Henry Cow was a purely instrumental group at the beginning; when and why did you decide to use lyrics?

Well, there were already songs soon after I joined. We did 2 of Fred's (with him singing) on the second John Peel show in, I think, 1972. Then there was Tim's 9 Funerals on the Leg End LP. At this time Fred and Tim both wrote their own texts. Otherwise, it's true we didn't see songs as our strength, and they soon disappeared from our repertoire. None of us was primarily a singer, though Fred, Tim, John and Geoff all went on to sing in later projects of their own. So it wasn't until we started to work with Slapphappy in 1974 that we began to think again about the song form (especially Fred and I - a mutual interest that led eventually to the Art Bears). It was in the middle of the Slapphappy/Henry Cow period that Fred, who had not written anything for a while, gave me a tape of a piece he thought could be a song and asked if I'd write a text for it. This was the first time I had done anything like that. This piece eventually became Beautiful as the Moon. And of course, once Henry Cow asked Dagmar to join the group permanently, songs inevitably crept back onto our agenda.

There was never any discussion of a text before it was written. Discussion of texts only became an issue before we left to record what was planned to be the fifth Henry Cow LP - and eventually became Art Bears' HOPES AND FEARS. For this we were supposed to record Tim's mammoth composition cheerfully then known as Erk Gah (it was eventually recorded 18 years later as hold to the zero burn). Fred and Lindsay were deeply unhappy with Tim's texts. So unhappy in fact that they insisted we shouldn't use them. They called a group meeting a week before we were due to leave for the studio in Switzerland, which ended in my being asked to write new words for the whole piece. Of course this was impossible. And, from my point of view at the time, not really desirable. Instead, I wrote a number of shorter texts and suggested we make an LP of songs. Most of the music and further texts were written en route to - and in - the studio. But, when we got back to London, this time it was Tim who called the meeting, to say that the result wasn't really 'Henry Cow' - that musically, it wasn't the way the group should go. The meeting ended with the group deciding - slowly - to disband. Two long pieces by Tim and Lindsay were kept for the next 'Cow' record and the rest the group agreed Fred and I could release (since in fact more or less all the compositions were ours). We recorded 3 more songs to make up the time and the Art Bears were born.

Two other lyrics I remember engendering discussion were Joan (on Hopes and Fears) which the feminist faction of the group was very unhappy about - and were only eventually accepted because Dagmar strongly defended them - and Albion Awake! (on The World As It Is Today, to which Dagmar strongly objected because they were too violent. Which is why the piece only appears on the record in instrumental form.

What function do you think lyrics have in rock music in general and - more specifically- in Henry Cow-Art Bears music?

Function ? My own texts are the only ones I can answer for. And there is no single answer. For me, different texts have different functions: narrative, atmosphere, affect, theatre - but they are all, in the first place, literary functions, related to storytelling and drama. I write to speak, and verbal precision is important to me, as is meaning - though I do accept that nonsense texts are not so unusual in rock and are perfectly valid, having musical and gestural rather than semantic value. One could equally cite scat singing in Jazz, where the voice suppresses its aspect of language-giver altogether and concentrates purely on it's attributes as singing instrument. Rock, Punk, Metal and Rap texts may also be used primarily to communicate attitude and swagger, where words bear an iconic rather than a coherent or literal meaning.

Broadly, however, I think of song as a species of speech, generally with a non instrumental purpose (I mean instrumental in sense of being used to get something done).

My own text are first and always concerned with content - and that means also, form. So that, on The World As it is Today I tried to write directly unambiguous political words (my own task to avoid copying Brecht) while on the new Science Group CD the texts are somewhat technical and deliberately ambiguous and opaque. For Domestic Stories I made a kind of an argument, in the form of a cycle tightly related to three mythical women; A Face We All Know is a narrative (psychodrama), Hopes and Fears is all in the first person while Winter Songs is wholly abstracted and descriptive. In fact each record essays a new form or sets itself some new problem. Also, when writing I try to think very concretely of the singer and the grain of their voice. I wrote very much for the sound of Dagmar's or Christoph's delivery, and of their accents in English, and their affective geste.

