Some investigations more or less launch themselves because they are precipitated by what David Binder and Paul Bergman have referred to as a "moment (or moments) of substantive importance." Binder & Bergman, Fact Investigation: From Hypothesis to Proof 191-195 (1984). These are events (in my usage here) that plainly (or very probably) have potential legal significance and, thus, justify the suspicion that investigation ought, at least initially, center on them. In many instances, however, the investigator is not so fortunate: no event leaps forth as a moment of substantive importance, as the obvious focus of an investigation.

Investigation is intrinsically hard because in an investigation one is trying to learn or acquire that which one does not yet know or have. The initial phases of investigation are particularly hard because (generally speaking) in the initial phases of an investigation very little is known. The hardest investigation to conduct may be the one that begins without a "moment of substantive importance." In such a situation the investigator faces the formidable task of determining a sensible investigative strategy when (s)he -- the investigator -- is almost completely in the dark.

So what is a prudent investigator to do in a condition of ignorance such as this?

One way to think about the initial phase of an investigation is to think of it in terms of "anchoring": one might say that an investigator in a state of relative ignorance is trying to find some relatively stable footing for his or her investigation. So perhaps one general sensible strategy is to pay particularly close attention to the little information, evidence, or knowledge that one does have. That initial information is, by hypothesis, insecure and uncertain -- but if it is more stable and secure and certain than anything else the investigator has, such partial and fragile evidentiary matter may be the right ground on which to stand, the right material in which to "anchor" investigative efforts -- at least until something better comes along.

But this notion of anchoring is a bit vague. It needs to be made more precise. If the investigator is to make any use of the notion of "anchoring evidentiary details," (s)he needs to appreciate the kinds of investigative directions or paths that evidentiary trifles might suggest. An investigator needs to have some notion of the types of questions that one characteristically needs to address and answer if one is to generate a sound answer to some question about some matter of fact. Without some sense of these things an investigator would not be able to appreciate what evidentiary trifles suggest.

The article "A Theory of Preliminary Fact Investigation" attempts to specify what some of those different kinds of matters are. But, in the meantime, let me state, quite dogmatically, that an investigator of factual issues associated with possible litigation needs to have knowledge of at least the following matters:

(i) law & legal requirements for liability,
(ii) time lines,
(iii) scenarios, (iv) identification of human and nonhuman sources of evidence,
(v) possibilities suggested by available evidentiary details,
(vi) credibility of human sources (witnesses),
(vii) probative force of all of forms of available evidence, taken together,
(viii) the preferences of persons such as the investigator's employer or client and the investigator's own judgments about the importance of the potential topics of investigation,
(ix) the extent of the available evidence and the possibility or probability of pertinent missing evidence, and
(x) the degree of the consistency or inconsistency of all of the foregoing.

These are some of the types of matters that an "anchoring evidentiary trifle" may suggest warrant further investigation.

But it is important to note that evidentiary trifles that are suggestive of some possible facts -- it is important to observe that such "anchoring" or "suggestive" evidentiary trifles do not produce unambiguous conclusions about what sort of investigative strategy makes the most sense under the circumstances.

Granted, one possible response to an uncertain proposition suggested by an evidentiary trifle is to investigate in an effort to make a better-grounded judgment about the proposition suggested by the trifle. This makes a certain amount of sense -- because, by hypothesis, the proposition suggested by the trifle at least has some discernible probability of being true -- and it might make sense for the investigator to make the judgment that he or she should put his or or her stakes on the proposition that has the best chance of holding up. However, it is equally sensible, I think, for an investigator who is in possession of a suggestive evidentiary trifle to be impressed by his or her greater ignorance about entirely different propositions that the investigator's intuitions suggest may also turn out to be important and, for that reason, to decide that, rather than attempting to shore up the proposition that has some evidential support, it is better to look for evidence about potentially important propositions about which there is (presently) less or no evidence. In short, the notion of an anchoring detail is a useful notion and it may help clarify thinking about the possible course of the early phases of an investigation -- but, by itself, the notion of an anchoring evidentiary trifle does not, I think, dictate the optimal course of an investigation.

In this connection -- in connection, that is, with the inability of evidentiary trifles to dictate the course of an investigation --, it is useful to consider that serendipity -- even in investigations -- is sometimes very useful, it is sometimes a productive investigative strategy. To wit: sometimes the best way of trying to determine how to pursue an investigation is to begin with undirected or random (investigative) activity -- i.e., investigative activity that does not (seem to) focus on any objective or hypothesis in particular, but involves a kind of "random walk," done with the idea of just seeing what might "pop up." See Pek van Andel & Daniele Bourcier, "Serendipity and Abduction in Proofs, Presumptions, and Emerging Laws," 22 Cardozo L. Rev. 1605 (2001). It is true (I suspect) that no useful investigative activity is completely serendipitous or entirely "random." But people can and do engage in relatively undirected (unfocused) investigation. For example, police often "cruise" through a neighborhood. This is done only partly because they wish to keep order by being visibly present in the neighborhood: police sometimes cruise (in a relatively random or undirected way) also because they (or their superiors) think that by doing that they may -- accidentally, so to speak -- see something interesting, they may "stumble across" indicia of criminal activity. Serendipitous investigation, in short, is sometimes a very valuable form of investigation -- particularly since it (practically by definition!) has less of a tendency to foreclose potentially fruitful possibilities. (But, like any other strategy in the face of uncertainty, this investigative strategy also involves significant costs and risks. {And what, pray tell, are these?})

Tillers' Home Page

Fact Investigation Course Home Page