29

CHAPTER 5

Retrospective Prophecies
Legal Narrative Constructions
PETER BROOKS

I currently teach an undergraduate course called “Clues, Evidence,

Detection: Law Stories.” It does not aim to be a “pre-law” course, or to
claim any true legal expertise—yet nor is it a “law in popular culture” kind
of course (of which there are plenty of examples), since I do want to teach
my students something about how “the law” goes about thinking about
certain kinds of problems, in this case ones concerned largely with criminal
procedure and the use of evidence. Discussion of these legal matters leads
me into narrative analysis: the way in which the law understands discovery
and recites the meanings derived from evidence raises problems in narra-
tive telling of the sort studied by some narratologists: an understanding
of how narrative presentation shapes events recounted, for instance, and
of how anticipated ends shape story events. For, after all, the narratives
presented in law as well as in literature are not themselves events in the
world but rather the way we speak events, the way we give them significant
order—a distinction too easily forgotten. Here there might be a productive
meeting of legal and literary traditions.
I have long thought that the place and status of narrative in the law, and
in legal studies, are strangely uncertain and ambiguous. On the one hand,
trial advocates know—have known, presumably, since antiquity—that
success in the law court depends upon telling an effective and persuasive
story. The discipline of rhetoric originated essentially to teach courtroom
3
9

practitioners how to do just that. Law schools courses on advocacy and

legal writing necessarily pay attention to narrative skills, urging the impor-
tance of “controlling the narrative,” though such courses are not generally
thought to be central to the legal curriculum. Academic study sympathetic
to “law and literature” has by now given considerable attention to narra-
tive and its uses throughout the law, as institution and as praxis.1 Still, one
looks in vain within legal doctrine, and in judicial opinions, for any explicit
recognition that “narrative” is a category for adjudication, that it is some-
thing that needs to be talked about and brought to bear analytically: that
rules of evidence, for instance, implicate questions of how stories can and
should be told. Not so long ago, Justice David Souter evoked a concept of
“narrative integrity” in one of his Supreme Court opinions—so far as I can
tell, the first overt recognition that the literary and cultural category of
narrative needs to be imported into legal thinking, and one that thus far
has had little in the way of sequel.2
The detective story is the most literary of genres (and the one most often
written by academics) for a reason: its processes of investigation and detec-
tion touch on the very rationale of narrative; it very often gives narratives
that speak to and exemplify the narrative process, or what is sometimes
called “narrativity.” To the extent that this narrative process is about dis-
covery and the creation of a meaningful sequence, it touches closely upon a
subject at the center of criminal justice. So before returning to generalities,
let me try an example:

“I knew we should find a ventilator before ever we came to Stoke Moran.”

“My dear Holmes!”
“Oh, yes, I did. You remember in her statement she said that her sister could
smell Dr. Roylott’s cigar. Now, of course that suggested at once that there must
be a communication between the two rooms. It could only be a small one, or it
would have been remarked upon at the coroner’s inquiry. I deduced a ventilator.”3

Some readers may have recognized this exchange, from Arthur Conan
Doyle’s The Adventure of the Speckled Band. The ventilator is crucial since
it suggests a communication between Dr. Grimsby Roylott’s bedroom and
that in which Julia Stoner died of mysterious causes. Add to the ventilator
a dummy bell pull that hangs down from it to a bed that itself is clamped
to the floor. And then find in Dr. Roylott’s adjacent bedroom an iron safe, a
bowl of milk, a horsewhip tied back to itself, and footmarks on the seat of
a chair under the other end of the ventilator. Now you have a bridge along
which the loathsome swamp adder can travel, from its home in the safe
through the ventilator down the bell pull to the immovable bed.

R E T R O S P E C T I V E P R OP H E C I E S [ 93 ]
4
9

Such moments are repeated throughout the Sherlock Holmes corpus.

