Hreinn Steingrímsson's research was based on field recordings of Icelandic folk music, especially kvæaskapur, made between 1958 and 1974. Although early research began at the beginning of 20th century, almost no serious study was ever attempted. One of the first collectors was the Icelandic composer Jón Leifs (1899-1968) in the 1920's. He traveled with his pony and a machine that recorded sound on wax cylinders. When Hallfreur Örn Eiriksson began his collection in 1958, obviously new technologies such as jeeps and tape recorders were great improvements for this research.
The performers Hreinn studied were all quite old before they were recorded. The vast changes in Icelandic society were so intense that only those born before 1900 escaped these influences if they lived in an isolated region. The first kvæamaur I met was Thórur Gubjartsson in Patriksfjördur, a small town in northern Breiafjörur. He was born in 1891 and had spent his life as a fisherman. At the time he was working eight hours a day in a fish processing factory since he was too old to go out on a boat. At night he returned home to his small house, cooking and tending to his daughter and grandchildren, and late at night he was teaching himself English by translating the Bible. Thórur had not performed in years, except occasionally for his family or for himself, but he obviously loved rímur and was anxious to kvæa for anyone who sincerely cared to listen. Fortunately, his performances were documented on several occasions between 1958 and 1984, and his voice and delivery remained exceptional. His only weakness was his shortness of breath because of age.
Gunnar Alexandersson was the other performer I recorded. Born in 1897, he lived mostly on the northern part of the Snæfellness Peninsula on the southern end of Breiafjörur. He had spent most of his life as a fisherman on a small boat. Gunnar generally spoke in complicated Icelandic metres, and he was know as a "skilled versified" (hagyrdinger ). A remnant from a distant past, he continually made verses in his head and performed them in the old way. On his fishing boat he would kvæa for long periods spontaneously conceiving poetry about each member of the crew, the fishing situation, the "heroic" captain, etc. When I met him, he recently had a stroke and was living in a nursing home in Kverjageri, and we brought him to Hreinn's home in Reykjavík for the recording. When we first spoke to him, his responses seemed quite slow. Then I realized that he was composing rhymed replies. He performed verses about the cake he just ate, or about how expensive the rental car must be and how beautiful the woman sitting next to him was.

