to these tunes as "croche" or, in English, "bent," and have a special appreciation for them. I am aware of two tunes in Guinchard’s repertoire that are also found in Québec. The tune he called "Old Boney" can be found on the independently released CD, Airs Tordus, performed by Les Têtes de Violon.[19] According to the liner notes, their version comes from the repertoire of fiddler Louis Boudreault — his name for it was "La gigue a Philibert" — and it is widely known in Québec.[20] Kelly Russell encountered the other tune in Québec City:

once while visiting the Museum of Civilization in Quebec City I heard a recording from the early 1900’s of a then popular Quebec singer, Madame Bolduc, and the melody she was singing was a tune Rufus always played called Sam Sinnicks’ Tune. (Russell 2000, v)

Many of Guinchard’s tunes such as "McBen," "Israel Got a Rabbit," and "The Hound’s Tune" have short motivic melodies, quantities of drones, and sections which are played three times instead of twice, characteristics which give the tunes a "French" aspect. Although I have not identified French or Québecois cognates for these tunes, it seems quite likely that they, too, have French roots. Fiddler and musicologist Anne Lederman believes the quality of "crookedness" in the western Manitoban Métis fiddle tunes she studies is related to the influence of Aboriginal musical aesthetics, and that it is possible that the asymmetry of some Québecois tunes may at least partly be due to "western Native influence working its way back along the fur trade routes into the east" (48). She suspects that the presence of crooked tunes in the repertoires of fiddlers in other regions of Canada may also have to do with the influence of First Nations musicians on the tunes (Lederman, pers. comm. 2003). While French or Aboriginal influences might contribute to the quantity of crooked tunes found in localized places in the province, it is not the whole story.

As I travelled through the province and learned more of our tunes, I discovered that dance melodies with irregular phrase lengths were more common than I had previously thought. Although the percentage seems to increase in regions that have enjoyed a French presence, crooked tunes exist in other areas, not just those in proximity to First Nations cultures or Gallic influence. Without in any way trying to be exhaustive, I made a quick study of a few published and unpublished sources available to me to try to ascertain whether or not they exist in any appreciable numbers. I excluded waltzes, music from the French Shore, and Scottish tunes from the Codroy Valley.[21] The quantity that I found surprised me.

There are many asymmetrical tunes among the accordion music I have examined: of Mrs. Belle Fennelly, accordionist from the Southern Shore, ten out of 23 pieces are asymmetrical; Cyril Flynn, Croque, four out of eleven. From Wilf Wareham’s collection: Leo Bonnell, from Lamaline, six out of nine; Mike Tremblett, St. Mary’s Bay, eleven out of 61, or 18 percent; Arthur Young, Branch, six out of eighteen. In this latter case, Mr. Young commences with several symmetrical tunes, which he discloses that he learned from recordings of Wilf Doyle, Harry Hibbs, and Ray Walsh. As he turns to local dance tunes, the music takes a much more circuitous route to the cadence. On the MacEdward Leach website: fifteen out of 22 instrumental tunes (including music played by Eddy Pimroy, Ray Noseworthy, Mike Kent, Mary Whalen, and Mike Whalen), or an amazing 68 percent, are crooked; this figure discounts the waltzes and the tune "St. Patrick’s Day," a set dance meant to be asymmetrical which came from Ireland in that condition.[22]

The second volume of Russell’s collection has comparatively few irregular tunes (for Newfoundland music) at 8 percent, still a significant figure in comparison with other dance music traditions. In contrast with accordionists, fiddlers have a higher percentage of "radio tunes" (tunes learned from the radio, as opposed to local oral tradition) in their repertoires.

In the first half of the twentieth century, the fiddle gradually gave way to the accordion as the instrument of choice for the dance. This was the result of two developments: the growing availability and popularity of accordions, and the changing venues for dances. Many community halls were built by such organizations as the Society of United Fishermen, the Orange Lodge, and the Knights of Columbus. In such large spaces, the fiddle simply could not compete with the louder accordion. In the words of Ruth Matthews, director of the Penny Folk Dancers (Burin), "a lot of people used to play the fiddle, but the fiddle wasn’t loud enough for a dance hall, and all the background noise" (Smith MD 20/07/04). The same shift has been documented in the Shetland Islands (Cooke 93). In addition, some venues would own an accordion for the express purpose of providing music for the dances, further limiting the opportunities of the fiddle player. Given that in many places the absence of electric power precluded the use of a public address system, an accordion seems a sensible alternative. Vince Collins, a fine accordionist from St. Anne’s, Placentia Bay, contributed that in his area this instrument was called the "priest accordion," as it resided with the priest when not in use (Smith MD 29/09/05).