“I think we may safely deduce a cab,” says Holmes after looking over the
layout of the rooms, corridors, and staircases in the Foreign Office, in The
Naval Treaty, all covered in a “creamy linoleum” that would have shown
marks of wet boots on this rainy night had the thief come on foot. But what
Holmes calls “deduction” really fits better into the discovery procedure
that C. S. Peirce called “abduction,” often glossed as “inference to the best
explanation.” If for most contemporary scientists abduction belongs to the
“context of justification,” for Peirce it was explicitly part of the context of
discovery: “Abduction is the process of forming explanatory hypotheses,”
he wrote. “It is the only logical operation which introduces any new idea.”4
A perfect Holmes example comes in Silver Blaze, where he says of the wax
vesta match he finds buried in the mud at the crime scene: “It was invisible,
buried in the mud. I only saw it because I was looking for it.” You would not
have found it without first hypothesizing that it had to be there, since the
criminal would have had to strike a light to perform his dastardly act of
wounding the race horse. The inferential process of abduction interests me,
not so much on philosophical grounds, but because it seems to characterize
a form of legal narrative reasoning that finds its hyperbolic statement in
the doctrine of “inevitable discovery.”
I have written about “inevitable discovery” elsewhere.5 In very brief defi-
nition, it is a doctrine that allows the prosecution to use results of an illegal
search, one in violation of the Fourth Amendment rules on searches and
seizures, if it can show that what was found would inevitably have been
discovered if legal means had been used to discover it. This makes pretty
good sense in the case in which the Supreme Court signs on to the doctrine,
Nix v. Williams, in which there was a search party on the ground, marching
westward through Iowa, some two and a half miles from the place where
the murdered child lay—a search called off when the perpetrator told the
police where the body was after an interrogation judged to be illegal.6 In
other cases, it has at times been a convenient way of saying that the find-
ing of the goods (drugs, very often) by whatever means is justified because
if the correct, warranted means had been used you would certainly have
found it.7 This sounds to me very much like the logic of a certain kind of
narrative explanation that derives beginnings and middles from ends. It is
the inferential logic of the: “it must have been like this.” It may be the logic
of telling rather than of happening, and the law can at times confuse the
two, turning the way events are told into the way things happened. I want
to pursue and refine this question by way of Carlo Ginzburg’s ambitious
essay on “Clues.”

[ 94 ] Methods
5
9

SEARCH AND THE HUNTSMAN’S PARADIGM

In a sinuous argument that touches on Sherlock Holmes, on Freud, and

on the prototype of a kind of discovery procedure used by both that was
elaborated by the art historian Giovanni Morelli—whose premise was that
in order to authenticate a painting, one should look to minute details such
as ear lobes and fingernails, where an artist’s unique characteristics would
be better revealed than in the ensemble—Carlo Ginzburg undertakes to
isolate and define a special form of cognition by way of clues.8 Knowing
by way of clues—following the traces left by one’s quarry—is of course
the detective’s method. It doesn’t work by deduction from a general law
(though it may call upon fragments of general wisdom, e.g., “the horse is a
gregarious animal”), nor does it quite work inductively from part to whole.
It is rather a science of the concrete and particular that achieves its discov-
eries through putting particulars together in a narrative chain.9 Ginzburg
identifies this science with the huntsman’s lore:

Man has been a hunter for thousands of years. In the course of countless pur-
suits he learned to reconstruct the shapes and movements of his invisible prey
from tracks in the mud, broken branches, droppings of excrement, tufts of hair,
entangled feathers, stagnating odors. He learned to sniff out, record, interpret,
and classify such infinitesimal traces as trails of spittle. He learned how to exe-
cute complex mental operations with lightning speed, in the depth of a forest or
in a prairie with its hidden dangers. (Spie, 166; Clues, 102)

Even in a post-hunting society, searches reach their discoveries by such

tracking of details, making them into a chain of meaning, uncovering their
connections. Ginzburg speculates that this kind of knowing may in fact lie
at the inception of narrative itself:

This knowledge is characterized by the ability to move from apparently insignifi-

cant experiential data to a complex reality that cannot be experienced directly.
And the data is always arranged by the observer in such a way as to produce a
narrative sequence, which could be expressed most simply as “someone passed
this way.” Perhaps the very idea of narrative (as distinct from the incantation,
exorcism, or invocation) was born in a hunting society, from the experience of
deciphering tracks. (Spie, 166; Clues, 103)

