Thanks to this post on Peter A. Mahler’s New York Business Divorce blog, I learned that in a recent opinion of the New York Supreme Court (despite the name, a lower court), a judge considered the implications of either … or. I wasn’t about to let that pass without chiming in—regular readers will know that I have a thing about the semantic confusion created by or.
The case in question is Mintz v. Pazer (here). It involved dueling shareholders groups in two companies. I’ll spare you the details. Here’s the language at issue (emphasis added):
In addition, if any Deadlock exists which has not been resolved pursuant to Section 4.8.2 hereof, then either the Mintz Group or the Pazer Group shall have the right to give the other Shareholder Group a Purchase Notice as to all of the Shares owned by the other Shareholder Group within ten (10) business days after such failure to resolve, in which event the Shareholders shall proceed under Section 8.2(b), (c) and (d) below.
This extract describes a mechanism by which one shareholder group could buy out the other. The question was whether having one shareholder group give notice served to block the other group from giving notice subsequently. The court decided that yes, it did:
The word “or” is “a disjunctive particle indicating an alternative and it often connects a series of words or prepositions presenting a choice of either.” The Oxford University Press dictionary defines “either” as being “used before the first of two … alternatives specified (the other being introduced by ‘or’).” The Merriam Webster dictionary defines the correlative conjunction “either … or,” as “an unavoidable choice or exclusive division between only two alternatives.” Since the meaning of “either” is clear and unambiguous, extrinsic evidence may not be considered.
Thus, in examining the plain language of the Shareholders’ Agreement, the court finds that the use of the grammatical construct “either … or” and the singular “a Purchase Notice” in the last sentence of section 8.2(a) … denotes that only one of the shareholder groups could effectively serve a Purchase Notice for the other’s shares, and thereby, exercise the singular Right of First Offer. … If it had been intended that both shareholder groups were permitted to serve Purchase Notices after failed mediation, as contended by Pazer, this section would have had to have provided that both the Mintz Group and the Pazer Group shall have the right to give each other Purchase Notices as to all of the Shares owned by the other shareholder group. Section 8.2(a) of the Shareholders’ Agreement, however, fails to contain this language.
Add this to the very long list of instances of judges putting too much faith in dictionaries. That’s something I discuss in this post and this post. Dictionaries are like word zoos, in that they present words divorced from their context. It’s unrealistic to expect that you can shoehorn into a sentence the dictionary definition of a construction involving the slippery word or and expect that every time it will reflect actual usage.
Allow me to explain:
As noted in MSCD 11.35, the literature on drafting says that or can be “inclusive,” with A or B conveying A or B or both, or “exclusive,” with A or B conveying A or B, but not both. So in the language at issue, the question is whether the or is inclusive or exclusive.
MSCD cites The Cambridge Grammar of the English Language
on this topic:
According to CGEL, or is most characteristically used when the speaker believes that only one of the component propositions joined by or is true, and as a result or typically conveys the implicature that not all of the propositions are true. An implicature is a proposition that is implicitly conveyed rather than being explicitly said. It is not strictly part of the meaning of the sentence itself: as CGEL says, “or doesn’t mean that only one of the alternatives is true.”
…
And CGEL uses the example They are obtainable at Coles or at Woolworths to suggest that even a speaker who knows that both propositions are true might nevertheless use or. The reason for using or here even though the speaker could have said and is that it reflects the fact that you have a choice as to which store you obtain them from.
Most pertinently for our purposes, CGEL says that either only tends to strengthen the “not and” implicature:
I’ll be seeing her on either Friday or Saturday conveys somewhat more strongly than the version without either that I’ll be seeing her on just one of those days. Exclusiveness nevertheless is still only an implicature: They are obtainable at either Coles or Woolworths emphasizes the choice but, like the version without either, could readily be used in a context where they are obtainable at both stores.
In other words, regardless of what any dictionary says, use of either still leaves room for uncertainty over whether a given or is inclusive or exclusive. So if I had been in the judge’s shoes, I would have mentioned the possible significance of either, but I would have focused on the court’s other arguments, which I found persuasive.
