Slurred eighth notes

I tried to send this via email yesterday, with no luck, apparently.

Are there any harpsichord related sources that tell us how to play eighth notes slurred by pairs, as one sees for instance in courantes by d’Anglebert or François Couperin? Equal, unequal, “reversed” unequal (which is how one often hears them)? I seem to recall that Gustav Leonhardt, in an interview on the radio, said we simply didn’t know.


There are several, though not all of them are harpsichord treatises and the topic is rather wide-ranging. One thing to look to is Hotteterre’s Principes and compare this with Saint-Lambert. I shall tell you why. Both indicate that, in conjunct motion, pairing notes was a natural thing in all walks of musical life. Hotteterre uses phonetics to demonstrate: tu-ru, tu-ru, tu-ru, pairing across the beat; Saint Lambert does it by placing fingers 3 & 4 (rh ascending) on the same lateral plain, indicating the pairing. Jean-Pierre Freillon-Poncein (1700) is also interesting:

In all types of bars, to make the melody more agreeable to the ear, when there are only four crotchets or four quavers, they should be played tu tu ru tu and ru on the following note; when the tempo is very fast, it is necessary to play them tu ru tu ru, with tu on the notes that follow them.

Both Hotteterre and Freillon-Poncein’s treatises imply a phrasing across beat and bar line. In terms of keyboard music, such an approach would be best achieved through fingering. Thus, in ascending passages, rh: 1-2, 3-4, 3-4, 3-4; lh: 4-3, 2-1, etc.; in descending passages: rh: 5-4, 3-2, 3-2, etc.; lh: 1-2, 3-4, 3-4, etc.

Slurs are different since they can indicate a variety of techniques. In some instances, they are not slurs but liaisons and indicate a slight overlap between the two notes. We see this in particular in connexion with the tierce coulé and port de voix, though Marchand, Clérambault et al. also use the effect.

Otherwise, it might indicate even notes when notes inégales would be normal or, as you say, it could mean a reversed inégalité, whereby a Lombardy rhythm is played.

The problem is that no one knows for certain and the approach is down to that arbiter of taste, le bon goût. I suspect it also has something to do with the nature of the music. I would be wary of using using a Lombardy rhythm in som fast music since it would be inelegant, but in courantes and sarabandes, it can be quite an arresting effect.

There’s a lot of information out there and all the treatises I mentioned can be found at Gallica, which is run by the Bibliothèque nationale de France.

Very interesting your question Dennis.

There is ONE thing we know from Hotteterre and others, thanks to the “tu-ru” and “tu” doctrine. This was partially explained decades ago in a famous booklet by Lasocki, and has been fully explored and explained in my “Playing the Baroque Harpsichord” book, Chapter 4.NOTES INÉGALES, section 4.4.SLURS IMPLY ÉGALES.

There are different evidences, both in Italian and in French composers, that often slurred pairs were played not égales but in Lombard rhythm. Leonhardt was correct: unlike inégales, with slurred pairs there is no direct evidence about when they were played égales and when Lombard. What we do know is that some 18th century composers, when they wished, say, a slurred pair of quavers to be Lombard, notated it as an appoggiatura tied to a crotchet.

Edit: needless to say, the same principles apply whether the notes slurred by pairs are quavers/fourths or semiquavers/eighths. As for crotchets, although in rare cases they are inégales, I know of no evidence that they can be Lombard if slurred.

Thanks, Jon & Claudio. Both of you mention Hotteterre and his “tu-ru”
articulations. But when it comes to notes slurred two by two, all he
says, as far as I can see, is that each pair is played with one “tu”
(“deux … notes passées d’un même coup de langue”). I don’t see how you
can draw any conclusions from this regarding the duration of the two
notes. And even if you could, there’s no way of knowing if this would
also apply to the harpsichord. Which is why I specified “harpsichord
related sources”.

You are absolutely correct, Dennis. In Hotteterre (and in no harpsichord sources either known to me) there is no suggestion to play slurred pairs Lombard. This might be seen as an indirect indication that Lombard in France was infrequent, but there is strong contrary evidence in both the use of appoggiaturas and the use of reverse-pointées (F. Couperin, Rameau).

Then you should look at Saint-Lambert. He doesn’t mention the tu-ru thing but if you look at the fingering in his scales, you will see that he places fingers 3-and 4 laterally, implying the pairing of notes. Such fingering produces a natural lilt that is probably more a more accurate means of playing notes inégales than the trochaic approach adopted these days.