On Ginzburg’s hypothesis, narrative would be a cognitive instrument of

a specific type, one “invented” for the decipherment of details of the real

R E T R O S P E C T I V E P R OP H E C I E S [ 95 ]
6
9

that take on their meaning only when linked in a series, enchained in a

manner that allows one to detect that “someone passed this way.” This is
what Sherlock Holmes’s searches—for a wax vesta in the mud, for a ven-
tilator high up in the wall—are all about. And the “huntsman’s paradigm”
may indicate in more general terms the use-value of narrative as a form of
speech and cognition: it is the instrument we use when the putting together
of particulars into a meaningful sequence seems to be the only way to track
down our quarry, whatever it may be. Working from Ginzburg’s sugges-
tions, Terence Cave argues that the huntsman’s or “cynegetic paradigm”
points us toward that most basic and enduring and useful of plots: the story
that leads to anagnorisis or recognition. “The sign of recognition in drama
and narrative fiction belongs,” says Cave, “to the same mode of knowledge
as the signature, the clue, the fingerprint or footprint and all the other
tracks and traces that enable an individual to be identified, a criminal to
be caught, a hidden event or state of affairs to be reconstructed.”10 Signs
of recognition in literature reach back to antiquity and forward to moder-
nity: see the scar on Odysseus’s thigh that enables his old nurse Eurykleia
to recognize him by touch, see the hidden birthmark of Shakespeare’s
Cymbeline, see the notorious la croix de ma mère of nineteenth-century
melodrama, the token which at the denouement allows the orphan to be
recognized, true identities established. It is easy to recognize that the law,
particularly when dealing with issues of evidence, must make use of the
huntsman’s paradigm, seeking to show how finding signs and deciphering
tracks will lead to the apprehension of what passed that way.
Fingerprinting, from early in the twentieth century onward, assumed
dominant importance in the tracking and identification of criminals. It
was imported by the British raj from Bengal and immediately accepted in
Europe and America:  other systems of identification, including the com-
plex measurements of “Bertillonage,” were never wholly reliable. It is inter-
esting, in the context of Ginzburg’s paradigm, that a decision by federal
judge Louis Pollak in 2002, in United States v. Llera Plaza, cast doubt on our
dependence on fingerprinting.11 Pollak ruled that fingerprint evidence does
not meet the criteria for scientific evidence established in Daubert v. Merrill
Dow Pharmaceuticals. He noted that other recent federal cases called fin-
gerprint identification “the very archetype of reliable expert testimony”
and “scientific knowledge,” but that it failed on the grounds of testability
and especially falsifiability: it is the test of falsifiability that allows one to
know not only that a proposition is true a good deal of the time but that
it is universally true. Fingerprint identification, while mustering a con-
siderable body of expertise, does not in the final analysis meet this stan-
dard. The final step in fingerprint identification—the determination of a

[ 96 ] Methods
7
9

match between two sets of fingerprints—involves a subjective judgment

rather than a scientific procedure: “it is a subjective determination without
objective standards to it” (at 37–38, citing expert witness David Stoney).
Therefore, Pollak ruled that experts may present analysis of fingerprints,
and point out observed similarities between prints, but “will not be per-
mitted to present testimony expressing an opinion of an expert witness
that a particular latent print matches, or does not match, the rolled print
of a particular person and hence is, or is not, the fingerprint of that person”
(at 69). This I  think had the effect of moving fingerprint evidence from
the domain of science to the realm of Ginzburg’s “conjectural knowledge.”
But Pollak’s decision inevitably met with ferocious resistance in the police
and prosecutorial communities, and he reversed himself three months
later. The absolute accuracy of the clues registered by our digital imprints
remains in doubt.12
Ginzburg notes an early relation of the huntsman’s paradigm to law in
his discussion of Mesopotamian divination, based on the minute investi-
gation of seemingly trivial details:  “animals’ innards, drops of oil on the
water, stars, involuntary movements of the body.” According to Ginzburg,
Mesopotamian jurisprudence was similarly oriented toward the interpreta-
tion of particulars: “Mesopotamian legal texts themselves did not consist
of collections of laws or statutes but of discussions of concrete examples”
(Spie, 168–69; Clues, 104). So that the same paradigm can be found in the
divinatory and jurisprudential texts, with this difference that the former
are directed to the future, the latter to the past. Ginzburg then stretches his
hypothesis to suggest that narrative modes of knowing (such as archaeol-
ogy, paleontology, geology) all make what he calls “retrospective prophecies”
(183; 117), which he sees as the key to the popularity of detective fiction.13
What Ginzburg calls the “conjectural paradigm” means that when “causes
cannot be reproduced, there is nothing to do but to deduce them from their
effects.” That working backward from effects to causes characterizes psy-
choanalysis in its entirety, as Freud was very much aware, and possibly
much legal analysis as well. The “case method” of American legal study—
introduced by C. C. Langdell at Harvard Law School shortly before Conan
Doyle began his Sherlock Holmes tales—resembles the Mesopotamian
approach in its insistence that argument be worked up from concrete par-
ticulars.14 And here, too, the concept of “retrospective prophecy” is rel-
evant: that which is plotted forward to the predictable outcome can be so
ordered because one in fact stands at the point of the outcome. The causes
become apparent in light of their effects.15 Langdell of course thought the
study of cases ought to lead to a legal science. Holmes suggests the same
thing in a late story, The Adventure of the Abbey Grange, when he complains

