top of page
  • Writer's pictureSarah MacDonald

Where does the consonant go?The tied quaver/eighth-note in 20th-century British choral music

Updated: Dec 18, 2021


This article was first published in The American Organist magazine in November 2016.

A perennial question with which choir members interrupt rehearsals runs as follows: “Where does the final consonant go?” This legitimate enquiry can be particularly irritating when asked by someone who only achieved 60% accuracy in the preceding phrase. Now, I am a stickler for final consonants, some might argue to a fault. Nonetheless, if by the end of a rehearsal, a choir’s consonants are perfectly together but the notes are still wrong, one wonders about their conductor’s priorities.

In theory, of course, this question should not need to be asked by anyone from a well-trained choir. However, confusion inevitably arises with that very British[1] of choral quirks, the tied quaver, or eighth-note. How many times have you seen this:

and agonised about where the ‘d’ should be placed? (This example is taken from Howells’ Like as the hart.[2]) Perhaps those of us who are particularly OCD about final consonants can be forgiven for having forgotten that 40% of the previous phrase was inaccurate.

Historically, these pesky quavers/eighth notes began to appear inconsistently in the early works of Elgar (there are a number of instances in the 1897 Te Deum and Benedicite Op. 34, for example). Vaughan Williams was the first to use them habitually, though his application of them is also inconsistent (see below). The tied quaver/eighth-note is especially ubiquitous in the works of Howells, Finzi, and Leighton, and their ilk, and has for many years perturbed singers and conductors alike. Fortunately, many contemporary composers (James MacMillan, Jonathan Dove, Cecilia McDowall, et al.) seem to have abandoned them.

So, what is that tied quaver/eighth-note for? Is it correct to place the consonant directly on the note, or should it occur on the ensuing rest? What were the composers’ intentions? And, assuming they had intentions for those consonants, what on earth did they mean by adding a tied quaver/eighth-note to a word ending in a vowel—as in this example from Vaughan Williams’ A Choral Flourish?[3]

My personal practice is consciously inconsistent. I consider each instance individually, and base my decision on tempo, style, harmony, and context, varying my verdict from bar to bar. I do not teach my choirs a specific ‘policy’, which is why that aforementioned question occurs. In an informal survey of cathedral and Oxbridge colleagues, I was not surprised to find that they follow this route as well. Those who do attempt to abide by a policy inevitably stray from it on a regular basis.

It is widely agreed that tempo is the most important factor in deciding where to place the consonant. In general, at a faster speed, the consonant is placed directly on the tied note. So, in the quick middle section of the Magnificat from Leighton’s Second Service,[4] placing the consonants on the tied note enhances rhythmic clarity, and keeps contrapuntal ensemble precise.

On the other hand, in a slower, more lyrical work, placing the consonant on the rest after the tied note creates subtle and less perceptible joins between phrases. For example, in Howells’ O pray for the peace of Jerusalem,[5] the consonants in the opening and closing sections are most musically placed on the rests (or at least this is true at the tempo at which I take the piece!).

Frustratingly, composers themselves were inconsistent. In Vaughan Williams’ setting of Let all the world,[6] the following two phrases occur within three bars of each other. Surely they are not meant to be sung differently?

Elgar’s anthem “The Spirit of the Lord” (from The Apostles)[7] contains the following three examples of tied quavers/eighth-notes, each of which demands a different interpretation.

For ‘poor’, a tied quaver/eighth-note is provided for no final consonant (the ‘r’ is not pronounced when singing with a standard English accent).

Putting the ‘s’ of ‘captives’ on the rest creates a more refined connection between the phrases.

The complexity of the syncopation and the rallentando in the accompaniment makes putting the ‘s’ of ‘righteousness’ on the tied note by far the simplest and tidiest solution.

A particularly awkward example of the phenomenon occurs in Howells’ Magnificat in B minor,[8] during the quicker middle section, which is conducted in minims/half-notes.

Howells’ original:

My proposed alternative:

Placing the ‘t’ on the tied note itself makes the gap feel unnaturally long, and inevitably someone comes in early after the rest. On the other hand, it is mathematically very difficult to place the final ‘t’ of ‘seat’ on an eighth-note rest when counting in half-notes. My (undoubtedly controversial) solution is actually to rewrite Howells (!) which fortunately works well harmonically in this case.

There is, of course, no right answer. Composers themselves were varied in their approach, and it is a relief that the practice seems to have fallen out of use. As to why it was introduced originally, one theory is that the tied notes were added to encourage less-accomplished singers to hold long notes with conviction right to the end. Since many of the works in which they first appeared were oratorios written for amateur choral societies, this conjecture seems plausible.

The late former Director of Music at King’s College, Cambridge, told me that he had actually asked Howells about the tied quavers/eighth-notes many years ago. Howells “didn’t comment in detail”, presumably “preferring decisions in this area to be made by conductors.” [9] Consequently, it looks as if we are required to make an informed and musical judgement in every individual case rather than trying to construct a universal policy—but mark up your choir’s scores before the rehearsal, in order to avoid being asked that old familiar but irritating question!


[1] Re: I do not believe that Gabriel Fauré was doing the same thing as RVW, Howells, et al. Fauré’s occasional tied quavers/eighth-notes simply indicate the duration of the note he wanted, i.e., a dotted-crotchet/quarter-note for the final syllable of ‘luceat’, with the ‘t’ occurring on the rest. [2] Howells, Herbert. Like as the hart. OUP. 1943. [3] Vaughan Williams, Ralph. A Choral Flourish. OUP. 1956. [4] Leighton. Kenneth. Second Service (Evening Canticles). Novello and Co. 1972. [5] Howells, Herbert. O pray for the peace of Jerusalem. OUP. 1943. [6] Vaughan Williams, Ralph. No. 5 from Five Mystical Songs. Stainer and Bell. 1911. [7] Elgar, Edward. The Apostles Op. 49. Novello. 1903. [8] Howells, Herbert. Evening Service in B minor. Novello. 1955. [9] Stephen Cleobury, personal correspondence, 29 August 2016; quoted with permission.

1,321 views1 comment

1 Comment

Jeremy Backhouse
Jeremy Backhouse
Sep 08, 2021

Really interesting article, Sarah! I am equally inconsistent with the tied quaver issue, dependent on the context. The one difference that I make in your examples above is in the VW Let all the world. On the word "shout", there is a crescendo in voices which is not easy to effect convincingly on just a crotchet; so I always extend that by a quaver to make a full minim. Each to their own!

bottom of page