The Process (Part 1)
Robert Calvert on Songwriting tips.
Composing, arranging, compiling, recording, editing, designing and promoting an album of songs - it's a complex process, and few people have covered each aspect in such depth as solo performer and sometime Hawkwind singer/songwriter Robert Calvert. In the first of this occasional series of articles covering the whole writing, demo-ing and performing chain, he shows you how it's done.
In interviews with people who write songs for a living, one of the questions most often asked is: 'Where do you get your ideas?'. It's a hard one. Who knows where ideas come from? Perhaps they come from a combination of everything one has ever experienced. More often than not, they don't come at all. But if you have deadlines to meet, or ambitions to fulfil, you have to have ideas. Fortunately, there are one or two techniques that can be used to bring ideas out into the open. It just isn't any good waiting for them to arrive.
The poet W.H. Auden once said (and I'm quoting from memory here), 'I don't know what I think until I see what I say'. This is an approach I often use in songwriting. I'm not saying it's the best way, or the only way - it's just the method I use myself when inspiration is not in evidence. Of course, it's wonderful when inspiration arrives and takes over; but you can't rely on it.
The way I go about it is to go into my little home studio at a set time every day and write. It's the same way I write plays or prose. When I'm building an album project I work the same way as any other professional author - continuously, experimentally, and highly critically.
To Produce an album of, say, six songs, as I've recently done (a mini-album) took several months of constant writing; rejecting a vast amount of unwanted material. This material, though, may surface again in another form, so none of it was wasted time. The main thing is to write, in the same way that a virtuoso musician has to practise. Too many people regard songwriting as a divine gift that needs no development. It's not just a gift - a large part of it is technique, like any other art.
The two main requirements for a home studio are an instrument of some kind, and a tape recorder. This latter need not be a portastudio, by the way; a portastudio is perhaps more of a producer's toy. Multitracking, bouncing and mixing down to stereo can be an unnecessary distraction for a songwriter - besides which, waiting around for a cassette to rewind in order to do a re-take can be very frustrating. Cassettes are also nearly impossible to edit; you can't reverse them to get special effects, and you can't make them into loops.
Personally, I recommend a good reel to reel machine with sound-on-sound facilities. The ideal is a Revox A-77, and you can pick one up quite cheaply these days, as everyone seems to be going overboard for cassette multitrackers. The Exchange & Mart is a good source of second-hand A-77s, and you'll need a budget of around £250-£300. If this seems more than you can afford, then £120 or thereabouts will buy you a Revox G-36. This is an earlier model, with a built-in valve amplifier. It has one or two disadvantages: it's not so easy to do editing on, and all the connections are phono - placed on the back of the machine. The advantage (apart from its lower price) is that you don't need an external amplifier. But personally, I'd go for the A-77. This is a fabulous machine that will never let you down, hardly needs any maintenance and produces high quality recordings, with six or seven overdubs (or even more, with practice) being quite possible.
You can, of course, get by with any recording device - a dictaphone would do - but you will need some kind of musical instrument, even if you're not a player. To be a songwriter you don't have to be a skilled musician; but you do need to be able to indicate some sort of backing for your songs. A good song, in fact, should be an inextricable combination of instrumentation and voice (unless, of course, it's an unaccompanied song).
Any of the Casio range or keyboards is useful. Even the VL-Tone, which is monophonic, can be quite effective with overdubbing. A guitar of some sort (preferably electric, as it can be directly injected into the tape recorder) will also help. A drum machine, too, is an invaluable asset - the Casios have a built-in rudimentary rhythm section, but a programmable drum machine is more desirable. It helps to set the whole tone and atmosphere of a song, providing a context in which to create something. But if you can't afford a programmable drum machine, then even a pair of drumsticks and a couple of microphones - together with cardboard boxes, tins and trays! - can be used quite effectively to create a rhythm track to build on.
It's often best to put the rhythm track down first. Simplicity is the key here - just the basic beats; you can add decoration later. If you're writing a rock song then it's a good idea to start with a riff: a repeated pattern of notes or chords. A good riff is a combination of notes with a drum pattern, the notes and the drum beats forming one musical whole that can't be separated. Listen to the riffs in your favourite songs, and see if you can work out how they're put together. To cite one example, the drums, bass and guitar in a song like Satisfaction work together like components in a piece of clockwork. It's not a case of one part being played on top of another - the parts complement each other, combining in a harmonic and rhythmic whole.
A riff can be arrived at by simply trying out repetitions of patterns until you find something arresting. Once you're got it - record it immediately. Don't let it escape; it's a rare beast, this!
It's quite possible to spend an entire work session trying out different combinations of drums and chords, only to find that you end up with nothing usable. Don't be discouraged. Regard it merely as practice for the moment when inspiration strikes and takes over the proceedings. And don't forget: once you've composed a magic and earth-moving riff, make notes of how you played it, in as much detail as possible - like a recipe, for re-creating the effect at a later date.
It's possible to construct a whole song around one striking riff, but changes are desirable to maintain interest. Again, it depends on the sort of song you're writing - it's a good idea to establish some feeling of the song's subject matter once you've got the basic riff to build on. As an example, I once wrote a song called Ejection, about a jet fighter pilot (featured on Robert's 'Captain Lockheed and the Star-fighters' album - Ed.). I tried taking the riff through various changes - into a different key for the chorus, and so forth - but I found that this distracted from the headlong forward motion required for the song's climax, when the pilot would eject into explosive sound effects. Therefore I maintained the riff without changes, other than those of emphasis, throughout three verses and choruses.
In another case, a song called Over My Head (which explores the difference between intellectual conversation and rock 'n' roll music), the riff itself goes through three changes of key before the vocal begins, and continues to change throughout the song while still retaining the character of the riff's original 'punkishness'.
Too much is made of chord changes for their own sake. A change should only be used to create a specific effect, not as a demonstration of musicianship - and the fewer changes the better is a good rule for rock music. Hugh Cornwall of the Stranglers once pointed out that Summertime Blues has three chords, and Tie A Yellow Ribbon Round The Old Oak Tree has forty. I leave it to you to decide which is preferable!
Once you've decided the direction your song's going in, you can try recording the riff and its sequence of changes. This should be a pattern which is interesting in itself - it's the bare bones of the song, but should have some life of its own. You can regard the whole sequence as just an extended riff, however many changes it goes through, and these changes should form a pattern with its own logic. Whatever the direction in which you take the basic idea, you should always keep in sight of the starting point, and keep referring back to it. Even if you go into an entirely different rhythm and chord (or note) pattern altogether for, say, the middle eight, it should always have some connection with the theme of the song. It should never be there for its own sake, or for the sake of dazzling the listener. It might be that you want to convey a different viewpoint of the song's subject from the one you started out with; or a sudden change of mood. A middle eight is often best used as the kernel of the song, where the song's whole meaning and intention is made clear in a direct statement of the lyrics; also as a bridge into an instrumental passage.
Anything is possible; but the music and words must work together to convey something. A song has to have some sort of coherence and reason for being, even if the subject matter is entirely surrealistic; otherwise it is not a song at all, and is only worthy of being erased to make room on the tape for something else.
I know this isn't a very fashionable view - as anyone who listens to daytime radio will attest - but it's a sound principle upon which songwriters throughout history have built their careers and their reputations. The current practitioners of what I regard as ersatz music obviously have very little faith in a future of any kind. Their art is more in line with advertising jingle writing... which is another subject altogether!
And the next stage of The Process? Watch this space to find out...!
This is the only part of this series active so far.
Feature by Robert Calvert
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!