R E T R O S P E C T I V E P R OP H E C I E S [ 97 ]
8
9

that Watson’s “fatal habit of looking at everything from the point of view of
a story instead of as a scientific exercise has ruined what might have been
an instructive and even classical series of demonstrations.”16 He querulously
proposes “to devote my declining years to the composition of a textbook,
which shall focus the whole of detection into one volume.” His reproach to
Watson misunderstands—and allows his creator to emphasize—the genius
of the case method. The point of the exercise, in a pedagogical and cogni-
tive sense, is to retrace how the outcome can be drawn, with the logic of
inevitability, from the “facts of the case” by the plotting of the story. And if
we enjoy the mental processes activated by detective fiction and legal argu-
ment, it must be in part because of the satisfaction derived from the dem-
onstration of inevitability: it had to be this way, and no other way.
Searches for evidence may always include a “retrospective prophecy” fac-
tor. Consider that application for a search warrant must contain a predic-
tion of what is to be found. The warrant application sets forth the evidence
that the police believe they (inevitably) will find if given permission to
search.17 Warrants must be based on “probable cause” that what is sought
will be found. In this sense, searches for evidence always involve a prior
story, a probable, hypothetical story which the search intends to confirm.
The doctrine of “inevitable discovery” offers a particularly clear instance
of “retrospective prophecy.” It makes the claim that a trail to the quarry
exists, and that the (hypothetical) following of the traces and tracks mak-
ing up this trail would (certainly) lead to the quarry. In other words, it takes
the logic of the huntsman’s paradigm—the logic of narrative knowing—
and, in its hypothetical application of the paradigm to a case in which the
quarry was not but would have been found, exposes the logic of discovery as
a narrative process. In the doctrine of inevitable discovery, we know that
the quarry is there, at the end of the trail. The question is whether follow-
ing the trail would inevitably have led to it. When you decide—as in Nix v.
Williams—that it would have, you sign on to the logic of narrative discov-
ery in a particularly telling way, accepting that the huntsman’s lore is infal-
lible, and infallibly cognitive. When as a legal decision maker you so decide,
you may be simply affirming the nature of the law as discipline: affirming
its belief in evidence as the meaningful entailment of tracks and traces.

NARRATIVE RETROSPECT

All the practices of identification by way of signs interpreted as clues in the

narrative of what happened, who passed by, involve a “retrospective proph-
ecy,” a construction of the story of the past by way of its outcome, what it

[ 98 ] Methods
9

was leading to. It is in the peculiar nature of narrative as a sense-making

system that clues are revealing, that prior events are prior, and causes are
causal only retrospectively, in a reading back from the end. The narrative
theorist Gérard Genette argues that narrative offers the determination of
means by ends and of causes by effects. If the narrative went nowhere,
never became a complete story, there would be no decisive enchainment
of its incidents, no sense of inevitable discovery; the units of the narrative
would cease to be functional. Such, Jean-Paul Sartre argued, is the differ-
ence between living and telling. To tell is to conceive life as adventure, in
the etymological sense of the ad-venire, that which is to come, and by its
coming to structure what leads up to it. It is worth quoting at some length
the reflections of Sartre’s fictional spokesman Antoine Roquentin on the
problem. When you begin to tell a story, you appear to start at the begin-
ning. But, says Roquentin:

In reality you have started at the end. It is there, invisible and present, it is what
gives these few words the pomp and value of a beginning: “I was out walking,
I had left the town without realizing, I was thinking about my money troubles.”
This sentence, taken simply for what it is, means that the guy was absorbed,
morose, a hundred miles from an adventure, exactly in a mood to let things hap-
pen without noticing them. But the end is there, transforming everything. For
us, the guy is already the hero of the story. His moroseness, his money troubles
are much more precious than ours, they are all gilded by the light of future pas-
sions. And the story goes on in reverse: instants have stopped piling themselves
up in a haphazard way one on another, they are caught up by the end of the story
which draws them and each one in its turn draws the instant preceding it: “It
was night, the street was deserted.” The sentence is thrown out negligently, it
seems superfluous; but we don’t let ourselves be duped, we put it aside: this is a
piece of information whose value we will understand later on. And we feel that
the hero has lived all the details of this night as annunciations, as promises, or
even that he lived only those that were promises, blind and deaf to all that did
not herald adventure. We forget that the future wasn’t yet there; the guy was
walking in a night without premonitions, which offered him in disorderly fash-
ion its monotonous riches, and he did not choose.18