Normally when we think of a melody, we imagine a specific succession of pitches which is always the same and fits a particular text. However, in kvæaskapur this is not the case, so Hreinn uses the term "variable interval" referring the very unusual performance practice of not attempting to articulate precise pitches. When Hreinn analyzed recordings transcribed on to graphs by a machine in Sweden, he had recorded examples of people who said they were singing and performers who demonstrated kvæaskapur. He found that normally even the most untrained "singers" tried to sing precise pitches corresponding to a scale. However, the kvæamenn from Breiafjörur clearly did not do this; they performed within certain shapes with constantly fluctuating intervals. One tone was always kept exactly the same and usually ended each stanza, the seimur (Old Norse, "thread"), and the melody was named stemmur (Old Norse, "stream, or fluid"). Thus around this central tone the melody is always changing slightly in details and in shape. Even though a kvæamaur might perform mostly the same melody over and over, it would never be executed in quite the same way; there would always be subtle differences.
Since the kvæmaur performed daily for many hours at a time for as much as seven consecutive months, I believe s/he was in a state of concentrated meditation. On one level the seimur was the one exact tone which the melody ornamented, and with a good performer, this one pitch varied very little. And on another level, the length of the breath determined the rhythm. A good kvæmaur in his prime often sang an entire stanza without breathing, and the rhythm emerged from the divisions of a single breath. Thus the concentration was on expansions of a tone within a breath.
Another anomaly in this style is the disassociation of the meaning and mood of the text and the delivery of the melody in performance. Although a clear articulation of the text was important, the music was not a subordinate layer functioning to clarify the story. Instead, there was in a structural counterpoint between the layers of text and melody, each proceeding somewhat independently. The common factor was the verse metre. The rímur metres have a rhythm resulting from long and short syllables within two to four lines, with particular prerequisites for both rhyme and alliteration. Metres fall roughly into two categories, and any melody fitting a category could be used regardless of the content of the text.1 When we recorded Gunnar, he sang three segments from different poems with the same melody: one was an heroic tale about Atila, another about a rhyme game between a poet and the devil, and the last was a slightly bawdy song about a young woman. There was no perceptible difference in the emotion of delivery. In fact, often the structures of the poetry and music do not coincide. For example, in most of the verse metres Hreinn calls "Category I," the stanzas are four lines long divided into two couplets. However, in most instances when the performer has to breathe, the breath is not at the end of the first couplet but in the middle of the second one. In addition, the contour of the melody often accentuates words that are relatively unimportant.
Hreinn's categorization of the verse metre is unique, as he is the first to equate performance practice with types of metres. For example, Bjarni Thorsteinsson cataloged 2267 kinds of verse meters in 27 categories. Since Hreinn restricted categories to performance with particular forms of stemma -contours, he reduced the numerous kinds to two basic types and a relatively new third subcategory. The first has short contours which combine into the shape of the overall melody, whereas the second group has long shapes that outline the entire stanza.
Listeners most likely heard one performer and only a few different melodies during the long winter period, so the question becomes: what kept their interest? Part of the answer was that there were relatively few dramatic changes in the performances. The melody was constantly different, but it never changed drastically, and thus the listener was never forced to pay attention because of an intense alteration in either the music or poetry. The verse metre, and possibly the melody, generally varied about every fifteen minutes. If one decided to listen intensely, there were constant variations, never quite predictable, and if instead one focused more on his/her work, the story and melody flowed on very naturally. The narrative progressed slowly and was generally known ahead of time. Whenever motivated, the listener could join in on the last note of the stanza, sometimes holding it for a very long time. So the seimur and breath of the kvæamaur was a constant drone in the small room. The atmosphere was extremely intimate with a timeless continuity.

Until the 20th century, most Icelandic music could be considered folk music or highly affected by it. Certainly there were many external influences since Icelandic fishermen traveled extensively, and Iceland was governed by Denmark. But these influences were largely absorbed into the music the same way foreign literature was absorbed into older Icelandic styles. When Iceland adopted Protestantism after Catholicism (in the 16th century), most of the clergy could not read music extremely well, and so even hymns that were written down gradually transformed into "the old tunes." However, in 1840 the first harmonium (organ) was brought to the Reykjavík cathedral and new melodies were introduced. This was the beginning of complete transformation to the modern world. The organ could not only produce precise pitches of a scale but also play the same tune over and over in more or less the same way. The words clearly fit the melody. However, the "old tunes" and style of singing were practiced in private homes almost as long as kvæaskapur because of an extremely revered religious text by Hallgrímmur Pétursson from 1661 known as the "Passion Psalms." These psalms, written in old Icelandic metres, were too precious to be sung with the modern tunes and manners of singing.2 Kvæaskapur survived the longest because it was the least socially bound; it could be performed in a wide variety of circumstances unlike church music.
As more people moved into the city, new kinds of living structures were designed with more modern materials; concrete replaced turf. Kvæaskapur had existed in places where there was absolutely no resonance: outdoors, on a boat, or in a small room with a turf roof that quickly absorbed any sound.3 The newer churches and building let the sounds ring longer. Early Western music that developed fixed scales and intervals was conceived in resonant spaces where absolute intervals resound so strongly that the room vibrates. The kvæamaur's sound world evolved around the subtle qualities that could be invoked in quite an opposite environment.4


1 This occurred in other cultures such as in early North American cowboy music. Often the same tune was used for dissimilar texts as long as the metre was the same.

2 Traditional church singing, as described by some older informants, consisted of everyone enthusiastically singing their own version of the hymn at the same time. Each individual's performance was generally the same in every rendition, but different than the others. One person with a particularly clear voice (the forsungværi ) would sing the words slightly ahead of the small congregation since there was often not enough light in the room for everyone to read the text. In addition, an individual from time to time might sing approximately a perfect fifth above the rest. Hreinn's grandfather was the primary singer (forsungværi) in a church. He said that many people stopped going when the organs were introduced because they considered it "obscene" to sing the exact same thing as someone else at the same time.
Unfortunately, I do not know of any recordings of this style. Perhaps a somewhat similar example might be found on the Disques Ocora recording entitled "Musique Celtique Isles Hébrides" (OCR 45) on side B3: "Family Worship."

