When you encounter confusion in contract language, the thing not to do is to stamp your feet and insist that your interpretation is the sensible one.
The confusion I have in mind is whether a deadline of, say, 5:00 p.m. passes once you’ve entered the first second of the five o’clock hour or whether it ends after the last second of the first minute of the five o’clock hour. [Updated March 22, 2015: For an example of a dispute caused by this confusion, see this 2009 post.]
Here’s what I said about that in this March 2012 post:
Today I’d like to advance a simple but ill-understood concept: A time of day isn’t a period of time. Instead, it’s a boundary between the period of time that comes before and the period of time that comes after.
So if a contract states that a given deadline is midnight at the end of March 13, 2012, the deadline will pass once the last second of the 11:00 p.m. hour expires and the first second of the 12:00 a.m. hour begins. Midnight is the boundary between those two seconds.
It seems that many drafters don’t understand that.
I mentioned this issue briefly in my recent post on midnight, prompting this comment by A. Wright Burke. With the acuity that you’d expect from someone with an M. Phil. in morology from the University of Wiltshire, he says the following:
Unless a drafter makes clear that clock times expressed in hours and minutes are points and not 60-second periods, the possibility of legitimate confusion remains, even with the expressions ‘midnight at the start of (date)’ and ‘midnight at the end of (date)’, since those phrases are often understood as synonyms for numerical clock times.
It is surely not ‘delusional’ to reckon that it’s 2:00 p.m. till it’s 2:01 p.m., so that a deadline of 2:00 p.m. occurs at the *end* of 2:00 p.m. and at the *start* of 2:01 p.m. A Canadian trial court so held and was affirmed on appeal. So it’s the drafter’s job, as ever, to head off possible confusion.
I hereby accept that he’s right.
Wright offers as a solution the following provision specifying a drafting convention:
In this agreement, clock times in hours and minutes are points, not periods. A clock time refers to the point 60 seconds before the next minute. Midnight at the end of any date refers to the point at the end of the day. Midnight at the start of any date refers to the point at the start of the day. Noon on any date refers to the midpoint of the day.
Here’s my take on this [adjusted June 22, 2015]:
In this agreement, a statement of a time of day denotes not a period of time but a boundary between periods of time, occupying no time. For example, “midnight” denotes the boundary between the last second of one day and the first second of the next day; it does not denote the first sixty seconds of the next day.
But for such a provision to be worth including, you’d have to be concerned about the possibility that someone might attempt to deliver a notice, or submit a bit, during that one potentially contentious minute. I’m not sure how often that’s enough of a concern to be worth adding two sentences.
And if anyone’s going to fight over that, they might fight over who had the right time. Acme might claim that Widgetco stopped accepting bids thirty seconds too early! I invite A. Wright Burke to come up with a provision stating that whether a party meets contract deadlines is to be determined by reference to the time of day according to, oh, the National Institute of Standards and Technology.
Note that neither A. Wright Burke nor I propose stating times of day down to the second, as in 12:00:00. Using that convention would in effect involve accepting that a time of day is a period of time, albeit in this case a period lasting only a second; I’m not willing to do that. Furthermore, humanoids don’t state the time of day like that.
Ken:
I don’t think I’d ever use this. Typically, my agreements will say “by” a certain time, which would be either (a) that time as a point or (b) the beginning of the period. Unless I’m missing, something, which is usually the case.
Chris
“Acme shall purchase the Widgets by January 2016.” By when does Acme have to purchase the Widgets? Before the first second of January? Or before the last second of January? “By” doesn’t preclude a fight.
Ken:
If “by” doesn’t work, then I would try “before.” If you say “before January 2016,” I can’t see an interpretation that allows January 2.
Chris
Sure, but that’s not the only kind of reference to time.
I’m with Chris. If I could see an example of where this matters in practice, it might affect my views. I don’t think an English judge is going to care about compliance with deadlines to the nearest minute, and nor should commercial parties. Whose watch are we using? Is it set to the atomic clock in Paris? Is there a time lag in getting the Paris signal through to darkest Wiltshire? Do we have affidavit evidence about all of this?
Here’s an example of people fighting over this: https://www.adamsdrafting.com/deadline-for-submitting-bid/.
Interesting. Still not sure that it is something that deserves lots of words in a contract.
Sure. I acknowledge as much in the post.
Wait!
Don’t trivialize the issue, which is ‘How do you specify a point in time precisely when it matters?’
Or if you prefer, ‘How do you describe a period of time precisely when it matters?’