In any event, because of these factors, many Newfoundland fiddle players maturing in the 1940s and 1950s opted to eschew the local dance-tune repertoire, which they seldom got a chance to play, in favour of "downeast," Scottish, or Irish tunes heard on recordings or the radio. Consequently many of these fiddlers will play "downeast" tunes like "Little Burnt Potato," "Maple Sugar," and "St. Anne’s Reel," or Irish tunes learned from McNulty Family records, but not local dance tunes. To quote Kevin Broderick of Bay de Verde, "I give up playing them tunes" (Russell 2003). The repertoire of dance music became associated with the accordion, an ironic turn of events considering that the ubiquity of fiddle for dance accompaniment only a few decades before had ensured that the title "fiddler" was bestowed on any dance musician, no matter which instrument he or she might happen to be playing.

The raison d’être of our instrumental music has been dance; consequently, the music and its style of performance has evolved for that purpose. Until the dancing ceased a few decades ago, most Newfoundlanders (with the exception, perhaps, of those living in the Codroy Valley and Bay St. George area) were not much inclined to listen to instrumental music for its own sake, or even for the sake of admiring the virtuosity of the player. Although the odd tune was played for enjoyment in the kitchen, songs were the entertainment of choice. Fiddlers and accordion players took pride in their ability to play for dancers. Some musicians felt they could not play at all if there was no dancer. Mac Masters comments, "When I was playin’, I want someone dancin’. If there was no one dancin’ I don’t want to be playin’" (MUNFLA tape 70-08 C674). Our relatively recent listening tradition, developed in the second half of the twentieth century, has produced musicians, now, who do not play for dances. This was not so in the past. The most accomplished musician was not the fastest or flashiest player, but the person best able to accompany the dance. In conversation with Wilf Wareham, Herb Reid recounts being requested to play at the Belmont in St. John’s, because the accordionist on stage, attempting to accompany a step-dance, played

a reel or something they plays up in Don Messer’s way sometimes, you know, only it wasn’t no good fer a step-dance.... but anyway he [the dancer] never stayed out two minutes, you know, couldn’t do nothing with it, you know. Like he said, ’twas a good tune fer what it was fer, but ’twas no good fer a step-dance. (MUNFLA tape 70-08 C771)

I have heard similar stories from different informants who have their favourite tale of "saving the day" for a step-dancer. Somewhat thinly disguised as praise for the step-dancer, the stories are actually an opportunity for a little modest self-aggrandizement.[23]

Music adapts itself to the requirements of the dance. This was brought home to me at a folk festival in England. Listening to the English band Waterson:Carthy backstage at the Warwick Folk Festival (and prospecting for inscrutable tune names, as usual) I abruptly realized that a melody they were playing was a version of a "single" tune widely known in Newfoundland and Labrador as "The Kissing Dance." Guinchard also had this tune, but as a double called "Uncle Manuel Milks the Cow." As played by Waterson:Carthy, the tune displayed a slowing down in one part, and then a recuperation of the tempo. I compared notes with Martin Carthy when the set was over. He told me that the tune was a Morris dance called "Balancey Straw," and that the ritardando/a tempo was meant to accompany a particular action of the dancers. He was curious about how the tune was played in Newfoundland, and whether our version included this alternation of tempo. When informed that our version had no such climax, and, moreover, was performed at almost double the Morris dance tempo, he opined that it no doubt had to do with the kind of dancing Newfoundlanders did (pers. commun. 2003).

