An article from the English Lute Society News that EVERY continuo player should read!!


Plucked Continuo: Some Personal Thoughts on the Current Art of Basso Continuo Accompaniment by Nigel North


What I would like to do here is not to present a paper, or propound a musicological thesis, but to discuss in a general way continuo playing on lute-family instruments. Most of all I want to talk about a few general questions which I have been thinking about over the last 30 years; I want to ask questions rather present answers.

The first thing I would like to say is that I feel that there is no single, right way of playing continuo. About 20 years ago, on one of my first visits to America to teach at the LSA summer school there was one student from the southern states who had been playing banjo all week. He was very enthusiastic; all he wanted to do was play Bach on the banjo (I can sympathise with this as I am myself guilty of making Bach transcriptions . . . ). He was an excellent musician, and I remember that his arrange­ments of Bach cello suites for the banjo were the best thing in the concluding student concert; I especially remember a slow movement from one of the violin sonatas. The music was not as Bach would have heard it, and perhaps even some of the Tealised harmonies were not as the composer would have intended, but the moral is that in the end it doesn't matter about being historically authentic-as long as we make music.

A second general point is that I have myself been guilty over rhe last 20 or 30 years of the some of the practices which I would like to question today. It is for this reason that some recordings I refer to are my own-not to publicise them, but to invite criticism of my own work rather than the work of others!

Talking about continuo playing is difficult because it is some­thing which just has to be experienced live. Its essence is not something that can be put in books, or learnt from books-I have written a book on continuo playing on the lute, but I never use it myself!

There are three general questions I have noticed in listening to music with basso continuo over the years. The first of these is the question of balance; how many instruments should we have in a continuo group; what combinations work? Secondly, should we have lutes in everything? And thirdly, have we gone too far in the use of lutes for continuo, or perhaps not far enough?

Before looking into these three areas, I would like to discuss an analogy which I have found to be very useful.


Language and harmonic accompaniment: a comparison

Some of you may have read a book published about 20 years ago, by Nicholas Harnoncourt, which consisted of a series of essays on baroque music. One of these has a very important essay on rhetoric in music, in which Harnoncourt suggests that the period of continuo playing, the l7th and l8th centuries, was a time when music attempted to copy speech. A number of other writers, such as Robert Toft and Robin Headlam Wells have taken up this subject since then [see Judy Tarling's talk in LN56-Ed.] We can compare what we are trying to do on our instruments with spoken language. If you have never thought of things this way, it may help you. In the next column is a sort of table of correspondences between language and harmonic accompaniment.


LANGUAGE                                 CONTINUO

Vocabulary/words                          Chords

Consonants                                     Articulation

Vowels                                           Resonance/sustain of a Chord

                                                      Grammar/syntax      How to put chords in an order;

                                                      sequences; cadential suspensions etc.

Punctuation                                    Phrasing

Different languages                         Different styles

Rhetoric                                         Performance, live music, sryle,

                                                      taste, i.e. EVERYTHING which

                                                      cannot be experienced by reading

                                                      a book !

Words / chords

The most basic element of language is words, while the vocabu­lary of continuo playing is chords. Just as we may ask how words are spelt, and how they sound, so we can think about the voicing and sound of chords. Supposing one has a bass note with a `6' written underneath it, in one `language' one might play a bare sixth over the written note, while in another `language' one might play a big, rolled or arpeggiated chord. It is notable that in the original continuo treatises that have come down to us, chords are almost the only thing discussed; so we can have a reasonable idea of how musical language differed, from Italy in 1600 to France in 1730.


Vowels and consonants

As words may be considered in terms of consonants and vowels, so chords may be considered in terms of articulation and sustain, respectively. Different languages have different relation­ships between consonants and vowels; Henry Lawes complained that English is `clogged with consonants'. Articulation in lute playing is a matter of how we pluck the string, and how long we let the note last, and how we make the sound. Arpeggiation, rolling of chords, and attack are all components of musical articulation. Verbally we may characterise articulation with descriptions such as `quick', `bright' or `lazy'. In fact I do not know of an early continuo treatise that talks about articula­tion, though it is certainly very important.

Vowels in language correspond to the duration or sustain of chords in music. If we are accompanying singers it is the `sustain' of the instrument which they will need to support them, but of course the lute is all attack and very little sustain -strong consonants and very weak vowels. When working with other musicians I have found that often articulation is not talked about enough in the continuo department.


