.jpg)
Two years after a group of mostly Saudi men flew commercial planes into One and Two World Trade Center, resulting in both buildings' collapsing, several New Yorkers packed into a courtroom, a mile from where the buildings once stood, in order to hear a court case on the semantics of the word occurrence. In July of 2003, Larry Silverstein, the leaseholder of the World Trade Center site, sued the insurers of the WTC site, claiming he was entitled to 7.1 billion dollars, twice the amount the insurers believed he was entitled to. According to policy, Silverstein was guaranteed maximum compensation for any occurrence that led to the devastation or destruction of the buildings, but the language of the contract was shaky about what constituted an occurrence. An occurrence was defined as “losses or damages that are attributable directly or indirectly to one cause or to one series of similar causes.” In this somewhat idiosyncratic rendering, an occurrence is basically a bad event caused by a person or persons, or some other event, say a natural disaster like an earthquake. The question in the courtroom that July, for the three Second Circuit Court judges, was this: Was the attack on One and Two World Trade Center one event or two?
Silverstein and his lawyers argued that the attack constituted two events. At 8:46 AM, the five terrorists in the cockpit of American Airlines Flight 11 crashed into the north face of One World Trade Center, and about 15 minutes later, five different terrorists crashed into the south face of Two World Trade Center. It's possible that either never happened, that no terrorist attack was ever planned and carried out, that One and Two World Trade Center are still standing and that the people inside are at work as usual. It's also conceivable that after the first building was hit that an air marshall aboard United Airlines Flight 175 recruited passengers to help take out the terrorists aboard and diverted the flight path away from Two World Trade Center. Silverstein and council believed that if it's possible that no attack was carried out, or that only one attack was carried out, then the destruction of both towers must be two events.
The World Trade Center insurers claimed otherwise. They argued that the terrorists aboard those commercial planes intended to destroy both buildings and as such was part of one large plan. Therefore, the attacks constituted one event, they said.
To the three Circuit Court judges, none of this was clear. Everyone agreed about the facts but here they were listening to lawyers have a conceptual debate. And it was up to the judges to make a ruling based on the argumentation about what an event really is. Even though it might not have been clear to the judges, it's not too hard to understand the conceptual distinction being made here. When we're talking about people, should we think about the events they bring about as what springs from their larger plan? Or we should we think in terms of the result of their actions and number what they wrecked? The judges eventually sided with the insurers, by fiat determining in a court of law that an event is really a matter of the successful execution of plans, at least when we're talking about persons.
But there really is no answer to the question of what events are, outside of some framework or other, for a couple of reasons. One reason is that when talking about what people do and why they do what they do, we appeal to the reasons why they act as they do, and in the absence of knowing what the reasons are, we reconstruct them. Conceivably, two otherwise incommensurate constructions of reasons for actions can account for the events that follow. It's possible, for example, to construe the attacks on 1WTC and 2WTC as the conjunction of two events, with the intention to carry them both out, and it is just as valid a construal to conceive of the attacks as part of one orchestrated plan, and so as one event. This is what the lawyers were doing regarding the insurance claim.
There's another reason there's no telling what events are, though, a deeper reason, and that is because there are no word-world relations that we know of. Another way to say this is there's no semantics in the formal sense, let alone a real meaning to what events are. On that July day, when the lawyers for both Silverstein and the insurers used the word occurrence, they used it in reference to the attacks, roughly in the same context but with vastly different intentions. But the different intentions are not internal to the word; they're internal to the people using them. Questions about meanings of a word are really questions of normativity, namely how we ought to use the word. Even if we were to grant that certain words have a definite range of meanings—say we assume that events are limited in scope to either successful execution of a plan or conformability to acts we can individuate—this range would still be limited to the internal structure of a word or the word's concept and tell us nothing about the word's deeper relationship to the world.