Unless Newfoundland is completely different from other traditional cultures, the character of our dance music is directly attributable to our manner of dancing. Consequently the formal asymmetry of so many of our tunes is puzzling. Why, in a musical tradition which until relatively recently was inseparably linked with dancing, would there exist tunes in which the first part would have nine bars and the second eight, or perhaps ten? Or five, or seven and a half? One would logically expect dance music to exhibit the usual eight-bar/sixteen-beat strains. It does in the British Isles, where the majority of our tunes came from. It is the case in Atlantic Canada (with the exception of Acadian New Brunswick) and on the Eastern Seaboard of the United States, areas that were settled by a mix of ethnicities quite similar to ours. In these places you may find a very rare tune with asymmetrical phrases, perhaps in New England where a specific contradanse tune might have been crafted to fit a particular dance, or maybe in Ireland where competition set-dances require a certain phrase structure. You will find nothing to match or even come handy to the quantity of odd and quirky tunes that exists in Newfoundland, unless you travel to Québec, but again, not all our asymmetrical tunes have French influence. How and why, then, did our dance music develop this strange and wonderful quality, which seems so at odds with the purpose to which it was put? Did these irregularities develop before or after the dancing ceased? Were the crooked ones perhaps used for listening, and not dancing? Were these tunes originally crooked, or did they develop this attribute in Newfoundland?

To ascertain whether tunes may have been asymmetrical when they made the journey across the water, I consulted the excellent research resource Sources of Traditional Irish Music c.1600-1855 (Fleischmann et al. 1998), a compilation which includes many collections published in Edinburgh and London. I found a qualitative difference in the non-symmetrical Newfoundland tunes and those published in Sources. Phrases in the non-symmetrical Newfoundland tunes are not confined to multiples of four bars. In contrast, the great proportion of asymmetrical tunes in Sources display phrases that are either four bars, or a multiple of four bars long. Almost all of those that do not are either "airs," that is, song melodies arranged for instrumental performance, or versions of the tune "The Black Joke," a very long-lived melody with strains of six and ten bars, first published in 1729.[24] As Newfoundlanders did not have much of a tradition of listening to instrumental music until the dancing ceased, the "air" is even now not commonly found here, although we do have the tune "The Black Joke."[25] This exercise was of limited use in determining whether many of our tunes made the transatlantic journey in a symmetrical condition. The fact that our "unbalanced" phrases don’t come in multiples of four bars argues for the conclusion that the asymmetry developed in Newfoundland. On the other hand, there is no way of knowing whether the tunes in Sources may have, when collected, displayed category 1 asymmetry, which might subsequently have been "corrected" by the collector or publisher. In addition, published sources do not represent the entirety of the music of the time, merely that which happened to be published. In addition, quite a few tunes are reprinted. The exercise does, however, give us some kind of indication of the relative tolerance of a given time period towards non-symmetrical tunes.

Sources includes an "Index of Tunes by Category," which lists not only which tunes are jigs, which reels, which hornpipes, etc., but also which are standard (i.e., having two equal 8-bar phrases) and which are not, as well as dates. Mathematical calculation reveals the proportion of non-standard traditional tunes published in various time periods, though the percentages may be skewed slightly, since some published collections contain music that was composed for the publication, and was not actually found in the tradition at the time.

Pre-1700: 7.5
1700-1749: 13.43
1750-1775: 18.9
1776-1799: 5.69
1800-1824: 4.94
1825-1855 1.7

In the mid-eighteenth century, the percentages of oddly proportioned tunes are quite high, and decline the closer we come to the middle of the nineteenth century. The data seem to document an aesthetic shift in the early nineteenth-century British Isles towards more symmetrical phrases, a shift that appears not to have happened in rural Newfoundland’s oral tradition.[26]

What is passed along in the musical oral tradition is not limited to a series of notes. A set of characteristics are communicated with the tune from musician to musician, which enables us to categorize types of tunes: for example, fast tunes in duple meter with an eighth-note subdivision of the beat and an emphasis on the off-beat will, in most of Newfoundland, be categorized as "singles." Aesthetic values are communicated as well: what constitutes a "good" or "correct" performance, as opposed to a poor one. In rural Newfoundland, an aesthetic defining performances that display symmetrical musical phrases as "good" is not as evident as in other places.

