It’s generally a bad sign when, barely after it begins, a sentence shifts into a set of enumerated clauses:
If (1) A, (2), B, or (3) C, then X.
Acme shall (1) A, (2) B, and (3) C.
In such sentences, the drafter is forcing the reader to make a connection between each enumerated clause and the stub beginning of the sentence. In the first example, it would be better to incorporate the stub into each of the enumerated conditional clauses. (It’s a separate question whether the conditional clauses should go after the matrix clause, then X; I mention that issue in MSCD 2.175.) In the second example, it would be better to create three separate sentences or find some way to expand the opening stub.
Something about this offends the anti-redundancy principle. I realize there’s a term for the rhetorical device that involves beginning or ending multiple sentences with the same phrase (my daughter the classicist could tell me in an instant, but I couldn’t raise her on the phone just now), but it seems to me that what Ken is looking for here can be achieved without either (notice how I avoid putting in enumerated subheadings here?) making the reader refer back to the trunk or invoking classical poetry. Could one not say that party A will do X in (any or all of) the following circumstances: (a), (b), (c)? Of course, this raises the ugly specter of leaving a subject (noun or pronoun) dangling in front of the subclauses, but I’ll leave that battle for others.
I prefer enumerated clauses to separate sentences. For example, if a seller promises to that it will follow five rules during an inspection period, I think “Seller will a, b, c, d, and e” is easier to follow the first time reading it, less likely to be ambiguous, more efficient, and has a tighter logic and better aesthetic than “Seller will do a. Seller will also do b. Furthermore, and without limiting a or b, Seller promises that it will be obligated to do c, etc., etc. But maybe I’m missing your point.
Paul: Try replacing your letters with provisions of average complexity and you might feel differently. Ken
I think that you are right in your particular examples Ken, but I would analyse the issue slightly differently.
I think the two things to consider are:
(i) Whether to ennumerate in the first place. For me, this is about whether it is useful to think of the items in the list as connected and whether they would be unclear in a single sentence; and
(ii) where the sentence should be broken. For me this is about whether the list items are each coherent within themselves, and this determines whether the “look back” problem arises.
Starting with point (i), I’ll use different versions of an invented (and fairly meaningless) clause to illustrate why I think this:
Version 1
“In the event of X, Acme shall write a letter to TopCo detailing the losses incurred on the transaction and listing all items unrecoverable from third parties.
In the event of X, Acme shall write down the amount in the Company books.”
Version 2
“In the event of X:
(a) Acme shall write a letter to TopCo detailing the losses incurred on the transaction and listing all items unrecoverable from third parties; and
(b) Acme shall write down the amount in the Company books.”
I think that Version 2 makes it easier to take in the concept that event X has two consequences, which is a useful thing to have a feel for. Also, if you made the two items clauses 22 and 23 in a 100-clause contract, you would need to read all the other clauses to see whether X had any other consequences. If the contract is well-drafted, a list should set out the consequences.
Conversely, if there is no meaningful connection (as with your second example), the list is just distracting – I would also separate them in that case.
For my point (ii), compare Version 2 with this Version 3:
“In the event of X, Acme shall write:
(a) a letter to TopCo detailing the losses incurred on the transaction and listing all items unrecoverable from third parties; and
(b) down the amount in the Company books.”
As I read it, only Version 3 requires the reader to look back. I think that a reader does not forget the idea in the stub (that we are looking at consequences of X), but does forget the rest of the actual words needed to make sense of a phrase in the list item. So long as the list items are genuine concepts (rather than just the ends of sentences chopped off at the point where there is the last common word), the list flows.