3 The only Icelandic folk music that used a somewhat fixed musical interval between voices was called tvísogvar, as it was generally sung by two people or as two lines. It appears to be a very old style, and was already mentioned in ca. 1325 in the "Laurentius Saga," where it was prohibited by the church in favor of plain chant. Although its origins are unknown (perhaps coming from French organum), the melodies were special to this form and passed down orally, and perhaps resembled somewhat the kvæa -melodies. These tunes also were not connected to the story but rather the verse metre. There is evidence that tvísongvar was practiced especially around the northern central cathedral school in Skáholt, where it seemed to exist the longest in its older form. Two performers would sing at first in unison and then one would sing up in the interval of a fifth; they would again join on the same pitch and then continue in fifths. Unlike European music which strictly forbids singing two or more fifths consecutively, the singers only used this one interval. To my knowledge, the singular recording stylistically similar to kvæaskapur in vocal style, ornamentation and register was made by the Helga Jóhannsdóttir for the Radio in 1965 in Skagafjörur of Pálmur and Ólafur Sveinsson.

4 There are other traditions that did not conceive of fixed pitch or scales. For example, in many tribes of American Indians solo singers do not sing regular intervals, but often keep one note the same usually related to the pitch of the drum. As in Iceland, the singers performed in settings that had little or no resonance (outdoors or in tents).



When Iceland was settled by Norwegians in 870, the spoken language was Old Norse, as it is now named. Several hundred years before it came from a fundamental Germanic language that latter evolved into Old English, Old High German, and Old Norse. Their poetries share many properties including the use rhythm and alliteration. The languages generally accentuated the first syllable of a word, so generating a rhythm through a group of identical consonances or vowels at the beginnings of word was natural. Poetry often was the means of transmitting history, legends, and ethics, and it was passed down by oral tradition. Rhythm, alliteration, and eventually rhyme were great assets for memorization of long poems.1
Old Norse existed from about 750-1200, and the written alphabet was "runic." However, committing something to writing was considered extremely sacred and few examples remain.
For some inexplicable reason, most of the oldest writings about the Norse tradition (historical and mythological) were from Iceland, generally written down in the 13th and 14th centuries. The collection of the most primary source is called the Edda,2 and it is a collection of poems from an extended oral tradition finally transcribed anonymously to parchment by various scribes. Each poem had constantly changing metres, but the rhythm as defined by alliteration was obvious. In the poem "Hávamál" the god Óthin is discovering the alphabet of the "runes" which possess the ultimate secret wisdom with special transmission through poetry. He hangs himself and wounds himself with a spear (sacrificing himself to himself for true knowledge). The poem reads:

Veit ec, at ec hecc vindgameii á
nætr allar nío,
geiri undar oc gefinn Óni,
siálf siáfom mér,
á theim meii, er mangi veit,
hvers hann af rótom renn.

Generally it has been thought that Eddic poetry could not be sung since the metre was changing continually. This notion was primarily conveyed by the prominent 19th century German scholar Andreaus Heusler who devoted his brilliant research to the Germanic languages and especially Old Norse. However, he never conceived of the idea of a variable melody which could adapt to fluid modulations of the verse-metres. In fact, it is quite probable that Old English poetry was sung in some manner, 3 and so it seems at least possible that the Eddic poems were transmitted through some form of vocalization rather than spoken. Modern Icelandic is very closely related to Old Norse, much more than any other Germanic language is similar to its original form, so perhaps kvæaskapur resembles a prototype of the oldest Germanic music. 4
"Skaldic" poetry originated by the 9th century before the settlement of Iceland and flourished for several centuries, and it was attributed to specific authors. There were rather strict regulations on metre and alliteration, but the subject matter was somewhat flexible. The poets were often highly revered and frequently written about in the sagas (the Icelandic prose literature from this period). The use of alliteration defining rhythm is obviously similar to rímur in this 12th century skaldic poem by Thóra Særeksson:

Ok gimsløngvir ganga
gífrs hlémána driífu
nausta blakks at næsta
Normanna gram Thori.