What’s the message for drafters in all this? Scrutinize each or construction in a contract to make sure that it doesn’t convey a potentially problematic alternative meaning. (MSCD chapter 11 contains the most detailed discussion of this subject to be found on planet Earth.) And don’t rely on either to do the job for you.
Also, spell out in complete detail buy-sell procedures of the sort at issue in Mintz so that there’s no question about how the different scenarios play out.
Ken:
Courts should be banned by law and custom from referring to dictionaries. But it seems like the court got the result right. If the construction allowed both, then the two shareholder groups would end up swapping shares, which would probably do nothing to resolve the dispute. Is there anything in your reading of the case that would indicate otherwise?
Chris
I think the court got it right. As I noted in my post, I found the court’s other arguments persuasive.
1/ I agree that it was poorly drafted in that it threw the matter into a zone of unnecessary uncertainty. Agnostic about whether the parties subjectively intended to create a race to give notice, but as the saying goes, the intention of the parties is to be found in the meaning of what they did say, not in what they meant to say but didn’t.
2/ The passage with “either … or” was also MSCD-noncompliant in other respects: “pursuant to” instead of “under”; nitcaps instead of lower case on “Section” references; failure to put names of cross-referenced sections in parens after number; undisciplined use of “shall”; “shall have the right to” instead of “may”; “Shares owned by” is passive and also fails to specify the temporal point of ownership (what about shares acquired since date of agreement?); “within 10 business days after such failure to resolve” instead of clearer description of period from point to point; and use of “below” instead of “in this section” or “in this agreement,” whichever applies.
3/ Loose criticism of dictionaries and courts that resort to them deserves flogging (with full pardons for Ken Adams and Chris Lemens for their outstanding services to humanity). The OED alone deserves an hour a month of sheer adoration. And drafters of contracts should be on notice that courts will consult dictionaries if drafters fail to be clear within the four corners of the agreement.
After reading the whole long decision, I think the case was wrongly decided. (Not that that’s the point.)
The nub of the error was this:
“[T]he court finds that the unambiguous terms of the Shareholders’ Agreement dictate that only the *first* Purchase Notice served following the deadlock and unsuccessful mediation was to be valid and effective….” (Twinkly emphasis added)(p. 41).
The court interpreted the agreement to mean that only *one* Purchase Notice was to be valid and effective, but the agreement said nothing about which of multiple purported notices was to be the valid one. The agreement certainly did not say that the first in time was to be that one.
In my opinion, the court should have decided the case as if the two claimant notices had been simultaneous, since they were both given in the prescribed 10-day period.
That would have forced the court to find a contractual but non-chronological basis for deciding which notice was valid and, if it found no such basis, to proceed as if neither was valid, i.e., as if no notice had been given in the 10-day period.
I commend Judge Schmidt for his skill and effectiveness as a mediator, as well as his extraordinary willingness to give freely of his own time in meeting with the parties and their counsel, but in this instance, Homer nodded.
The correctness of the decision is an entirely different question from the lesson for us drafters, which at the highest level is always and ever “Be clear.” At the level of this case, it’s “If your deadlock-breaking mechanism is a race, say so clearly, and if it’s a bidding war with bidders limited to the parties to the agreement, say so clearly, and if it’s a public sale using appraisers and brokers, say so clearly.
If you fail to specify, your deadlock-breaking mechanism may be the creation of a judge with a dictionary, who necessarily claims he’s just carrying out what you agreed.
Mathematicians probably think we’re idiots for not simply using “xor” to denote the exclusive or and “or” for the inclusive or. (My undergraduate degree was in math.)
Another math idea that would be beneficial: Use “iff” for “if and only if,” with the single-eff “if” denoting a non-exclusive if-then possibility. EXAMPLE: You can have dessert iff you finish your vegetables. EXAMPLE: You can take senior status on the federal bench if you reach age 65 with at least 15 years of service [that’s not the only way you can do it].
Don’t get me wrong: I have no illusions that contract drafters will ever adopt either practice. Still, it’d be nice to see that happen.