The only way to say whether an event ought to be one thing or another is just to define it a certain way so that it fits the framework you want to use it in. If you want to make use of the word to understand how something works, you can't be agnostic about what it is. This defining of words ahead of time to fit some explanatory framework is not as strange as it might sound. For example, we talk about the gravity of a situation, or, a person's gravitas. This usage predates Isaac Newton's use of the word gravity, but it made sense to use it in the way that he did. He needed some name for what he was looking for, namely some name for force that attracts bodies. But regarding uses of the word event, we won't know how to make use of it unless we know what we're looking for. So we can't know what it is until then.
This has larger implications for rational inquiry in general. Think of Alfred North Whitehead's attempt to construct an event ontology as opposed to an object ontology, or Donald Davidson's event semantics. Or the lawyers' squabble over whether the WTC attacks were one event or two. We need to know what we're investigating here. What is an event? The question should be posed to those people seeking to understanding something. The appropriate response to anyone seeking to make use of the term when they want to know how the world works is: You tell us.
NOTES
An article on Larry Silverstein's claim against the WTC insurers can be found in Dan Ackman's article in Forbes titled “Larry Silverstein's $3.5 B Definition” (23 July 2003).
Discussion of the semantics of the word event is in Steven Pinker (2007), The Stuff of Thought, New York: Viking, pp. 1-24, and also in the context of the September 11 attacks.
Donald Davidson's “Actions, Reasons, and Causes (1963)” in Essays on Actions and Events (2001), Oxford: Oxford UP, pp. 3-20, was helpful to me in formulating how we explain human behavior in terms of reasons for action.
For more discussion of the possibility of semantics in the formal sense (in terms of word-world relations), see Noam Chomsky, New Horizons in the Study of Language and Mind (2002), New York: Cambridge UP.
I learned of Isaac Newton's use of gravity as metaphor from James Gleick, Isaac Newton (2004), New York: Vintage.
For more information on Donald Davidson's event semantics, I highly recommend Terence Parsons, Events in the Semantics of English (1990), Cambridge, MA: MIT Press. (NB. Amazon list price is $500. Best to check it out from your local library.)
Billie Pritchett is a writer and English professor with interests in moral and political philosophy, philosophy of social science, and phenomenology. He maintains his own blog called si hoc legere scis... and is on Twitter via @b_pritchett.
It’s interesting reading the case from a philosophical perspective, because I work in insurance law. But to take a step back, and at the risk of asking a dumb question: what exactly is the philosophical issue here? Is the question: how much are apparently neutral concepts affected by our own needs and interests? For example, when interpreting contracts, courts typically try to figure out what the parties would have decided before entering the contract, if the parties had addressed the issue during negotiations — which helps to explain some cases that might otherwise look weird to the non-lawyer.
Here’s a good discussion of the philosophical issues about events:
http://www.iep.utm.edu/events/
Hello, Dan:
Thank you so much for the question. You asked what the issue was in the article. It’s essentially an elaborate case that demonstrates that there’s no fact of the matter about what would constitute an ‘event’ or an ‘occurrence,’ that it’s a conceptual issue, and I don’t so much think the concepts are neutral so much as they are already value-laden, already dependent on needs and interests.
Philosophers have tried to formalize the ontology of events in what is called (misleadingly, I think) event semantics. You can read more about event ontology and event semantics through some of the links I included at the bottom of the article. You can also check out the articles that Alan Cook posted as replies here, and another good source is the Stanford Encyclopedia of Philosophy entry on ‘events.’
To give you a quick-and-dirty summary of how work in event ontology and event semantics has gone, there’s a group of philosophers who believe that one primary way we process information, or maybe *the* way conceptually, is in terms of events in the world. The story goes that our conceptual system primarily allows us to perceive agents in the world doing things, making things happen. Philosophers have tried to formalize this in symbolic logic, along the same lines that they did sentential and predicate logic.