It is interesting that after the 11th century all of the poets in the courts of Scandinavia were Icelandic. This implies that there was a special performance practice for poetry, for if it were merely recited, why would the performer be imported from Iceland? It is noteworthy that until the 20th century, the only artists in Iceland that might be classified as "professional" were the kvæamenn .
The practice of kvæaskapur flourished throughout the country, but persisted mainly in the region of Breiafjörur. This area was one of the most stable during the famines and plagues (occurring especially in the 18th century) and few emigrated, so perhaps traditions could survive better here. However, even within this small region there were varieties of style. In the southern part, the music tended to be highly ornamented with flourishes on almost every note. In the north, there was far less ornamentation, and in Vestfirur there was almost none.5 There is also evidence that a similar style accompanied dance (dans ), but dancing was banned in throughout the country by the 19th century, so little is known. In the 17th century the term dans was synonymous with poetry.


1 See: Pope, John Collins. The Rhythm of Beowulf, Yale University Press, New Haven, Connecticut, 1966.

2 The Edda refered to here is sometimes called the Elder Edda, as there was another book called Edda written in the 13th century by Snorri Sturluson which served as a textbook for skalds and later for rímur poets. It is Snorri's Edda which Hreinn refers to later in the book.

3 Op. cit.

4 At Yale University from 1968 and 1970 I studied Old English with John Pope and Old Norse with Konstantin Reichardt. We discussed possible performance practices of the poetry, and it was relatively clear that Old English verses were chanted accompanied by a small harp (which kept a drone). There was little evidence of how Old Norse poetry was presented, so we were unable to draw conclusions. I had never studied Hreinn's findings, and when I met him and we discussed this issue, I found it quite ironic that he had spent years speculating on the same topic. I do not mean to imply that kvæaskapur is a direct link to Medieval Germanic poetic practices; however, to my knowledge this possibility has never carefully been pursued and might prove extremely interesting.

5 Compare the transcriptions of Gunnar Alexandersson (No. 14) from the southern area and Thóur Gubjartsson (No. 11) from Patriksfjördur in the north.



Once Hreinn had abandoned his attempt to find scales with precise pitches he needed another way to penetrate the music in order to find the underlying consistencies of this tradition. Why were certain melodies not suitable for kvæaskapur but only for singing? In fact what defined a "kvæa -melody" in some basic way since it was always changing? Hreinn told me of an occasion when Thórur Gubjartsson was being recorded by an Arabic producer for a disc of Icelandic folk music. After a short time the producer stopped Thórur somewhat angrily and said he was forgetting the melody and changing it constantly. Thórur was very polite, so he sang a melody exactly the same several times and then left. No traditional definition of "melody" quite applies to this style.
Hreinn had spent years transcribing the recordings over and over in musical notation, and gradually he began to see kvæa -melody as a succession of several fluid shapes rather than as lines of rigid points. Certain contours began to occur more frequently, almost always associated with certain verse metres. The contours could provide a somewhat flexible mold for the amorphous melodies.Verse and contour are inseparably connected, as rímur was always performed and never recited.
At this stage, Hreinn's analysis became more and more reductive. In this chapter he proposes that since large contours govern the macro-structure, then perhaps they in turn are generated by smaller micro-contours. For the most part, they are diminutive representations of the larger shapes. For example, each half-line of verse might have an individual shape, but when combined, they fashion the overall contour. It is in this domain of the micro-shapes that the persistently elusive kvæa -melody begins to form.
The inevitable problem arises that kvæaskapur was an organic living and evolving tradition with endless possibilities. Clear boundaries fluctuate, and any "theory" is but an attempt at better understanding and cannot ultimately provide a complete definition. Hreinn was meticulously mistrustful about his conclusions, and so in great detail he explains the progression of his research and where it led. He was completely aware of possible limitations in his theory of contours, but he felt it offered a glimmer of truth. The core contours provided the skeletal background of all of the melodies, and the microcosms define the flexible boundaries that help makes them recognizable.