Were these tunes, in this condition, played for dances? Yes, without doubt. The Leach collection attests to that. Many of the instrumental tunes Leach collected in 1951 and 1952 (68 percent of which, as I have mentioned, display asymmetry) were given to him to illustrate what was played for the square dance. On 14 August 1969, John Dollimount recorded a square dance at a wedding in Francois (MUNFLA 71-02), where the accordionist played three tunes with asymmetrical strains. In any event, there are just too many fine players who execute their tunes this way for it to be otherwise. Some would argue that a great majority of the metrical eccentricities, especially those of the first category, can be attributed to idiosyncrasies of individual performers. That, however, leads to the inevitable question: why do Newfoundlanders perform their tunes with so many rhythmic peculiarities? If one musician plays in this manner, it is an idiosyncrasy. If many do, it is a style. Logically, a certain percentage of the category 1 anomalies can be ascribed to incompetence, or lack of recent practice on behalf of the performer. This covers the situation in which the informant, who could have been providing the dance music as a service to the community and not because he or she felt particularly called to it, might not have played the tunes since the dancing ceased. Ned Keeping, a lovely and accommodating gentleman of Boxey who hadn’t played in many years, struggled to remember a few dance tunes. Many hesitations are evident on the recording he made for me, as he delved into his memory and eventually unearthed a few local gems (Smith MD 24/07/02). If I were to learn these tunes and incorporate the hesitations in my performance, the resulting changes in the tunes would be an example of "corrupt rendition," according to Bayard: "the result of faulty learning or bad performance of the tune"[27] (17). Certainly some of the contortions in our tunes can be ascribed to this. However, this "corruption" does not explain asymmetry in the repertoires of those musicians who didn’t "put it down" during the years when the dancing vanished and the music went unappreciated — fine musicians such as Rufus Guinchard, Mike Tremblett, or Belle Fennelly, who are obviously very competent and have a continuous record of playing the tunes.

There remains the question of whether the eccentricity of our music should march under the colours of musical style, or under the banner of performance practice. In some performances, especially of category 1 tunes, strain lengths were apt to vary from one performance to the next. I was fortunate to play with Rufus Guinchard fairly frequently in the 1980s, whenever he came into St. John’s for festivals or other performances. In some of the tunes he played, the beat structure seemed to be a somewhat moveable feast. You had to be on your guard not only because the tunes were so intrinsically metrically unusual, but also because he might, without notice, add or subtract a "jog" at the end of the first strain, leaving you flousing around in the second part of the tune wondering how you got cast adrift. Norman Ricketts, the accordionist providing the music for dances in Francois (MUNFLA 71-02) occasionally does the very same thing.

On MUNFLA tape 70-08 C674, on which Mac Masters is instructing Wilf Wareham in the finer points of dance music, he vamps in between strains of the music: he plays a couple of "jogs," then draws a note on the same button, which produces the supertonic note (it sounds like an attempt to get more air in the bellows) all the while rhythmically beating time with his feet. The impression given is that the exact length of the strain is not of great importance — what is more important is the consistency of the beat. As I discovered while playing with Rufus Guinchard, this practice of varying the strain lengths can cause difficulties in ensemble performance. It would have caused no problem for a dance musician who would generally have played un-accompanied. The custom of playing solo contributed to the fiddler’s freedom to lengthen or shorten strains, as there was no requirement to keep in time with another musician.