I don’t know how to type the logical symbols here, and in any case it might not be too helpful, but just to give you a taste for how to formalize some bits of ordinary language in event semantics, let me give you an example from the Terence Parsons book I mentioned as a note to the above article. Take the ordinary sentence, “Brutus stabbed Caesar with a knife.” For the sake of ease, I’ll avoid tense here, but the way to read this sentence according to event semantics is like this: “There is an event, such that it’s a stabbing, the subject of which is Brutus, and the object of which is Caesar, and the means of completion of the event is/was with a knife.”
You don’t need to understand event semantics to understand the following but it might help to understand it in order to see why an event semantics or event ontology might be so meaningless. The main point comes down to the fact that it’s not clear what an ‘event’ is. You can ask, for instance, if the attack on One and Two World Trade Center was one event or two. You can also ask if the attack on the Pentagon was part of that event or another event. If you admit there’s no definite answer to this kind of question, then it means that the whole event ontology apparatus is going to be relative to needs and interests, and also means that there’s no isomorphic relation to a concept or word like ‘event’ and something out in the world. For ‘event’ and so many other commonsense notions, they’re words that are part of expressions used by people in given contexts, and not much more.
The word ‘event’ can always be used technically, of course, but then it has to be defined. It lacks definition in the contract, and the word ‘occurrence’ is awfully vague too. And it mattered a lot what an ‘occurrence’ or ‘event’ was because 3.5 billion dollars was riding on the issue.
Now you mentioned that typically contracts are interpreted by asking what the parties would have agreed to before entering the contract. Since you work in insurance law, perhaps you could provide some much needed insight into this particular case. I’m curious, then, what you would make of the case. Do you think that, assuming Silverstein and the insurers had the foresight to do so, they would have jointly agreed to have classified the collapse of both towers as one event?
Thank you for your very generous reply!
You asked, “Do you think that, assuming Silverstein and the insurers had the foresight to do so, they would have jointly agreed to have classified the collapse of both towers as one event?” After glancing over the cases for a few minutes, I think I have an answer: YES insofar as the parties agreed to the WilProp form’s broad definition of “occurrence”; and NO insofar as the parties agreed to the Travelers binder’s narrower definition of “occurrence.” I think this is basically what the Second Circuit held. (FYI, “WilProp” is short for “Willis Property,” and Willis is an insurance broker that was involved in negotiating insurance coverage for the WTC, so the “WilProp form” refers to the insurance form that Willis negotiated on WTC’s behalf with the insurers.)
It sounds like the opinion at issue here is World Trade Center Properties v. Hartford Fire Insurance Co., 345 F. 3d 154 (2d. Cir. 2003), which can be found at: https://goo.gl/BdmIfo. There, the court held that certain insurers (NOT Travelers) had agreed to a WilProp form, which specifically defined “occurrence” to include “all losses or damages that are attributable directly or indirectly to one cause OR TO ONE SERIES OF SIMILAR CAUSES.” Id. at 160 (emphasis added). The language I’ve emphasized led the Second Circuit to hold as follows: “no finder of fact could reasonably fail to find that the intentional crashes into the WTC of two hijacked airplanes sixteen minutes apart as a result of a single, coordinated plan of attack was, at the least, a ‘series of similar causes.'” Id. at 180. I think that’s basically right. The “causes” here were the two terrorists flying the planes at the same time as part of one conspiracy, so they have to be “similar” causes.
However, Travelers did not yet agree to the WilProp form at the time of the attack. At the time of the attack, Travelers had basically said, “we’ll agree to cover the property under some default boilerplate language, but we’ll hammer out the details later.” That default language was called a “binder.” In contrast to the WilProp form, the Travelers’ binder didn’t have a particular definition of “Occurrence,” and just defined the scope of coverage as follows: “[t]he Company will pay for direct physical loss or damage to Covered Property at premises …, caused by or resulting from a Covered Cause of Loss. Covered Cause of Loss means risks of direct physical loss.” Id. at 189.