Note: The concept of contours as essential to perceptual and emotional response in music is not a new notion. In music theory, Heinrich Schenker (1868-1935) felt that certain primary shapes govern all the structural levels of a composition in many ways similar to Hreinn's theory, and this was also true with the highly influential theorist Josef Schillinger (1895-1943). John Lomax (1867-1948), the legendary ethnomusicologist who specialized in folk music of the United States (especially cowboy music and early African-American blues styles) suggested similar ideas.
However, perhaps of even more interest, certain neuroscientists have described specific innate contours that seem to be genetically encoded, and when perceived by an individual, they are linked to particular emotions. Notably Dr. Manfred Clynes, neuropsychologist, engineer, and musician, has written extensively on this subject (see the bibliography).



Clearly kvæaskapur was a complex performance practice weaving together poetry, melody, voice, and presentation. Hreinn was eventually left with recordings of only nine performers whom he considered represented the tradition in relatively complete ways. There were many recordings with remnants of the older style, but only these nine had all of the qualities including the proper vocal timbre. Hreinn believed that kvæaskpur was extremely difficult art form only mastered by a few, and that if analyzed, the vocal style was not merely the strange quiverings of old people, but a deliberate skill practiced usually starting in childhood.
When we recorded Thórur, he was the last know person who could both kvæa and sing. His posture and demeanor were very different in each case. When he sang a hymn, his head tilted slightly downward with his eyes somewhat focused on the floor. He was slightly slumped over, and he appeared to be reminiscing. His pitches were quite precise and did not vary from verse to verse, and he sang with a much deeper voice. When he would kvæa , he sat much straighter with his neck erect and his chin tucked inward. His concentration appeared to be more internal, and he focused intensively on his breath. He frequently stopped between stanzas to catch his breath, so he could kvæa the entire next verse with only one very short pause. He sang the hymn quietly and did not try to project, but his kvæaskapur was astonishingly loud and effortless. He had always sung hymns in public, but he recalled that the first time he attempted kvæaskapur in public was when he was 14 although he had been practicing as long as he could remember.
Since this style evolved in conditions where the kvæamaur performed without resonance from the room often for many hours at a time continually, this environment influenced the vocal quality. Obviously, the voice must project well without straining the muscles.
Research about the sound quality of vowels discusses vowel "formants," or specific frequency regions which distinguish the sound of one vowel from another. Many art musics and folk musics alter the vowels sounds (formants) drastically in order to attain a specific timbre in the music. This is indeed the case with kvæaskapur where the sound of the vowels are so altered that understanding the meaning of many words becomes almost impossible. The sound quality is concentrated in the upper spectrum where small details of change become highly apparent. Vocal quality is as important as the poetry or melody, and mastery of this technique requires remarkable control.
The need to relax the vocal muscles for prolonged performances evolved in to a style that centered on small intervals. A common ornament (a vocal tremolo) rearticulates the same note very rapidly, a dillandi .1 This same fast throat articulation could also produce extremely fast ornaments with many pitches, and these ornaments are often tiny mirrors of the larger shapes. Many criticisms of kvæaskapur mention a strange "quivering" voice which probably refer to this style of ornamentation.
The state of mind of the performer and listener is not dissimilar to meditation on sounds found in practices such as in Zen Buddhism where often the concentration is focused on minute modulations and variations within the exhalation of the intoned breath.
The last criteria for the kvæa -voice, the most inexplicable, could possibly be labeled "strength" and no doubt is closely related to breath control. The kvæamaur was part of a Nordic tradition that valued heroism. It is well worth repeating the quote of Pétur at the end of this chapter -- "...if great power is put into will be braver."


1 Although this ornament is not usual in western classical music, it is found frequently in many folk music styles. For example, A.P.Carter (1891-1960), the American "hillbilly" singer used this ornament as frequently as any kvæamaur.