On this statement, any narrative telling presupposes an end that will trans-
form its apparently random details “as annunciations, as promises” of what
is to come, and that “what is to come” transforms because it gives meaning
to, makes significant the details as leading to the end.
Roland Barthes once suggested that narrative may be built on a general-
ization of the philosophical error of “post hoc, ergo propter hoc”: narrative

R E T R O S P E C T I V E P R OP H E C I E S [ 99 ]
01

plotting makes it seem that if b follows a it is because b is somehow logi-

cally entailed by a.19 And certainly it is part of the “logic” of narrative to
make it appear that temporal connection is also causal connection. This
indeed may be one of the uses of narrative: we need to be able to discover
connections in life, to have it make sense, to rescue passing time from
meaningless successivity. One of the projects of complex narratives—such
as novels—has often been to question such connections, to ask about the
possible randomness of existence. Novels often appear to stage a struggle
between chaos and meaning. But their very existence as novels, as writing
about life rather than life itself, must generally assure that they conclude,
however tenuously, in favor of meaning.
In the inevitable discovery doctrine, the law comes down firmly on the
side of meaning, conjuring away the specter of meaninglessness, a chaotic
universe in which searches would not necessarily lead to anything. It pre-
supposes an infinitely knowable world, one laid out in tracks and traces—
like the gridlines marked off on the county map by Agent Ruxlow in his
search for the young girl’s body in Nix—waiting to be deciphered. If this is
a contestable picture of the world, it may be an accurate picture of the law,
which assumes that its quarry exists, and that its discovery procedures, if
patient and thorough enough, will find it. In the doctrine of inevitable dis-
covery, then, the law is merely affirming (in fairly spectacular form) its own
nature. By this, I mean that law is inextricably bound up with the making of
“retrospective prophecies,” using the shaping force of outcomes to construct
narratives from tracks and traces into a coherent plot with a meaningful
ending. Inevitable discovery is in this sense what the Russian Formalists
might have called a “laying bare of the device”: one of those moments that
images the procedures and the very nature of the text in question.
When we speak of “the narrative construction of reality”—in Jerome
Bruner’s terms, how narrative “operates as an instrument of mind in the
construction of reality”20—we must mean, among other things, the ways
in which narrative sequence, plot, and intelligibility are used by humans
to make sense of their lives and their world. It was precisely his reflection
on the workings of narrative structure in the creation of intelligibility and
meaning in human action (a reflection continued in his autobiography, The
Words) that led Sartre eventually to renounce the novel as a genre, since it
came to appear to him a violation of existential freedom, a misrepresenta-
tion of the open-endedness of becoming. Yet one might respond that the
renunciation of narrative is not an option, since narrative construction of
reality is a basic human operation, learned in infancy, and culturally omni-
present.21 For better or worse, we are stuck with narrative and its ways

[ 100 ] Methods
0
1

of making sense. The conclusion would then seem to be that we should

become better narratologists—better analysts of the stories we tell, the
ways they work, the effects they have.
Bruner notes that the way the human mind processes knowledge as
story “has been grossly neglected by students of mind raised either in the
rationalist or in the empiricist traditions.”22 One wonders if it has been
neglected as well by students of the law. One tends to find attention to the
creation and critique of legal storytelling mainly consigned to the courses
on courtroom advocacy and legal writing, which often have marginal sta-
tus within law school. And, as I suggested earlier, one scans legal opinions
in vain for any mention of narrative as a category that needs thinking
about.23 Justice Souter’s riff on “narrative integrity” in a case called Old
Chief v.  United States remains exceptional. It is the only moment I’ve
encountered where the Court considers what has been called the “episte-
mology of the particular”: the way in which stories create certain mean-
ings that cannot otherwise be conveyed.24 At issue in the case is whether
defendant Johnny Lynn Old Chief can “stipulate” to a prior crime and con-
viction, thereby preventing the prosecution from presenting the details of
that old crime, which is very similar to the new charge. Souter registers
the need for “evidentiary richness and narrative integrity in presenting
a case.”25 He goes on to say that “making a case with testimony and tan-
gible things … tells a colorful story with descriptive richness.”26 And he
continues:

Evidence thus has force beyond any linear scheme of reasoning, and as its pieces
come together a narrative gains momentum, with power not only to support
conclusions but to sustain the willingness of jurors to draw the inferences, what-
ever they may be, necessary to reach an honest verdict. This persuasive power of
the concrete and particular is often essential to the capacity of jurors to satisfy
the obligations that the law places on them.27

Souter’s opinion develops into a striking reflection on the nature and the
force of narrative:

A syllogism is not a story, and a naked proposition in a courtroom may be

no match for the robust evidence that would be used to prove it. People who
hear stories interrupted by gaps of abstraction may be puzzled at the missing
chapters. … A convincing tale can be told with economy, but when economy
becomes a break in the natural sequence of narrative evidence, an assurance
that the missing link is really there is never more than second best.28

R E T R O S P E C T I V E P R OP H E C I E S [ 101 ]
02
1

For Souter, the full and convincing tale must be responsive to the jurors’
expectations “about what proper proof should be.”29 These expectations
preexist service on a jury: jurors bring them to the courthouse, “assuming,
for example, that a charge of using a firearm to commit an offense will be
proven by introducing a gun in evidence.” Failure to meet what we might
call jurors’ “narrative competence” can undermine the prosecution’s proof.
The need for a complete and convincing narrative indeed has an ethical
dimension for Souter. Pointing out that jury duty is unsought and often
difficult:

When a juror’s duty does seem hard, the evidentiary account of what a defen-
dant has thought and done can accomplish what no set of abstract statements
ever could, not just to prove a fact but to establish its human significance, and
so to implicate the law’s moral underpinnings and a juror’s obligation to sit in
judgment.30

As I  read this, Souter seems to be saying that the force of narrative evi-
dence, the kind of evidence of guilty mind and act that it alone can provide,
may be tied in a deep way to the very act of passing judgment on one’s fel-
lows. It is only on the basis of seeing and feeling what someone did in com-
mitting a crime that we can understand the morality of passing judgment.
If that is what Souter means, he implies that law without narrative would
be not only disabled but lacking in moral force.
Curiously, though, this major statement on the importance of narra-
tive in the law is used in the instant case to claim that while a jury needs
to hear a full narrative of events, too much story, for instance that of Old
Chief’s past felony and conviction, could “overpersuade” the jury. That
might lead the jury to convict on evidence of his bad character, rather than
on proof specific to the crime. Old Chief’s past felony, Souter decides, is not
a chapter in this story: it is “entirely outside the natural sequence of what
the defendant is charged with thinking and doing to commit the current
offense.”31 Story is dangerous as well as persuasive. The law needs to keep
it channeled or, you might say, under erasure: visible when you look for it
but only barely legible. If it is in some way fundamental to the morality of
criminal law, to recognize its constitutive force might be to concede that
law rests on rhetoric.
Certainly where Fourth Amendment jurisprudence is concerned—when
we are talking about searches and seizures and how we understand their
workings in relation to constitutional “rules”—the narrative construction
of the reality is the reality, and how it is constructed makes all the differ-
ence in the defendant’s story. And when we start probing the interesting

[ 102 ] Methods
03
1

piece of Fourth Amendment doctrine known as “inevitable discovery,” we

find implicated within it a larger problem of legal narrative, which is in
turn a problem of narrative as a human function, cognitive instrument,
and moral compass.
Stories are not events in the world, but rather a way in which we speak
the world, and in so doing give it shape and meaning. Yet in their telling
and their reception as the given story of what happened, they become
events in the world. This issue extends far beyond Fourth Amendment
cases. Consider the obvious instance of rape. I have taught the famous case
from Baltimore, Rusk v. Maryland and Maryland v. Rusk (recently the object
of a fine essay by Jeannie Suk32), where the “same story” is retold four
times, by majority and dissent at two different appellate levels, with cru-
cially differing outcomes. Rusk was convicted at trial; the Maryland Court
of Special Appeals reversed and vacated his conviction; then the Maryland
Court of Appeals (the highest court in the state) reversed again, reinstating
the conviction. In both appeals courts, there was a majority opinion and a
dissent: the story is retold four times, twice as “consensual sex,” twice as
“rape.” “What happened” was never in dispute. What it meant made all the
difference—and that meaning was all in the way the happening was told,
with different narrative connectives, with a differing understanding of how
events fit together to make the chain of meaning. And it is easy to see that
in all branches of the law, from trial court to Supreme Court, the telling of
“the facts” underlies all adjudication. If the need to “control the narrative”
has become something of a cultural cliché, that should suggest the need to
unpack and analyze the prevailing narrative.33
One more example (if that’s needed) could be found in the writing of
postconviction petitions, which must claim that a story officially pro-
claimed at its end should be reopened. In Mickens v. Taylor, which reached
the Supreme Court as a case of conflicted representation (counsel who
had been representing the murder victim on a charge of assault against
his mother was, immediately after the murder, appointed to represent his
alleged killer), Anthony Amsterdam had to produce a petition for a writ
of certiorari that would forcefully suggest from the outset that the record
compiled at trial by Mickens’s counsel could have and would have been
different with unconflicted legal representation.34 The lawyer’s conflict
of interest was the legal entering wedge; the more compelling issue was
that the lawyer had allowed Walter Mickens to take the stand and deny
any involvement with the killing although he had already confessed to it
and the evidence against him was overwhelming. Mickens was convicted of
aggravated homicide, murder made “wantonly vile” by forcible sodomy on
the victim, and sentenced to death. Amsterdam’s petition retold the story