Syntax and grammar

Next we must consider the grammar or syntax of music, that is, how to put chords together in a meaningful order, with

sequences, cadential suspensions, and so on. The better histori­cal treatises do give clear ideas about how musicians thought one should put together the verbs, nouns and adjectives of music.

A very important element in musical grammar is what came to be called `the rule of the octave' which sounds rather daunt­ing, but simply related to how a normal musician would expect to harmonise a given degree of the scale, in a given key. For example, if you are playing in G minor, and you have an F sharp in the bass, you would expect a 6 or 6/5 chord for instance. The chords expected varied according to time and place, and there is a great deal of information about this in the old treatises, which we can and should study carefully.



As punctuation is in spoken language, so phrasing is in music. It is interesting that at exactly the time continuo playing was emerging, punctuation was beginning to be standardised in something like its modern form. Sixteenth-century English poetry seems to have been punctuated for metre rather than sense (so Bob Spencer pointed out), with for instance, a full colon marking the middle of a line, where modern sense would not require it. If you work with Dowland songs in facsimile, and compare the punctuation of the 1597 edition of the First Book ofSongs with later editions, you will notice that the punc­tuation changes a lot over just a few years. While early written punctuation was nonetheless not terribly consistent, writers on rhetoric and performance stressed that punctuation is very important in performance. If one plays a strong chord leading to a weaker one, one could play them in a legato way-unless perhaps one wanted to give the effect of an exclamation mark, rather than a comma. This aspect of continuo playing is not much discussed in the old treatises, but part of what we need as continuo players is to go beyond what is written.

Language and Style                                                                                                                  Different spoken languages are analogous to different sryles in music. The musical language of l8th century music for D minor lute is different from the language of music for 8-course lute of a century earlier: the contents of chords will be different and musical clichés will be different. For example, one does not find chains of 7 chords (on the circle of 5ths) in Dowland, but if you are playing continuo for a Vivaldi opera you would be playing them all the time.



Finally, rhetoric. I used to find this a terrible daunting subject; I was not taught the meaning of rhetoric at school or college. All it means, I would say, is an exaggerated way of singing, talking or playing which gets the attention of the audience. To do this we have to really believe what we are singing, talking about or playing. I think this it is all just common sense, but at the same time it cannot be written down, and is a most fascinat­ing area of musical performance. One has to follow the nature of the text, and to try to express moods such as sadness or anger. Occasionally there are examples in the old treatises, but these are few and far between. The things which really bring performance alive were not written down, so we will never know how the musicians of old played.


I would now like to turn to the general questions I mentioned a little earlier.


Lutes with everything?

Do we now want to have lutes and theorbos in `everything' in order to be fashionably authentic? Have we gone too far in adding lutes to almost every continuo section?

There is a recent recording of the Bach harpsichord concertos made by the Academy of Ancient Music, with Bill Carter play­ing theorbo. They have their fair reasons for including a continuo instrument in the band, but I found it rather shocking when I first heard it.

Other recordings and performance which cause one to wonder if we have gone too far include solo songs by Purcell accompanied by 3 or 4 instruments (Purcell is known to have played the harpsichord, and may have composed his songs on that instrument) or Monteverdi operas which often seem like a continuo concerto.

Of course, as a lute player it is a treat to play in all this lovely music, but we have to use our ears, and ask, does the lute blend? Does it balance? Do we need it?

Another way of approaching this questior, is to ask what is most natural to the lute. As far as balance is concerned, the lute works very well as a single instrument accompanying a single voice, and we know that historically it was in this role that the lute and more particularly the theorbo was adored in the l7th century. Harpsichordists may resent this, but nearly all the title pages of English songs collections in the l7th century specify theorbo, theorbo-lute or bass viol. It is not until around 1680 that the harpsichord starts to be regularly mentioned. Of course the modern way of merchandising music creates a problem, because one has to produce a whole CD of 20 or so songs, one after the other-and modern taste dictates variety. Varied continuo orchestration in solo song is not mentioned, for example, by Samuel Pepys in his diaries.


Nigel North then played as a music example, a recording of `The Plaint' fram Purcell’s `The Fairy Queen; in which the continuo section consisted of a bass viol, harpsichord and archlute.