This chapter is by far the most complex. It seems inundated with minute detail generally explained by charts and mathematical formulas. However, underlying this intricate maze of thought is a strong romantic and metaphysical belief -- behind the chaos on the surface of kvæaskapur there is fundamental unity governing the art. This unity provides the foundation for its artistic and cultural identity. On another level, the kvæamaur, the performer, is at the core and allowed true individuality with a great freedom of possibilities within this environment of fluidly fixed limits.
Hreinn attempts to define the smallest elements (the individual pitches and rhythms) as variable points within vacillating boundaries. He came to the conclusion that the pitches sung were not arbitrary but could be placed within certain borders. A variable melody was not a sequence of infinite degradations of pitches, but rather a succession of only a few pitch areas. Abandoning traditional music notation, Hreinn developed a system using numbers to define these pitch areas. Since the seimur always remains the same, he represents it as 0, and motions above and below he notates for example as +1 or -1. Everything is a movement either away from the seimur or back to it, and the amount of fluctuation can vary a great deal. Ultimately there is a balance since every expansion eventually contracts to the unchanging seimur . 1
The formulas and premises Hreinn created were intended to explain what he had heard. He listened to the 39 examples and transcribed them hundreds of times and had them thoroughly memorized, so he was capable of cross-referencing the material instantaneously. Each performer was like a member of his family; he thought about them all of the time. It is quite remarkable how closely his predictions come to the actual statistics when analyzing of the material. For example, Table 2, by statistically analyzing about 100 transcriptions, defines the tonality of the melodic system in terms of a hierarchy of intervals. This corresponds extremely closely to his theory of the background rules governing the system.
Rhythm becomes the next area of microscopic investigation, and it is viewed very much like melody, i.e., constructed of a few small units that combine and expand to create larger and larger structures eventually governing the self-referential form. And like melody, the expansion is not arbitrary but bound by many interpenetrating restraints of verse metre, vocal quality, breath control, and consistency.
Consistency is defined not by repetition of rhythm and melody, but by a kind of fluid "improvisation" within the encompassing restraints facilitating the creativity in the kvæamaur . Since kvæaskaur was always variable in many ways, Hreinn finally focused on the invariants of its constant change. First he examined how much the melody could expand within a stanza (how there was a balance between the amount of change between lines). Then he compared the rate of change between stanzas. One of the most interesting conditions of the background rules is that the rate of change between stanzas must remain basically the same in a single performance. The kvæamaur might vary the second stanza considerably from the first, and then the amount of difference between each stanza will essentially be maintained. Conversely, the second stanza might be very closely related to the first with little variation, and the following stanzas will only change at about the same small rate. In a performance, the performer might alter the rate of variance considerably but only in a new section, and then the rate will be maintained throughout that part. This is what accounts of the lack of dramatic structure in the performances; since there are no drastically obviously perceptible changes the listener is never jolted into listening, but when paying attention, s/he is always provided with a plethora of mutable details. In the background remains a classic narrative story expressed by melodies within traditional contours and styles; in the middleground are the stemma , the melodies emerging that reflect the basic contours in inscrutable but consistent ways; and the surface is the creation of the performer rendering an incredibly intricate art form with her/his virtuosity.


1 Hreinn concluded that when a second tone was added to the fixed seimur, there were two interval-bands that when combined, could create a third. Essentially the background of the melody is comprised of these three regions, and variations could generally be considered expansions or contractions. Interestingly, Morton Feldman in his essay No. XXIV from "Anecdotes and Drawings" describes a very similar structure to his compositional method based on three intervals (originating from two) and then developing the music from this triadic core.