Since the demise of the dancing, Newfoundlanders and Labradorians have been developing a tradition of listening to, as opposed to dancing to, instrumental music. This has the effect of fixing the strain lengths of the tunes as musicians agree upon an arrangement for performance purposes. On occasion, this can lead to lively discussion. I have discovered that my colleagues often hear the beat structure of tunes differently than I. Comparing my own transcriptions of Newfoundland tunes with those made by Kelly Russell, Colin Quigley, or Evelyn Osborne, I find many differences in the "barring" of the tunes, and some rhythmic and metrical differences. This, I have discovered, is not because any of us has transcribed the tune incorrectly, but because the musical notation system fails us, and also because our ears have been trained to expect a beat structure, that in reality is not there. In fact, Newfoundland dance tunes (except for waltzes, which entered our tradition quite late) are not divided into bars at all, but into phrases consisting of a long sequence of single beats. This differs markedly from Western art music of the common practice period which has evolved patterns of metre such that some beats are more important than others. A waltz, to give a readily comprehensible example, will have a strong/weak/weak beat pattern, as anyone who has studied ballroom dancing or has been required to learn to waltz to attend a wedding, will immediately recognize: ONE two three, ONE two three. Four-beat patterns exhibit a strong weak, medium, weak pattern: ONE two Three four (To MARKET to market to Buy a fat pig). Newfoundland tunes, however, particularly our singles, lack this strong/weak beat pattern. Performance practice has evolved so that all beats have identical stress. The effect is of listening to music that has a long series of single beats. Instead of hearing a "ONE, two, ONE two" or a "ONE, two, Three, four" pattern, we hear "One, One, One, One, One, One," etc. Newfoundland tunes, which according to established musical practice, are transcribed in 2/4, 4/4, or 6/8 time, would more accurately be transcribed in 1/4 or 3/8. In some cases, particularly where the harmonic implications of the tune are ambiguous, this lack of strong/weak beat patterns essentially makes it impossible to discern whether the initial beat is a pickup or a downbeat. What may have been originally intended as a pickup beat may be understood by the learner to be as important a beat as any other, and repeated as such. I believe this particular practice in our oral tradition facilitated the process of change in at least some of the tunes. Problems arise, though, when a decision must be made for transcription or accompaniment purposes: when the music must be made comprehensible to our 21st-century ears and aesthetics. As I was rehearsing with Jean Hewson last year, the question arose whether "John Joe’s Single" (Smith MD 21/07/04) should be accompanied as if it started on a downbeat, as follows, or on an upbeat, thus: This lack of strong/weak beat patterns contributes to a more liquid understanding of the concept of a phrase, and so encourages and facilitates the odd phrase-lengths. There is much less compulsion to stop at eight beats when there is no metric structure which indicates the end of the phrase, and when, as was the case in Newfoundland, the musician plays unaccompanied.

Example 6a : John Joe's Single

-> See the list of figures

or on an upbeat, thus:

Example 6b

-> See the list of figures

How this practice developed is an interesting conundrum. I will attempt an explanation based on my knowledge and experience: my window gives me an interesting vantage point. I have played for dances, although certainly not as many as Guinchard, nor for as practiced dancers as he did. The dancing that happens in Newfoundland these days is more of a revival tradition and less of a community cultural endeavour. Even so, certain things come to light. The first is that if you are playing for several squares at a time, they will invariably get "out of synch" with each other. Rarely will you have a square of dancers which will keep strictly to eight- or sixteen-beat patterns. Even more rarely will two squares do this at the same time. In other traditions, where dances have been codified and taught as a performance art, such as Scottish Country Dance, phrase lengths are strictly observed: phrases of dance and music synchronize exactly. Any deviation is corrected. In England, the widespread popularity of dance collections, which began to appear as early as 1652, attests to the importance placed on proper execution. Instruction was available in towns. Dance was also widely instructed in eighteenth-century Ireland. Quite a lot of our Irish ancestors immigrated as fishing servants in the mid-eighteenth century, coincident with the apex of the itinerant dance masters who travelled the roads in Ireland, stopping in towns and villages for six weeks at a time to teach children and adults for pennies. Arthur Young, travelling in Ireland in 1776, noted both that dance instruction was very common among the Irish of all classes, and that 5,000 Irish left from Waterford each season to work in Newfoundland (Young 120, 135). It is very likely that at least some of our English and many of our Irish immigrants did receive dance instruction in the "old country," and this fact more than any other argues that tunes were symmetrical there, and acquired their "abnormality" in Newfoundland.

Dancing was taught in Newfoundland as well, at least in St. John’s. The Royal Gazette of 31 October 1811 contains a notice advertising the services of "Mr. Latour," who gave lessons in dance, German flute, and fencing.[28] On 13 August 1812 a notice announced a "Ball and performance of Scholars." The following advertisement was published in the Newfoundland Indicator, 12 July 1845:

Mrs. T. Preston begs leave to inform her friends and the public that she is now making arrangements for the purpose of instructing Young Ladies in that delightful and healthy accomplishment — DANCING, in all its diversified forms; and she would assure those who commit their children to her care that every possible attention will be paid to them. Mrs. Preston would have no objection to give lessons privately if required. Further particulars made known or applications to Mrs. Preston at her residence, Duckworth St.