The problem is that, apart from François Couperin, there are no other harpsichord treatises. There are instructions provided in some organ books but this shouldn’t put you off. Hotteterre is a perfect example since there is no reason why a musical language would change because of the medlia used for its performance. Especially in France.

I am always wary of François Couperin since he really only discusses his own music (and a little neurotically for my taste). But, as I mentioned, you can do whatever taste allows. One of the things I have noticed after studying ornament tables for more years than I care to remember is that everyone contradicts each other. Boyvin (1690), for example, has his own cadence and insists that pincements are always prefixed with a port de voix. He also stipulates an on-beat performance. Saint-Lambert says they are played before and are prefixed with a doublement de gosier. And so it goes on.

Another problem is in differentiating between large note and written-out ornaments and symbols. Charles Babel’s 1704 manuscript (GB-Lb Add Ms 39569) uses ‘slurs’ to indicate a descending tierce coulé in his copy of the Hardel gavotte, whereas Louis Couperin uses a dotted-quaver–semiquaver figure, and so forth. Every composer had their own ideas, which is one of the things that makes grand siècle French music such a hoot!

Anything elegant is not wrong.

Hi Jon. I have read Saint-Lambert (Principes du Clavecin 1702) and know it backwards, of course. And he is notorious among scholars for stating a few things that go counter every other contemporary writer (pre-beat ornaments) and also counter evidence: he categorically states that the two manuals of a double harpsichord are absolutely identical to all purposes: it looks as he never played one and did not know that the French lower manual around the year 1700 had 2 stops and a coupler! His name is also absent from any list of contemporary harpsichordists in France.

As for his fingerings, Rebecca Harris-Warrick (English translator) observed that his scale fingerings coincide with F. Couperin’s. Both, of course, favour the proper articulation and rhythm of inégales. Saint-Lambert’s two fingered pieces do not give us a clue about Lombard, unfortunately. The Gavotte has two quaver pairs slurred but, again, no clue as to whether they are égales or Lombard. The clues (but again with no clear “do or don’t” rule) are found in other composers, however, as I commented before.

Hi Claudio!

There’s a lot to unpack here but I have to make things brief. I think you are doing Saint-Lambert a slight disservice here since he was obviously in the loop: his books were each published by Christophe Ballard, who was a canny businessman and not likely to take a risk publishing someone without a good reputation. Indeed, he also wrote a laudatory poem to Louis Marchand, which accompanies his second book of 1702. That we know nothing of him is of no importance. F-Pn , z/1h/657 is a poll tax document from 1695 that lists most of the known harpsichord teachers and organists in Paris and while his name does not appear on any fascicle, there are scores of others whom we do not know. So, just because Saint-Lambert does not appear in any list of known harpsichord players does not mean he wasn’t known or respected.

His fingerings do correspond with François Couperrin, which should come as no surprise since he was brought up in an era that proscribed the approach. Of course, Couperin also demonstrated the new style, if I remember correctly. However, the lack of explanation for the ‘slurs’ might be addressed by his explanation of the liaison. While all is vague, we have to look at what Principes was: a guide for beginniers. There are many subtleties that go unanswered since these were things that mature players would acquire as their understanding of le goût developed. This is why ornament tables are so facile: playing a port de voix before the beat, for example, is a terribly difficult thing to do properly and subject to myriad variations, according to the effect one requires. The simplest explanation (including D’Anglebert’s very fussy attempt) was all that was required. For example, one wouldn’t expect a player such as Lebègue to take any notice of an explication just as one wouldn’t expect him to take any notice of the sort of things Couperin advocated.

I will stick to my guns by saying that ornaments (including slurs) meant different things to different people and, as they say, for today’s performer anything goes so long as it is tasteful.

Hi Jon,

As I already have mentioned numerous times, this music has to be felt. That goes for the ornaments and all the rest. The ornaments should sound improvised, and form an integral part of the discourse. As an old American country preacher once said, “It’s more felt than telt”.


Needless to say, guess we all agree with Frank Mento. In absence of specific advise as to where to use Lombard in slurred pairs, informed feeling is the best. The examples by Rameau I have quoted in my books (and well known anyway) provide good hints also. Other examples are the few cases where a composer wrote a passage with appoggiatura-plus-crotchet and later an identical one with a slurred quaver pair, obviously meaning the same thing and in both cases to be played Lombard. These passages provide a clue (alas, just a clue!) as to when harpsichordists felt it proper to play Lombard.