Behind all of the theories, formulas, and conjectures offered in this book is Hreinn's underlying certainty that kvæaskapur is a truly special art form. Kvæaskapur is a style involving an elaborate collaboration between poet and performer, each adhering to the complex demands of the tradition. The poet composed within the clearly defined boundaries of metre, rhyme, and alliteration, with freedoms not dissimilar to the kvædamadur's. Hreinn concentrates on defining the background poetic structure of rímur, its limits and flexibilities, and this bears much similarity to the logic governing the melodic forms. The poet writes for potential performers and attempts to establish all of the conditions that lend the verse to proper performance possibilities. One of the most important tools for the poet is the kenning (pl. kenningur), generally an expression that is a metaphor for a specific image. For example, a "horse of the sea" indicates a boat. These expressions are often used like formulas to set up clear alliterative and/or rhyme progressions to follow in the verse, and on another level they are aids to memorization. Generally there are several or many kenningur for one metaphor, and the poet uses them like ornaments -- the choice of kenningur sets the tone of the interpretation and the nuance of understanding. The poets, like the kvæamenn, are few in number, and it takes exceptional skill to attain perfection of the craft. The poetry was never meant to be recited but intended to provide the text for performance joined with a kvæamaur's bragur and special kvæa -voice.
The final result was the outcome of a multitude of interpenetrating conditions governed by metre, stemma, vocal quality, and aesthetic restrictions. This interpenetration of parameters creates a three dimensional counterpoint focusing on minute momentary variations rather than on a tune. Each layer has its own demands: the verse metre defines the possibilities within invariants of rhyme, alliteration, and rhythm; the melody evolves from the fluid boundaries of contours and specific variable interval progressions; the vocal quality outlines potential style and ornamentation; breath and strength govern the projection of the text and rhythm; and aesthetic concerns dictate uniformity in domains such as volume and timbre. The kvæamaur was not only an entertainer but a guardian of a cultural heritage.
When I was with Thórur and Gunnar, I felt I was in the presence of extremely special persons. Without pretense but with a clear focus on their performances, they immaculately captured the essence of centuries of their profound art form. They spoke mostly with older formal Icelandic grammar and vocabulary, and their facial expressions always exhibited great curiosity and self-reliance. Both had memorized enormous amounts of poetry written by themselves and others and quoted lines and stanzas regularly. When Thórur performed his own poetry, he always said that "no one wrote it." What amazed me was how their experiences were constantly transformed through composing and performing verses -- how completely inseparable they were from kvæaskapur . Their performances were beyond comprehension with their streams of complexity of details, but they conveyed something far deeper with their art than language can communicate. Undeniably, kvæaskapur was not random, quivering melodies from old people but a deliberate and highly practiced virtuosity that thrived for almost six centuries, if not much longer. Very likely, until modern times many poetic practices throughout the world were inseparable from performance and distinctly different than recitation. In Iceland the rich cultural heritage thriving within the isolated fertile environment for centuries fostered this development that was remarkable and highly unique.