R E T R O S P E C T I V E P R OP H E C I E S [ 103 ]
0
4
1

to suggest forcefully that the sexual encounter of Mickens and his victim
was consensual, occurring in a well-known trysting place, and that the vic-
tim was indeed probably a male prostitute, thus disqualifying Mickens for
a capital sentence in the Commonwealth of Virginia.35 All the other courts
below had accepted the story created at trial. Amsterdam effectively dem-
onstrated that it could be told otherwise, and with greater plausibility. That
argument took the case to the Supreme Court, where a five–four major-
ity decided that Mickens had to demonstrate that his legal representation
had an actual adverse effect on his trial—a “but for” unprovable by its very
nature. As Justice Stevens puts it in his dissent, “we will never know” what
the story made at trial might have been (at 186). The stories told at law,
once accepted by juries and judges and then retold and ratified by appellate
courts, become inert realities, rarely revised.
Courtroom advocates, appellate judges—including justices of the
Supreme Court—“know” in some unacknowledged way that there are
moments when the particulars of how “the facts” are told make all the dif-
ference. There are times when they would subscribe to Joseph Conrad’s
famous dictum for novelists: “My task … is by the power of the written
word, to make you hear, to make you feel—it is, before all, to make you see.
That—and no more, and it is everything.”36 But an explicit awareness of
the crucial, sometimes decisive role of narrative in passing judgment rarely
is visible, and almost never discussed. If the law regularly issues “retro-
spective prophecies”—reaching back to show how the present outcome was
necessarily entailed by happenings and decisions in the past—it doesn’t
recognize the kind of narrative act involved in this reasoning. For instance,
an acknowledgment that usually there are no facts of a search independent
of the narrative form given to them might seem to be an important ele-
ment in the intellectual toolkit judges bring to the resolution of appellate
cases. The narrative construction of reality as an explanatory system surely
is omnipresent, and it needs to be recognized as a construction rather than
accepted as one of the givens of the real. Narrative analysis needs to take a
place within legal analysis.

NOTES

1. See, e.g., Guyora Binder and Robert Weisberg, Literary Criticisms of Law
(Princeton, NJ: Princeton University Press, 2000); Anthony G. Amsterdam and
Jerome Bruner, Minding the Law (Cambridge, MA: Harvard University Press
2000); Peter Brooks and Paul Gewirtz, Law’s Stories: Narrative and Rhetoric
in the Law (New Haven, CT: Yale University Press 1996); Richard Posner, Law

