The earliest layer of responsories?

GKS 3443 8°: Ordines Romani, XIIIa
https://permalink.kb.dk/permalink/2006/manus/101/eng/

In my last post in this series on the Matins responsories, I looked at the evidence around the earliest Roman reading pattern for Matins, captured in Ordos XIV and XVI, and I noted  that the latter Ordo contains what is probably the first explicit surviving reference to the psalm based responsories (regardless of whether it dates from the seventh or the late eight centuries!).

And since Ordo XVI, regardless of its date, does represent a plausible way of integrating Ordo XIV's 'calendric' listing of the books to be read at Matins, it is potentially helpful to us in exploring some of the key questions around the history of the psalm based, or 'de psalmiis' responsories.

In particular, perhaps the most important question is whether the 'ferial' responsories, or 'historia' sets associated with particular books of the Bible developed in the context of the reading cycle in Ordo XIV, or only the psalm based responsories were used in this period; or whether they developed around the later reading cycle that replaced that of Ordo XIV (I will come back to the question of the date of the later reading cycle, but for the moment it is worth noting that it seems to have emerged in the early to mid eighth century).

How old are the history responsories?

One of the most important plank of the case for the use of the de psalmiis responsories as a ferial set, ferial responsories is the claim that the psalm based responsories represent the oldest layer of responsories in the Office, and there were not a sufficient number of 'history' (and festal) responsories to support the Benedictine Office until very late indeed.

The first statement of this claim, I think, can be found in René-Jean Hesbert's pioneering compilation of  responsories from twelve of the earliest antiphoners, Corpus Antiphonalium Officii (1963-1979).  But variants of it have been repeated many times since, not least in James McKinnon's The Advent Project, Robertson's The service-books of the Royal Abbey of Saint-Denis and Jesse Billett's The Divine Office in Anglo-Saxon England.

In this post I want to look at some of the evidence for different chronological layers in the responsories, with the help of Ordo XVI.

Understanding Ordo XVI

Ordo XVI, though, it has to be said, is a somewhat cryptic document that is not easy to interpret - if it does indeed date from the late seventh century, it reads as if it is perhaps an attempt after the fact to capture the various quite diverse topics touched on at John the Archcantor's Anglo-Saxon chant workshops, rather than as lecture notes or instructions per se.

All the same, since it is our earliest piece of evidence on their use, it is important to interrogate the document, not least since regardless of its credentials, it 

Ordo XVI's most important provisions, from the point of view of the history of the use of the de psalmiis responsories, relate to the reading cycle after Epiphany, and its reference to the de psalmiis responsories.

On the de psalmiis responsories, you will recall, it says:

Reliquo tempore in anni circoli praeter quod memoravimus ipsis psalmis responsuria nende [The rest of the time during the year, except for those that we have mentioned, the psalm responsories are used.]

In my last post I mentioned that the most likely interpretation of this is that it is referring to feasts without their own responsories.  

There are other possible interpretations though, so today I want to focus on where, other than on feasts, the psalm responsories could have been used.

The Advent Project?

I've previously noted that Ordo XIV, the earliest Roman reading cycle, simply lists the books to be read.  Ordo XVI seems to make it explicit that there are responsory sets associated with each of these sets of readings.  

In its first paragraph, it refers to responsories associated with Isaiah (read from the start of December in Ordos XIV and XVI), consistent with the approach to Advent in the other early Roman books in the lead up to Christmas, and responsories for them.  

It says the same thing for Jeremiah and Daniel, which then follow.

The subsequent references to the ferial cycle don't explicitly mention responsories associated with them, but it is reasonable to assume it is implied.

That said, the alternative possibility that it means exactly what it says, and the responsories for Advent and Epiphanytide referred to in Ordo XVI reflect their seasonal status rather than the books being read, and the de psalmiis responsories were used the rest of the time.

One of the key assumptions around the history of the Office responsories, after all, has long been that there probably wasn't much - if any - continuity between the responsory sets used by St Benedict, and those that emerged in the course of the seventh century or early eighth keyed to what is more or less the modern reading cycle.

We've already looked at some evidence that casts doubts on this theory from the fifth to sixth centuries, in the form of the early establishment of sets of festal responsories; the establishment of Office homiliary sets for feasts that align with this; references to an established 'canonical' set of responsories; and early specific references to a Kings responsory.

 How old are the de psalmiis responsories: psalms vs Kings

When it comes to the ferial cycle though, the evidence for different chronological layers in the responsory cycle takes a little more digging.