If you posit that it never matters, it’s worth no ink.
But there are often lease terms, deadlines, expirations, option periods, or notice periods where exactness matters.
In those cases, the question is what’s the best and concisest way to achieve precision.
That’s proper grist for MSCD, which seeks to standardize optimum usages.
At a high level of abstraction, one concise approach is an interpretive rule like this:
‘[In this agreement,] any point in time expressed as a period of time refers to the start of the period.’
Under that 17-word interpretive rule, which Ken has described as ‘brain-imploding’, many concise but murky time references become pellucid.
For example, ‘from 10 a.m. to 11 a.m.’ means ‘from the start of the 60-second period known as 10:00 a.m. to the start of the 60-second period known as 11:00 a.m.’
‘From January 1, 2019 to January 1, 2020’ means ‘from the start of 12:00 a.m. on January 1, 2019 to the start of 12:00 a.m. on January 1, 2020’.
‘From June to September’ means ‘from the start of 12:00 a.m. on June 1 to the start of 12:00 a.m. on September 1’.
I don’t mean to imply that the rule is the panacea. It has problems that need addressing, such as what to make of ‘from April *through* September’.
But it’s one contender among several for MSCD-worthiness.
Defining clock times is a less general approach. If a drafter goes that way, the concisest approach is best. We’re not there yet, in my view.
A third approach is defining one or more of the common prepositions: after, at, from, on, starting, before, by, on, to, through, and until, then using them consistently as defined. ‘Through’ seems like a special case.
A fourth approach is to precisely define on site every point in time and period of time in the contract. Even in concise formulations, this is likely to spill much ink, but if there are few such references, it may be better than a ‘drafting convention’ with only one or two applications.
Remember also that contracts often have periods for which the dates and times of the start and end points are unknown at the time of agreement, like an acceptance period that starts upon notice of an offer.
Another time-related drafting danger is the ‘Australian passport problem’ of stating a period in two inconsistent ways: the law says the passport is good for ‘ten years’, but the document says it’s good from the moment of issuance to the end of the same date ten years later, which makes the period ten years and some hours.
Anyway, the subject of time deserves its chapter in MSCD. Ken should pick the best approaches and formulations and recommend them in MSCD 4th edition for uniform use.
Perish the thought that I should trivialize the issue! Instead, my point is that it’s appropriate to assess the risk. If it seems remote that the parties to a contract would squabble over an attempt to deliver something during that contentious minute, then it would be reasonable for a drafter to elect not to eliminate that potential confusion.
But beyond that, yes, I’ll consider all the possibilities for purposes of MSCD4.
How about this variation of Ken’s concept: “Unless otherwise clear, a stated time of day refers to exactly the stated time.”
We could add: “HYPOTHETICAL EXAMPLE: A certain action must be taken at or before 5 p.m. The action is completed at 5:00:00.01 p.m. The action is untimely.”
I agree that this won’t often come up — but it might. For example, in the U.S., patent-related electronic filings must be filed at the USPTO Web site before 12 midnight Washington D.C. time on the deadline date. A while back, due to client delays, I did an important filing with 30 seconds to spare.
Won’t work, for the same reason this won’t work: ‘Unless otherwise clear, a stated year refers to exactly that year’. What needs to be clear is that a time of day is a point in time, not a period of time, and when that point is. The method of examples is often useful, though: ‘Midnight is the point that starts the day, and 12:01 a.m. is the point sixty seconds later. Subsequent clock times stated in hours and minutes are points calculated accordingly.’
Taking this a step further, to an absurd level, what frame of reference do we use to take into account relativistic effects on time? For example, time is measured differently if you are traveling at a high rate of speed – it appears that your clock is running slower than someone at rest at earth (from your perspective). The faster the traveling person is going, the greater the effect of this time dilation effect. The traveler’s watch might say 11:59 while the stationary person’s watch might say 12:01, and both would be correct! Should we also have to state the frame of reference which is the measuring stick? But then, of course, we need to deal with the twin paradox problem too!!!! #mindexplosion #overlawyered
And what if one party travels down a wormhole, so that a day or so for them equals seven years on Earth? I’m now an expert in such things, having seen “Interstellar.” I expect this to have a meaningful effect on my drafting practice.
One last shot, and I’ll pack it in: it’s not about nanoseconds and relativity, it’s about how to clearly describe *any* period of time of *any* length. For example, do the words “from 2020 to 2030” describe a nine-year, ten-year, or eleven-year period? What’s the most efficient way for a drafter to make clear which option governs? Wright out.