Dear David, Many thanks indeed for your reply to my e-mail about Campion's "It fell on a Summer's Day". Your idea of adding c' (d3) to the 3rd chord of my extract is an ingenious way of ironing out what happens before the E minor chord, but it still leaves us with the problem of what follows it.
There is a basic rule of harmonisation, which from your message you seem to understand, but which will need clarifying for others trying to follow the discussion: "Notes which do not move by step must be harmony notes." Moving by step is going up or down one note at a time, as you do when playing scales. If you have two notes further apart than that, for example ______a__ __c______ _________ _________ _________ _________ they must each belong to a chord. The harmony of the two chords doesn't have to be the same, for example ______a__ ______a__ __c___c__ __c___a__ __d___d__ __d___c__ __c______ or __c______ __a___a__ __a______ _________ ______a__ C major twice, or C major followed by G major, are both acceptable. Now, you can fill the gap between those two notes with what's called a "passing note", without that note having to belong to a separate chord: ______a__ ______a__ __c_d_c__ __c_d_a__ __d___d__ __d___c__ __c______ or __c______ __a___a__ __a______ _________ ______a__ The passing note f' (tab d2) is fine, even though it does not fit the harmony of the two chords either side of it. If a note doesn't move by step like that passing note, it has to be a harmony note and have its own chord. In other words, you can't have ______c__ __c_d____ __d___d__ ______a__ __a______ _________ (bad) because d2 doesn't move by step to c1. Instead you'd have to have a separate chord for d2, for example __________c__ __c___d______ __d___a___d__ ______a______ __a_______e__ _____________ (good) This means that you are absolutely right to say: "try putting middle c in the 3rd chord of the lute part. Then you have a descending line (d, c, b) that makes sense polyphonically", because those three notes move by step. Any one of them may be taken to be a passing note, and any clash with another voice part (e.g. C/B) is acceptable. Unfortunately this doesn't solve the problem of what happens after the E minor chord, since the b natural (tab c3) does not move by step to any note in the following chord (D major). You cannot add an a (tab e4) to the D major chord, or you'll have consecutive fifths. You could conceivably insert another c' (tab d3) as a passing note between the E minor chord and the D major chord, but even though this has the merit of some imitation with the voice, doing that is starting to get a long way from what Campion wrote in the first place, and we are still left with an ugly C/B clash at the end of the first bar. My solution of avoiding E minor in the lute part by changing c3 to c4 seems to be the neatest solution, with minimum editorial interference. After all, having a letter on the wrong line must be one of the most common mistakes in French tablature. Best wishes, Stewart McCoy. ----- Original Message ----- From: "David Rastall" <[EMAIL PROTECTED]> To: "Stewart McCoy" <[EMAIL PROTECTED]> Cc: "Lute Net" <[EMAIL PROTECTED]> Sent: Sunday, September 21, 2003 2:59 AM Subject: Re: It fell on a summer's day > On Friday, September 19, 2003, at 05:34 PM, Stewart McCoy wrote: > > > Dear All, > > > > It fell on a summer's day is a well-known song by Thomas Campion - > > naughty but nice - published as No. VIII in Philip Rosseter's _A > > Booke of Ayres_ (London, 1601). It presents a problem towards the > > end. This is how the voice part ends: > > > > > > _h__e__f_______e__e_f_h_f_c_____a___ > > _____________|________________|____|| > > _____________|________________|____|| > > _____________|________________|____|| > > _____________|________________|____|| > > _____________|________________|____|| > > > > > > and the lute tablature has: > > > > _c__a__a_______a________________a___ > > _e__a__c_____|____e_c_e_f___e_|_a__|| > > _f________c__|____f_______d___|_c__|| > > ____c________|________________|_c__|| > > _c_____a__e__|____c___________|_c__|| > > _____________|________________|_a__|| > > > > > > It has always troubled me that the first bar of my extract ends with > > a chord of E minor on the lute, while the singer is singing c". I > > might be tempted to change the b natural (tab c3) to middle c'(d3), > > to have a chord of C major (1st inversion) instead, yet I can't do > > that, because the lute chord rings on into the next bar, where the > > singer has b' natural. I don't think B's and C's clashing against > > each other is an option, so something needs to be done. I propose > > changing the c3 at the end of the first bar to c4, to produce a > > chord of e (e5) and g (c4). These two notes are common to C major > > and E minor, so will sound well with the singer's part. I know of no > > other sources of this piece to offer inspiration, but having a > > tablature letter on the wrong line is a common and plausible error. > > > > In a word, I propose the following for the lute: > > > > > > _c__a__a_______a________________a___ > > _e__a__c_____|____e_c_e_f___e_|_a__|| > > _f___________|____f_______d___|_c__|| > > ____c_____c__|________________|_c__|| > > _c_____a__e__|____c___________|_c__|| > > _____________|________________|_a__|| > > > > What do people think? > > I'm not sure what you're hearing: are you saying that the e and the b > in the lute part are the root and 5th of an Em chord, and the c in the > vocal part is an added discordant note? Harmonically speaking, e, b > and c are more likely to function as a C Major 7th chord. But in 1601 > we're not quite there yet. So, if you're listening for chord > progressions, then "B's and C's clashing against each other" are going > to sound a wee bit heretical for 1601. I think that at the point > you're referring to in the music, the lute part isn't supposed to make > sense as a chord progression. The b and c seem to me to be a point of > intersection between voices that are moving, not static chords. > > I think if you want to hear chords, that's what you'll hear; but > that's no reason to change the music. If you have to change it, try > putting middle c in the 3rd chord of the lute part. Then you have a > descending line (d, c, b) that makes sense polyphonically. > > Regards, > > David Rastall