Kvæaskapur relied heavily on the intricate, detailed improvisation of the kvæamaur. "Improvisation," by its very nature, focuses the attention of the listener and the performer on the momentary experience. Certainly, many styles of American jazz are also about exploring the spontaneous regions of the mind through sound. The general aesthetic environment with its cultural boundaries and artistic definitions, clearly and uniquely limits the scope. The performer is free to focus on details. Or, as Anton Webern said, "We want to say 'in a quite new way' what has 'been said before'".1 Hearing a story over and over was of no interest, but listening to the details of variations in the poet's verse and the kvæamaur's presentation was possible because in the background kvæaskapur was grounded in the historic narrative and the complex rules of the traditional art form.
This is not to say the the performer was in any way self-conscious about the "rules" governing the system. On the contrary, an improvisational performer concentrating on the restrictions is never convincing. The criteria defining the aesthetic judgments evolved in Iceland in a relatively consistent and isolated place for centuries, and in turn the inventiveness of the poet and kvæamaur affected the evolution. Gradually, foreign influences had greater impact leading to the demise of kvæaskapur in the same way the old architecture of the farms expired except in museums.
The famous singer Jón Lárusson (1873-1959) had a great influence on changing the public's perception of kvæaskapur. He came from a highly regarded lineage of poets and singers from the north, and he had a beautifully engaging voice and a marvelous sense of rhythmic ingenuity. The Radio often played his recordings during the revival of rímur, and he helped significantly in renewing the public interest in this style. However, he introduced new reforms that probably aided in making kvæaskapur more accessible to the public. He memorized melodies and sang them consistently with only minor variations, and he would often completely change the melody in the middle of a section (a practice that Thórur said was "an unfortunate blunder"). This became a paradigm for the new kvæamenn separated from the older style. Jón was much more "entertaining" in the traditional European sense, since his presentation was intense and dramatic and better suited to short periods of listening.
The complexity of the older kvæaskapur results from the multitude of simultaneous aspects: the metre, contour, bragur, rhythm, voice, dynamics, etc. Each can be analyzed separately, as Hreinn has done, and beneath each structure there are similarities in restrictions and flexibilities. Hreinn only once performed kvæaskapur for me. We were in a long line at the University cafeteria, so he performed very softly. It was clear he had spent endless hours practicing, but he knew he was not a true kvæamaur because he chose to analyze the art, thus compromising his spontaneity as a performer. In this situation his "individuality" was limited because he was self-consciously thinking of what was proper, and he knew this contradicted the inexplicable qualities of a kvæamaur. Historically, Icelanders have had great pride in self-reliance and individuality, and the performer necessarily represented these qualities.
The last recordings Hreinn and I did were special because Hreinn had years to study the previous documentations, and so his questions to the informants were highly focused on the issues he was examining; sadly, these were the final recordings of rímur as performed in previous times. Gunnar's penetrating blue eyes were mainly focused inwardly while he perpetually composed kvæaskapur, always glorifying his surroundings. He projected an almost heroic optimism by elegantly praising everything. Thóur,2 then 83, was gregarious and an amazing performer of many of his traditions. He was certainly one of the best kvæamenn , was a brilliant story teller, could sing, could interpret traditional hymns, could imitate almost anyone, played the role of Santa Claus every Christmas, and he would take care of your every need. When we arrived, Thórur had arranged for a house for us to stay in, a place to eat, a place to record, and he even brought sodas in an ice chest to the recordings so we would not be dehydrated. At first, when we arrived at his home, he was cautious in accepting us. He had met Hreinn before, but since I spoke English, he was skeptical because I might be from England. At that time Icelandic fisherman were almost at war with British fishermen for fishing too close to Iceland. After we were invited into his house, he could not be more gracious or entertaining. Perhaps, some threads remain of kvæaskapur . The last time I saw Thórur in 1974, he was helping to raise his granddaughters. He said that they were just beginning to kvæa .

At this time, there are few recorded examples available to the public that illustrate kvæaskapur. Currently the only obtainable CD is a selection from the Árni Magnússon Manuscript Institute ( "Raddir" ["Voices"] SMK 7). Of the examples representing kvæaskapur the only two performers discussed in Hreinn's study are Thórur Gubjartsson and Margrét Krisjánsdóttir. ( However, these are not the specific examples analyzed in this book.) Margrét is from the south part of Breiafjörur, and like Gunnar, ornaments almost every note, whereas Thórur, from the northern region, has far fewer inflections. These recordings are interesting because the other performers all embody some aspects of kvæaskapur, perhaps the voice or variable melody, but not the whole. When listening to Thórur and Margrét carefully, the qualities of true kvæamenn become clear.
The California Institute of Arts Library in Valencia, California U.S.A. has most of the recordings transcribed in this book as well as the texts accessible to the public. In addition, there are related articles, and earlier editions of this manuscript.

Unfortunately, Hreinn Steingrímsson did not live long enough to see the continuance of his research. The possibility of this art form being remembered is precarious. It is already so distant from modern concerns, and very few remnants remain. It was Hreinn's hope (and mine as well) that this book, representing a lifetime of passionate and brilliant research, will provide the seed of an even deeper understanding of kvæaskapur .

1 Anton Webern. The Path to the New Music.Wein 1960.

2 The celebrated Icelandic poet Jón Úr Vör grew up in Patriksfjördur with Thórur as his step-father. His book of poems Thorpi is about his experiences with Thórur and dedicated to him. Reykjavík 1956.