I was fortunate enough to play for many years with a friend, the harpsichordist John Toll, and very often we played continuo together on harpsichord and archlute. It was enjoyable to blend the sound of the instruments and share this music, but since then I have wondered whether the songs would have been expressed differently with just a single accompanying instrument, and what it added to have two instruments playing together. (Certainly having two instruments playing together raises some technical questions, such as whether it is acceptable to play in octaves; we do know from harpsichord treatises and some writ­ten out realisations that players played lots of notes when they wanted to play loudly, and that they played such things as suspensions in octaves.)

Another case in point concerning the appropriateness of lutes is the music of Biber. Everyone seems to assume nowadays that you have to have a theorbo to accompany the violin sonatas of Biber, but in fact there is no mention of theorbo on the title pages. The most commonly mentioned instrument for accompa­nying Italian, Austrian and German music of the l7th century is probably the organ. Do we need to add any further instru­ments? I can think of only a single example where additions to the organ accompaniment are suggested; a set of violin pieces by Walther, in which the composer says you may add a bass viol or a lute.


Nigel North then played two more musical examples, reeordings of viol^n sonatas by Biber, one with organ alone, and one with organ and theorbo. He expressed a preference for the reeording without theorbo, and pointed out that one of the sonatas begins with a long pedal note, suggestive of organ accompaniment, witb­out plucked strings.


Another area where current practice is open to dispute is early opera. There have been cases of opera productions with perhaps 20 people playing in the orchestra, several of them on the continuo line, with only three people singing on the stage.

To be fair, some historical inspiration for big musically multi­coloured continuo groups comes from Agazzari's treatise, Del sonare sopra'l basso ("To play above the bass'), published in 1607, the same year that Monterverdi's Orfeo was first performed. He divides instruments into two categories, `foundation' in­struments, which can play everything, bass and harmony, and other instruments which can be ornamental. The lute and theorbo are included in the `ornamental' category, along with the violin-these are instruments which you can add if you want different colours. So according to their taste, modern-day musicians and directors may choose all sorts of musical colours.

A letter survives from Monteverdi, in which he considers writing a piece for a certain tenor; his original idea was to have a single chitarrone accompanying, but he says that if the work is performed in a larger space, a different type of singer will be needed, and Monteverdi then says that three chitarrones will be necessary, for reasons of balance. So choices had to be made then too, according to considerations of balance. (One may note also that he put three identical instruments together, some­thing which we might hesitate to do.)

Corelli is another interesting case. Theorbo or archlute is specified for his trio sonatas (Opus 1 and 3), but his Opus 5 violin sonatas say clearly on the title page that they are for violin, with harpsichord or violone. The `or' could mean `and/ or'-but there is no mention of baroque guitar, or theorbo!! As the sonatas are for solo violin I suspect that considerations of balance may have dictated the modest accompaniment section.

One more piece of evidence suggesting the importance of balance is in Thomas Mace's Musickš Monument ( 1674). Dis­cussing what instruments you should have in order to have a complete collection, he says that if you have two violins, you should also have two bass viols and `two lusty full sized theorbos'. (Of course, it is possible that some Italian violin players really did play so loudly, they needed more than one instrument to accompany them!)


He then played a recording of a Vivaldi violin sonata, with a cello, a harpsichord and a tbeorbo accompanying. This is the sort of tbing we hear nowadays, and it gives a rather orchestral sound, even tbough it is not what is suggested on the title page of the sonatas.


One final case which sheds light on questions of instrumenta­tion and balance are trio sonatas for two oboes and bassoon by Zelenka. He is a very interesting composer, who wrote some wonderful music. He worked alongside Weiss at Dresden. Some of Zelenka's larger chamber works, including these sonatas have two continuo lines, one marked `basso continuo' and the other marked `theorbo and violone'. We know that Zelenka was a violone player, and that there were two types of violone, one small one for playing in chamber music and a large one for play­ing in larger-scale ensembles. Perhaps, since the oboes would have been quite loud, and the theorbo is quiet, the theorbo and violone were added to the (keyboard) basso continuo part sim­ply to add weight and volume. The bassoon sometimes doubles the continuo part, and sometimes plays separately.


He then played a recording of a double oboe and bassoon sonata by Zelenka, in which the continuo section consisted of theorbo and violone (at 16 foot pitcb) alone, without keyboard.


Zelenka's scoring is a challenge, because if one was told simply that these sonatas are for two oboes and bassoon, one would probably not think of adding a theorbo, yet that is what the composer specified.