Sometimes I hear a text as a small play or drama, requiring a singer to act - to adopt a role (Dagmar was good at that) - and sometimes texts need to be neutral and unemotional, to manipulate a singer through their internal complexity rather than their overt content (for instance the singer may have to find a way to deliver a text that does not close it's interpretative possibilities by expressing one of them). In general I have to say I write less and less for expressive singing. If the emotion is in the text there is no need to underline it. But that is not a rule; different songs call for different strategies. In a case like I tried to reach you, emotion is absolutely demanded - but an emotion that goes against the emotion of the text (which otherwise would be trivial). I think I write texts that need to be in a state of tension with their settings - and ideally, in tension with themselves in order to work. And I like a text which, when read on the page seems un-singable. Such texts make life more interesting for composers, I hope, since they appear structurally as problems while still being clear as to content.

Did your lyrics come first, then music, or did you write words on the basis of a melody?

At first, for me, the music came first, but after that it was always the text. Composers seem to prefer it that way. A text gives them plenty of limitations to work with, (especially my texts I think, which have a lot of internal structure). I guess it's easier to be second to work - the ball is already in play.

What are your literary roots and influences?

The usual canon - from Greek mythology and the Bible through William Blake, WB Yeats and TS Eliot. I was particularly inspired by the work of Gerard Manley Hopkins and Lawrence Ferlenghetti. I stole shamelessly from Shakespeare (all those final rhyming couplets). Of songwriters I would have to cite Dylan of course, though I never tried to emulate him (he was a singer-songwriter while I was writing texts on the page). But I loved the way Dylan would veer into uncloseable linguistic forms ('Someone called for an ambulance and one was sent; somebody got lucky, but it was an accident). I think unconsciously I took a lot from Alfred Jarry too (describing commonplace things in unexpected or transfigured ways). On News from Babel particularly but in many other places too, I return obliquely and often to the ideas of critic and writer George Steiner.

I noticed you avoid the term 'lyrics' and you seem to prefer 'text'; may I ask you the reason of your choice? What's the difference, in general, between the two terms in anglo-american rock culture (historically, and in their ordinary use)? There must be some connotation I'm not able to catch.

That's an interesting question - I suppose I think in part about priority; when the music is written first, or at the same time as the text, I instinctively think of the words as lyrics. However, I usually write words intended to be sung, and think of them as song texts in the absence of any music. So, to me they are texts, which will later be set. I can't say if Anglo-American rock culture would follow me in this. And of course, a listener can't know by listening whether the music or the text came first. It would be a question then, whether there is some experiential difference that distinguishes 'text' from 'lyric' in this case. I suspect there is. It would be a structural question of the way the music and the words relate to one another. And also, I suppose, one can say a good lyric may read badly, but a good text always reads well - even if, when set poorly, it does not sing.

When I asked what function lyrics (texts) have in rock music, my question was in the spirit of Simon Frith's: "Why do songs have words?". Possible answers: rock has a text just because there's a voice and the voice must sing 'something' (anything); rock has a text because words 'explain' or 'clarify' or 'illustrate' the sense of music, its 'message' etc. etc.

Hmm. I don't really identify with either position. And a critical orientation - to do with the text as text - is entirely absent. I feel happier with the Wittgenstinian idea of a family of associated forms than with a clearly definable category. Rock songs belong in general to the family 'songs'. It's a family where the words may occupy all stations between being of primary importance (as a kind of storytelling, where the music just adds colour or emotional support) and of being merely incidental (as is arguably the case in a fair amount of commercial music). In general, however, I think words have meanings and songs carry meanings and singers are expressing meanings, even where the meaning - as is so common at present - is essentially an inchoate expression of 'attitude'. I doubt if even these words explain, clarify or illustrate the music. Songs are more hybrids - things that exist in two worlds at once - than conspiracies where one aspect backs the other up.

You say your texts have 'a lot of complex internal structure'. What kind of structure? How do you organise them, formally?

I simply mean that they have strong internal rhythms (determined by alliteration and, more often, vowel sounds) that cut across the formal stresses and syllabic structure. Also that meanings emerge from omissions and suggestion and association that are not on the surface of the text. I am always thinking of at least two things at once when I write. It helps me to focus if I have continually to satisfy two imperatives, one overt and one covert. In a way the covert is a control for the overt. But yes, 'a lot of complex internal structure' is rather overstating something fairly commonplace.

Could you tell me more about 'Winter Songs' and 'Hopes and Fears' genesis? (sources, concept, composition, etc.)