Colin Quigley also came across some dance school advertisements in the Evening Telegram from later in the century. Quigley’s opinion is that this kind of remunerated instruction may have had some influence outside St. John’s, but that the social attitudes of the day did not allow for the modern types of dance to be accepted readily in the rural communities; therefore the older dances held sway (16). That is not to say that a returning sailor might not try to initiate his community into the latest dance trends he might have encountered on his travels. Ned Keeping of Boxey (Smith MD 24/07/02) told me that whenever the "bankers" were in, they might try to instigate the "quotillion [sic]," a dance that he would have no part of and for which he refused to play. Mac Masters, in conversation with Wilf Wareham, describes the dancing of the "American Eight" in Harbour Buffett. Because the local people were unfamiliar with it, "you had to call it off, see, we call it, you got the feller singin’ out what to do." Obviously "calling" the dance was rare enough that it needed explication. Belle Fennelly, who played many wonderful dance tunes for me, introduced the music for the American Eight as follows:

Now the next is American Eight now, Mr. Dick Curran used to ask for it, poor man he’s dead now, from Riverhead in Fermeuse. I’d play it and he’d call it off. (Smith MD 16/07/02)

It appears that as a latecomer into the Newfoundland dance repertoire, the American Eight needed to be called. My conversation with Ruth Matthews brought light to this situation:

I remember seeing the American Eight in Burgeo, that was brought to Burgeo by a man who worked at Argentia, p’raps in the mid thirties. And he brought that dance back, and he said the people down there, they’d told him that they’d got the dance from crews of American fishing schooners, banking schooners, who came into port occasionally. (Smith MD 20/07/04)

In any event, my experience with "calling" a dance is that it in no way guarantees that the dancers will synchronize their phrases with the music. Mrs. Matthews also mentioned "an old lady neighbour [who] said her papa had a book, the Lancers was in that, the Saratoga Lancers."[29] She also related that some simple dances, such as Speed the Plough, were taught in school by the nuns. These twentieth-century examples demonstrate that outport Newfoundlanders had some access to newer dance trends, but the access falls short of professional dance instruction. For most Newfoundlanders, until it declined, dance was not instructed, nor was it "called"; it was entirely in the oral tradition. Because of this, and because dance was a social occasion rather than a performance event, synchronizing the dance exactly with the music was not a priority.

In the Newfoundland tradition, everyone danced primarily for enjoyment, and virtually nobody had any formal instruction. If you had a novice or an elderly person in the square, it is quite likely that an extra beat or two might be required to get around the figure. Quigley notes, "In general practice, I have never seen music phrased exactly with the dancing, although ‘keeping time’ was frequently upheld as a mark of good dancing" (28). The following conversation between Herb Reid and Wilf Wareham is also pertinent:

Herb Reid: Up here in Creston, oh ’tis shockin’ up here in Creston. Wilf Wareham: They don’t, not like home, say, do the same thing at the same time? Herb Reid: Oh no, they don’t, you know. Some feller is swingin’ and another feller is dancin’. (MUNFLA C 771 70-8, 6:35)

And according to Evelyn Osborne, who describes a dance in New Bonaventure,

although one can see that the music and the steps do fall together, dancing in a small space with other people moving around, interrupting to say hello; someone getting confused and joining back in; a perfect 8 or 16 measure movement may go askew ... Nevertheless, the dancers keep dancing as long as the beat is held steady and simply waited at the end if there was extra music to be played. (68)

On MUNFLA tape 71-02, of the dance at a wedding in Francois, the dancers are quite audibly out of synch with the accordionist. The end of each section of the dance is marked with a sharp unison stamp of feet. Rarely do the stamps coincide with the ends of the strains of the music.

Speaking from the musician’s point of view, you have little choice but to shrug your shoulders and realize that the important thing for those on the floor is that their feet hit the ground in time with the music, and not that they get around their figure in eight or sixteen beats. If the dancers are not adhering to eight- and sixteen-beat patterns, musical phrase lengths and strong/weak beat patterns become much less important. Indeed, as a musician, you might like to de-emphasize a strong/weak beat pattern and regular phrase structure because it is at odds with the dancers; this, in practice, results in our long series of single beats. With the necessity of eight-bar phrases removed, and no strong/weak beat pattern, an asymmetrical tune will do as well as any other tune to accompany the dance.