I've previously noted that there doesn't seem to be any musicological basis to the claim that the de psalmiis responsories are older than the rest of the 'historia' sets that have come down to us: a study by Brad Maiani found that it was impossible to differentiate musically between the de psalmiis responsories and the Kings/Chronicles responsories in terms of the structures and melody components, and the way they were used.

One claim, for example, is that the de psalmiis set are significantly shorter than the other responsories, reflecting their origin as psalm refrains.  Maiani, however, has suggested that this is largely driven by the text: the Kings prose is more verbose than the poetry of the psalms.  

Even so, the differences in length are not great: there is really only extremely short psalm responsory in the set, noted in my previous post.  

The table below provides a very crude measure of this: it compares the number of lines the transcriptions of the respond sections take up in the 'Nocturnale Project' versions of them, for Roman Sundays; Gregofacsimil for the remainder. 

Respond length (lines)	De Psalmiis Sunday Roman	De psalmiis weekday	Monastic’ Sunday	Kings	Hartker Kings additional
1 - 2	-	1		-
2 – 2.9	2	7		2
3 – 3.9	2	7	1	-	3
4-4.9	3	3	1	3	1
5 -5.5	-	-		2

It shows that the Sunday psalm responsories used in the Roman Office (as well as the additional responsories used in the monastic Office or listed in the monastic Hartker MSSS are generally around the same length range as the Kings ones, though the Kings set has two longer responsories.  The weekday set are generally shorter, but not systematically so.

On the face of it, then, the psalm based responsories and the Kings set were probably composed at the same time, not one much earlier than the other. 

The eighth century layer and Ordo XVI

Maiani did, however, identify what look like a number of later compositions that seem likely to have been added to fill in gaps created by the extra time allocated to the reading of some books in the reading cycle employed from (at least) the eighth century onwards.  

There is another side of this coin though: I want to suggest that it is also possible to find traces of the contraction of the amount of time allocated to some books of the Bible in Ordo XIV compared to the later pattern, and of a less 'properised' set of responsories.

Kings-Chronicles

Consider, for example, the Kings/Chronicles cycle after Pentecost in the post-Trent books.  

The 'core' group in the post-Trent Roman Office (unlike the Benedictine) does not include any responsories using texts from Kings/Chronicles.  

Instead, three additional responsories, two based on Chronicles texts, are added on weekday days from the Fourth Sunday after Pentecost onwards. 

There are several more Chronicles responsories preserved in early manuscripts, presumably remnants of a more extended cycle from the time that the reading of Kings/Chronicles extended up to mid-October.

Lent

Similarly, Amalarius' introduction to his antiphoner describes his struggle to try and 'properise' the responsories for Septuagesima and Lent, a process that eventually led to the cycle based around the Genesis Patriarchs we still use.  

But the early manuscripts also preserve a considerable number of responsories relating to the other six historical books originally read at this time: in the surviving manuscripts, they appear in other contexts, such as Commons and for particular feasts.

More work would need to be done on their musicological features to verify this, but it is at least plausible, it seems to me, that they had their origins in the lectio continuo of Genesis to Judges envisaged in Ordos XIV and XVI, but were then displaced by later properization associated with the new reading cycle. 

The creation of Epiphanytide and Advent

One important example of the possible redeployment of surplus responsories may have occurred for the period of the year that most concerns us, namely after Epiphany.

In Ordo XIV and XVI, three books were read in the time between Epiphany and pre-Lent: Job, Ezekiel and the Minor Prophets.  In the later schema all of these were moved to other times of the year: Job to September; and Ezekiel and the Minor Prophets to November.  

Although the minor prophets feature in the later 'Prophets' historia set though, and Job has its own full set, there are no Ezekiel responsories at all in November.

So is this perhaps an example of where the de psalmiis responsories were originally used?

In fact though, there are at least a full set of responsories based on texts from Ezekiel, but in the modern Office, they turn up in other contexts, such as the feasts (and Commons) of Evangelists, Advent, and Lent.

The problem, it would seem, is that there simply wasn't enough space in November because it also had to accommodate seasonal responsories for the new five week Advent that was put in place in the early eighth century.

The epistles?

All of this suggests that although there was a great deal of fluidity in the responsory cycle as it developed over time, there is no obvious place for or evidence of the use of the de psalmiis responsories as an invariable ferial set in the six and seventh centuries.

There is one other possibility I want to explore though, namely that they were indeed used throughout the year, in the context of the third Nocturn reading of the Pauline epistles.

But more on this anon.