Of course, we can list some repertoires which certainly do call for a lute or theorbo accompaniment: the solo song reper­toires of England c.1580-c.1680 and of Italy 1580-c.1660, and the French 10-course air de cour l basso continuo airs repertoire; the sonatas of Salomone Rossi for 2 violins and chitarrone of c.1600-1610, Locke's Little Consort, Lawes's Royal Consorts, Fontana's compositions (c.1640) which use chitarrone as a bass instrument, and the trio sonatas of Corelli (Op. 1 and 3) and his predecessors (these call for theorbo as a bass instrument, rather than to play harmonies).

But of course, tastes and practices changed and by the mid l8th century the continuo section had generally been distilled into a single harpsichord (though in fact Quantz's orchestra still included a theorbo), so when it comes to some of the other repertoires, I wonder, are we historically, or hysterically, informed? Do we tend to ignore what original sources say about which continuo instrument is preferred, notably the almost universal standard of `organo' in many l7th instrumental and vocal repertoires?


What is natural for the lute?

Another way of approaching the lute as a continuo instrument is to ask what is most natural and idiomatic to instruments of the lute family. When my book on lure continuo came out, I was told that it was more or less `banned' at the Schola Cantorum in Basel because it contained ideas not derived from l7th or l8th sources or treatises. But then, as rather few treatises on lute continuo survive (and these give only the vocabulary and grammar of continuo, and very little on the punctuation or the rhetoric), we have to simply take inspiration from the nature of the instrument itself, and what it can do.


Number of voices

Playing in three parts often seems the most normal and natural thing on the lute, given the physical limitations of the lute family instruments. Keyboard continuo treatises often encour­age the continuo player to play in four parts, and occasionally this also seems possible on the lute.

However, to play in three voices, we may have to decide which note to leave out; the third, the fifth or the octave. Three-part accompaniment takes some practice! Much baroque (D minor) solo lute music is written in only three parts, with the excep­tion of broken chords (in Weiss for instance) which might have 7 or 8 notes, though not played simultaneously.

Playing one note at a time also seems very natural on the lute. English song accompaniments found in some l7th century MSS quite often have the bass note alone on the main beat, so that one waits for the other notes of the chord to be resolved; alternatively the bass note may be played alone as an `after shock'. And then of course, in the l7th and l8th centuries there is the style brisé in which the notes of a chord are played separately. Playing continuo in this style means we get to hear the individual notes of the chord very clearly, and perhaps gives an opportunity to make more sound and resonance; maybe we should be playing continuo in style brisé more often. And of course we can arpeggiate.

Varying the number of voices seems to be part of an idiomatic lute style, even though thís may lead to either íncorrect voíce leading, or indeed a style of playing which is purely chordal and not at all contrapuntal.

The fact that the lute is a non-sustaining instrument gives us the freedom to play harmonically, without worrying about voice leading-very different from the situation of keyboard players. (Indeed some original tablature song accompaniments, or dance pieces for solo lute, contain consecutive octaves and even consecutive Ffths to an extent that would never have been tolerated in vocal music.)

A great asset of the lute is its expressiveness, deriving from the fact that we are touching the stríngs dírectly with our fingers, and of course it is natural in accompanying to exploit the flexibilty of the lute's sound, tone colour and dynamic range.


A guestion of balance

We have already touched on the question of when and where lutes should be included in the continuo section. Even when there is just one lute, it is important to think about whether or not we are overwhelming or overshadowing the person we are accompanying. I have already suggested that balance ín the ensemble is very important. A further indication that this was the l7th century view is found ìn Bacilly's Remargues curieuses sur lárt de bien chanter (Paris, 1668). He wrote:

Among the instruments used at present to sustain the voice are the harpsichord, the viol, and the theorbo, the harp being no longer in use. The viol and the harpsichord have not the grace and accommodation found in the theorbo, which is necessary for accompanying all kinds of voices. This may be because the sweetness of the rheorbo adapts itself to weak and delicate voices, while the other instru­ments tend to obscure such a voice. The question therefore arises: is it necessary to be accompanied by a theorbo in order to perform a song propetly?

Undoubtedly the beauty of a song is not set off to good advantage when ít ís accompanied by an ínstrument which obscures the voíce. The ínstrument ought to accompany the person síngíng the melody (or the tteble part of an air) for the purpose of outlining the harmonies properly. This type of accompaniment is much more serviceable than the type in which the union of voice and instrument serves only to suffocate the fine points of the song in the resulting confusion, even though the result may be harmonically appropriate. In place of this confusion the theorbo merely sustains the voice pleasantly without detracting from either its beauty of the delicacy of its ornamentation.