Hopes was mostly written on the run, Henry Cow had an LP to record and no material - having decided at the last minute not to do what we had planned to do (see Q1). So I proposed a song record as the only thing that I could realistically help produce at such short notice - but also, I admit, because it was a long held ambition of mine. What's more, I did think it would be an interesting and unlikely project for the - at that juncture - symphonic-epic Henry Cow. I had no guiding idea of the overall content of the LP until I had to write the extra songs with Fred to complete the record. Then I looked for a common thread and reinforced it with the final texts (the thread was labyrinths and first person affective voices). Though much of the work was done in a single burst, there were also bits and pieces gathered from previous work; it was not a completely integrated process. The following year, when Fred and I came to make a real Art Bears LP from scratch, I determined to start with a clear idea from the outset, and ended up basing the texts on the carvings from the stylobate of Amiens Cathedral (and a few related carvings from the same period and culture elsewhere). Firstly, because medieval culture and the iconic way of seeing the world is so estranged from ours. Secondly, because I have a strong feeling for that era's symbolic language, and for allegorical and metaphorical explication. Thirdly, because the carvings are frozen moments and their dynamic and tension and meaning has to spring from the precision and ambiguity of the moments chosen: nothing is explained - but dynamic relationships are invoked - it is precisely the muteness in these carvings that makes them speak. And fourthly, because it imposed a discipline on me, a constraint, a problem to solve, and produced ideas by closing off possibilities.

And of course the pictures could be printed and thus add an extra dimension of signification to be worked against (as opposed to being reinforced). I wanted to work with the idea that some essential link was missing. When I sent the completed texts to Fred he set them all in a single bout of work. Then we met at the Studio, Fred played what he had (it was the first time Dagmar or I had it) and we began immediately to record. Most of Hopes, as a Henry Cow LP, had been recorded in a conventional way: we learned the songs and played them live, all together, keeping, and maybe adding to the best versions. When it came to making the extra songs, Fred and I decided to adopt a different procedure, which we retained for the making of all the other Art Bears records: first we put down a click track, then the key chord structure, then the vocal melody as a guide for Dagmar. She would take this away to another room and learn it with the words, while Fred and I continued to build up the track, adding only what we thought was essential, one thing at a time. We had one rule - also deriving, negatively, from our experiences in Henry Cow: no discussions. If someone had an idea, we would immediately put it to tape. Then we'd listen and decide whether to keep, abandon or modify it. This made for very fast and easy work. The voice we added as early as possible, so that the rest of the music could work around it. The drums were usually put on last. Also we got the sounds we wanted at the recording rather the mixing stage, thanks to the extraordinary abilities and sensibilities of our engineer Etienne Conod. Finally, when we had to make musical and sonic decisions we worked a lot from the pictures, developing a kind of visual metaphorical shorthand to describe effects we wanted to achieve (pertinent to this, there is an interesting analysis of 'First things First' in an unpublished thesis, and I recall giving some talks a long time ago about this work-method with concrete examples). The entire album represented a single burst of highly concentrated energy (we hardly slept during it's making) taking 12 days from first piano run throughs to the boxing of the final mixes.

As for sources:
Hopes was made in a piecemeal way and only took shape as I began to discern the thread that led to the next text. I researched the history and mythology of mazes and labyrinths - and curiously the way I followed the trail of the texts as they arose was not dissimilar to finding a way through a labyrinth. More or less all the songs being in the first person, expressing individual 'hopes and fears' formed my second 'subject'. Research for this was hardly necessary, though I adopted the kind of subjective projection technique (not to 'express myself - which I think I find a rather meaningless idea - but to express some 'other') which any writer of fiction or drama must adopt. The labyrinths were variously internal, urban and mythological, another 'family'.

For Winter Songs, I took the idea of writing seemingly plain descriptions of medieval carvings, mostly taken from the stylobate of Amiens Cathedral which, while remaining faithful to the carvings and to the iconography that informed them, also expressed some aspect of the loss of the world they portrayed and which portrayed them - and the dislocations endemic to contemporary life. Primary sources were an unpublished thesis by Francesc X Puerto, numerous books on symbolic iconography, Fulcanelli's Mystere des Cathedrals, Treatises on the Labours of the Months, Chambers' The Mediaeval Stage and R.D. Laing's Sanity, Madness and the Family.