To make a long story very short, the Second Circuit held that this language was ambiguous as to whether the two plane attacks counted as one “Covered Cause of Loss,” so that the jury could decide that the two plane attacks constituted either a single “occurrence” or two “occurrences.” Id. at 190. The jury ended up finding that the Travelers binder provided coverage, and the Second Circuit affirmed that finding in a separate opinion: “These forms were designed with different interests in mind and, not surprisingly, yielded different results. In our opinion, the jury’s determination that the insurers provided different coverage is not a manifestation of judicial error,” but rather “a reflection of the fact that the parties were at various stages of negotiating coverage when the two hijacked airplanes destroyed the WTC.” SR INTERN. BUSINESS INS. v. WORLD TRADE CENTER, 467 F. 3d 107, 140 (2d Cir. 2006), available at https://goo.gl/2G1qqK.
I think this other decision is also correct, since parties agreeing to an insurance contract typically intend to have broad coverage unless the policy specifically states otherwise — and here, the policy didn’t “state otherwise.” So, at least with regards to this latter decision, a role was played by background assumptions about “what the parties would have wanted.”
FYI, here’s an article on how courts determine the number of “occurrences” under insurance policies: http://daintorpy.com/download/single_multiple_occurrences.pdf. I think you would find a lot of evidence in the article for your argument that “event” is a hopelessly ambiguous notion.
Hello, Dan:
I was happy to read your response and be better apprised of some of the circumstances of the case.
Apparently, the quote I nabbed from the Forbes article on the definition of ‘occurrence’ comes from WilProp, and I found the ambiguous Travelers Indemnity definition as well, which renders an event or occurrence as “direct physical loss or damage to Covered Property at premises … caused by or resulting from a Covered Cause of Loss. Covered Cause of Loss means risks of direct physical loss unless the loss is excluded … or limited [in other policy language],” about which the Second Circuit Court said, “A jury could find that the words ‘direct physical loss or damage’ does not refer to the ‘event’ that triggers coverage at all, but rather sets forth the scope of damage resulting from the ‘event’ that the insurer will pay for, namely, direct physical damage as distinct from remote or incidental damage”–in other words, event as immediate damage rather than successful execution of one plan or conspiracy, say.
For other readers interested in the case, these quotations come from a lawyer named Doug Simpson and his website which contains law analysis. The URL is here: http://www.dougsimpson.com/blog/archives/000357.html.
Readers can also track the progress of the story after the initial win for the insurers and the subsequent appeals, the first of which Silverstein won (http://insurancenewsnet.com/article.aspx?a=top_news&lnid=244529091#.VOVLAELpwbE) and the second of which he lost (http://www.insurancejournal.com/news/national/2006/10/19/73411.htm).
And thank you, Dan, for that article on determining ‘occurrences.’ I’ll save it and read it.
On another note, I did want to know about this. As a lawyer, do you have access to an online law corpus, and is that what you used to help pull up information on this case?
Best wishes, and Happy Lunar New Year,
BP
I generally use Westlaw for work. Westlaw is basically a big database of opinions and statutes and such, and it has “official” curated versions of the text. However, free “unofficial” versions of all that stuff is on the internet too, if you just look for it. I’d suggest the following free resources:
Google Scholar basically has copies of any caselaw opinion you could ever want: https://scholar.google.com/
The Legal Information Institute has free copies of statutes, constitutions, and regulations (or links to free copies): http://www.law.cornell.edu/
Justia also has opinions, statutes, etc.: https://www.justia.com/us-states/
The Law.com dictionary will define most terms: http://dictionary.law.com/Default.aspx
Scotusblog will give a quick and dirty explanation of recent Supreme Court cases: http://www.scotusblog.com/
You can usually find ordinances and such on a city/town’s website.
When in doubt, just Google it. Wikipedia is also generally accurate.
Between all that stuff, you can usually find anything without paying for it, and 95% of the time it’s right. The only problem is that these free resources don’t explain whether a case has been overturned, and might include statutes and such that have been repealed or held unconstitutional. To get that extra information, you have to “shepardize” the material using a Westlaw or Lexis search engine.
And here’s an example of a real-world application of this kind of speculation:
http://motools.sourceforge.net/event/event.html