The oral tradition in the 21st-century is different from the one Bayard analyzed in "Prolegomena." Transmission of music these days is more likely to take a detour through print or electronic media. Our recently developed "listening" tradition for instrumental music has taken this path. Printed sources for music, in my experience, have an effect upon the number and kind of variants of a melody. The existence of a published (notated or recorded) version of a tune seems to limit the distance which that tune can wander. Variants tend to be assimilated into the published version. Tune books gradually encourage musical literacy as people learn to read in order to gain access to the collections. Eventually, as more people learn to read musical notation, even those who do not read are encouraged to make their versions conform to those who are known to play what is perceived as the "correct" or book version. This process seems to inhibit the number and diversity of tune variants. In the Codroy Valley, Newfoundland’s enclave of Scottish tradition, fiddler Joe Aucoin told me "since I retired about eight years ago I learned to read music a little bit, enough to be able to get the tunes right, y’know" (Smith MD 27/07/02).

In contrast with Scottish and Irish tunes, very few Newfoundland dance melodies were written down until quite recently. Tunes that are obviously cognates of ours can be found in collections of music published in the British Isles, but as we have seen, they are often quite difficult to identify because, in the great majority of cases, the name of the tune has changed. Therefore musicians who learned Newfoundland dance tunes did so completely "by ear" from the oral tradition, and had no "correct" written texts with which to compare their version. In the UK, where much of the repertoire has been in print for a couple of hundred years, and where instruction in both music and dancing have been readily available, the tunes and the dances have a much more symmetrical condition.

To conclude, a high percentage of Newfoundland tunes display asymmetrical phrases because of the oral transmission of both dances and tunes, and because of the demands placed upon the music by the manner of dancing, in particular, the lack of synchronicity between dance and musical phrases, which fostered a decline of a strong/weak beat pattern in musical performance. An aesthetic that did not highly value symmetry may have facilitated the development of the melodies, as did the practice of having individual musicians, and not ensembles, provide accompaniment for dance. The manner of song performance may have played a role, as well as the lack of published sources for the tunes. Changes in the melodies are consistent, in location and type, with patterns noted by Bayard in "Prolegomena." From the Newfoundland and Labrador example, one may postulate that the greater the distance from an instructed or published tradition, the more likely we are to find asymmetry in the manifestation of music and dance culture: that symmetry is an artifice imposed by Western art music, and it will decline in traditional cultures not directly under its influence. A comparison with Southern Appalachian or Québecois dance music traditions may corroborate this premise.

The prognosis for Newfoundland and Labrador maintaining these unusual tunes is not optimistic. Radio and media have had a huge influence, to the point that most people under the age of seventy now hear the tunes as "wrong." Indeed, one of the fetters that kept the music in obscurity was the feeling by many musicians that, because it was not symmetrical like the music on the radio, it was somehow incorrect and not worthy of performance. The listening tradition specifies that instrumental music shall be accompanied, and asymmetrical tunes are a challenge for guitarists, and difficult for a band. There is almost no surviving oral transmission of these tunes. With no access to their own music, younger players, particularly in St. John’s, are learning more technically challenging tunes from CDs of Irish music. Fiddle and accordion music is becoming an instructed tradition in Newfoundland and Labrador. Generally teachers are not old enough to have heard the dance tunes, which, with the exception of Guinchard’s repertoire, have never been recorded or published. There are some shining rays of hope, such as Daniel Payne of Cow Head, a fine musician who has made a study of collecting local tunes and passing them on to younger players. The inclusion of some dance tunes in Kelly Russell’s collection of fiddle tunes is another. At the annual Vinland Music Camp, directed by Eric West, a team of teachers pass along local songs and tunes. And the recently developed North American appreciation and embracing of asymmetrical tunes may encourage musicians to bring our dance tunes out of the closet, "dress them up a bit" as Vince Collins says, and record them. Those of us involved in developing the Newfoundland and Labrador "listening tradition" have the responsibility to learn, play, and teach these tunes as we discover them.