However, it is necessary to establish the fact that if the theorbo is not played with moderation- the player adds to much confusing figuration (as do most accompanists more to demonstrate the dexterity of their fingers that to aid the person they are accompanying) it then becomes an accompaniment of the theorbo by the voice rather than the reverse. Be careful to recognise this, so that in this marriage the theorbo does not become an overpowering, chiding spouse, instead of one who flatters, cajoles and covers up one's faults.


I really think that some baroque opera continuo groups nowa­days fall foul of Bacilly's warning not to obscure the voice, and it surely is very annoying if one cannot hear the singer's words, and follow the story. I agree with Bacilly that the continuo player is there to support the singer, not to express the text

when the singer cannot do so-or to show of the dexterity of his or her fingers. On this latter point, one is reminded of Couperin's remark that he preferred playing harpsichord solos to playing continuo, because when you play contínuo well, nobody comes up and thanks you afterwards. If you do a good job no one will notice you!

Some very good general advice can be found in Gasparini's Lármonico pratico al c^mbalo (Venice, 1708). It is written for the keyboard player-if only more treatises like this had been written for the lute-and the main subject of the treatise is how to play when no figures are given. Gasparini was a harpsichordist in Rome, at the same time as Corelli, and a solidly practical musician; he composed music, and played in opera orchestras. He writes as follows:


In order to perform the accompaniments of recitatives with some degree of good taste the consonances [chords] must be deployed almost like an arpeggio, though not continuouslv so.

Of course we have to interpret what is meant by `almost Iike an arpeggio' but it obviously means the chord is somewhat spread so that all the notes can be heard. He goes on:

once the harmony has been heard, one must hold the keys fast,

in other words, one lets the chord ring on rather than stopping , it. He continues:

You permit the singer to take the lead, singing at his discretion, and in accord with the expression of the words. Do not annoy or disturb him with a continuous arpeggio or with ascending or descending scale passages.

I think some modern continuo players certainly fall foul of this warning:

I do not know if I should call these performers trivial, who in their desire to display their facility create confu­sion, and imagine that it is inspiration.


Too far, or not far enough?

So have we gone far enough with anything? Nowadays, we have many more competent lute continuo players than 30 years ago, but are we good enough musicians, with strong harmonic and stylistic sense, in comparison with keyboard players? Dare we be simple in our accompaniment style? Dare we be simple in our instrumentation choices? Dare we say no to playing in `ín­appropríate' repertoire?

Current and continuing misconceptions for plucked string players

I would like to conclude by listing a number of misconcep­tions that I think some lute players have in relation to our role when we play continuo. Important mirconceptions, are, I think:

 I) that we can be `replacement' keyboard instruments-for the reasons mentioned above, we cannot be;

2) that performance instructions found in a specific historical source can apply across the board. For instance, the combination

of organ and chitarrone is known to have been used in Monteverdi's Orfeo, and many people have taken this to mean that this was a common combination, but Mace, for instance, mentions either organ or theorbo as alternatives for accompa­nying sacred song, not as a combination;

3) that to have a lute/theorbo in almost anything is `authentic';

4) that we need to vary the continuo sound either to keep the audience interested (in recordings of, say, 12 Sonatas, one after the other) or to make an `innovation'-rather than concentrat­ing on good simple music making!

5) that the continuo needs to express the text of the song, rather than this being the singer's responsibility!

6) that good continuo playing still means to many, many people, being inventive and melodic, even though Gasparini and others make it clear that the role of the continuo was to unobtrusively support the singer or soloist;

7) conversely, that many of us also may think that simple chordal/harmonic accompaniment is not good enough and may even be boring to modern ears;


8) that we must not double suspensions or moving voices which the soloist may have-in fact some of the old treatises make it clear that doubling was often common practice;


9) that one universal continuo style applies to all 200 years of music (c.1580-c.1780).

Nigel North, May 2002


Footnote: Nigel North's Book, Continuo Playing on tbe Lute, Archlute and Theorbo

Few individual books are requested so often as this one. It was first published by Faber, then by Indiana University Press, but the original print run has now sold out. However, it is available on a `POD' (print on demand) basis, from a US firm called Replica. They sell through bookshops rather than to private individuals, so one should be able to order it via Borders, or Barnes & Noble. The ISBN is 0 2~3 31415 1. Replica's website is at