[ 104 ] Methods
05
1

and Literature: A Misunderstood Relation (Cambridge, MA: Harvard University

Press 1998).
2. See Old Chief v. United States, 519 U.S. 172, 183 (1997). I have also discussed
Old Chief and some of the rare cases that cite it in Peter Brooks, “Law and
Humanities: Two Attempts,” Boston University Law Review 93 (2013): 1437.
3. Arthur Conan Doyle, The Complete Sherlock Holmes (New York: Gramercy, 2002),
417. My thanks to Georgetown Law student Elizabeth Sebesky for her fine
research assistance on this chapter.
4. Charles Sanders Peirce, “Three Types of Reasoning,” in Collected Papers of Charles
Sanders Peirce, ed. Charles Hartshorne and Paul Weiss (Cambridge, MA: Harvard
University Press, 1965), vol. 5, §172. Peirce continues: “Deduction proves that
something must be; [i]nduction shows that something actually is operative; [a]
bduction merely suggests that something may be”; ibid., vol. 5, §171. See also
the seminal discussions in Umberto Eco and Thomas A. Sebeok, eds., The Sign
of Three: Dupin, Holmes, Peirce (Bloomington: Indiana University Press, 1984):
Umberto Eco, “Horns, Hooves, Insteps: Some Hypotheses on Three Types of
Abduction,” 198–220, and Thomas A. Sebeok and Jane Umiker-Sebeok, “ ‘You
Know my Method’: A Juxtaposition of Charles S. Peirce and Sherlock Holmes,”
11–54.
5. See Peter Brooks, “‘Inevitable Discovery’: Law, Narrative, Retrospectivity,” Yale
Journal of Law and the Humanities 15 (2003): 101–29. Also Peter Brooks, “Clues,
Evidence, Detection: Law Stories,” Narrative 25:1 (January 2017).
6. See Nix v. Williams, 467 U.S. 431 (1984).
7. See for instance United States v. Feldhacker, 849 F.2d 293, 296 (8th Cir. 1988);
United States v. Andrade, 784 F.2d 1431, 1433 (9th Cir. 1986); United States
v. Levasseur, 620 F. Supp. 624, 632 (E.D.N.Y. 1985); State v. Butler, 676 S.W.2d
809, 813 (Mo. 1984). On these and a number of other cases, see Jessica Forbes,
“The Inevitable Discovery Exception, Primary Evidence, and the Emasculation
of the Fourth Amendment,” Fordham Law Review 55 (1987): 1221–38. It is
important to note that United States v. Levasseur was abrogated by United
States v. Heath, 455 F.3d 52 (2d Cir. 2006), which created a heightened standard
for the government to meet in order to rely on the inevitable discovery doctrine.
The Second Circuit in that case expressly eschews the “reasonable probability”
standard, and states: “we conclude that illegally-obtained evidence will be
admissible under the inevitable discovery exception to the exclusionary rule
only where a court can find, with a high level of confidence, that each of the
contingencies necessary to the legal discovery of the contested evidence would
be resolved in the government’s favor.” Heath, 455 F.3d at 61.
8. Carlo Ginzburg, “Spie: Radici di un paradigma indizario,” in Miti Emblemi Spie
(Torino: Einaudi, 1986), 158–209; “Clues: Roots of an Evidential Paradigm,” in
Myths, Emblems, Clues, trans. John Tedeschi and Anne C. Tedeschi (London:
Hutchinson Radius, 1990), 96–125. I have modified the Tedeschi translation in
places in order to give a more literal rendition.
9. The method identified by Ginzburg here resembles what C. S. Peirce called
“abduction,” “hypothesis,” or “guessing.”
10. Terence Cave, Recognitions: A Study in Poetics (Oxford: Clarendon Press,
1988), 250.
11. The first Pollak opinion comes in United States v. Llera Plaza, 179 F. Supp. 2d
494 (E.D. Pa. 2002), vacated by 188 F. Supp. 2d 549 (E.D. Pa. 2002); see also

R E T R O S P E C T I V E P R OP H E C I E S [ 105 ]
06
1

Daubert v. Merrell Dow Phamaceuticals, Inc. 509 U.S. 579 (1993) (subsequent
page references are to the first Pollak opinion).
12. For a thorough study of the current situation of fingerprinting and similar
forensic evidence, see Jennifer L. Mnookin et al., “The Need for a Research
Culture in the Forensic Sciences,” UCLA Law Review 58 (2011): 725–80. The
best book on the history of fingerprinting that I know is Simon A. Cole, Suspect
Identities (Cambridge, MA: Harvard University Press, 1986).
13. The Tedeschi translation says: “the ability to forecast retrospectively,” which
I think downplays the striking paradox of Ginzburg’s “retrospective prophecy,”
which seems to me a key part of his argument.
14. I am grateful to Simon Stern for bringing this parallel to my attention.
15. I have argued that this is true also of the way in which legal opinions recount
the history of the law as applied to the case at hand; this may be especially true
when the history of constitutional interpretation is at stake: the result must
seem the inevitable conclusion drawn from prior cases. See Brooks, “Law and
Humanities,” 1456–63.
16. “The Adventure of the Abbey Grange,” in Arthur Conan Doyle, Sherlock
Holmes: The Complete Novels and Tales (New York: Bantam Dell, 1986), 1:1010.
17. See, e.g., the standard form for a warrant in federal court: http://www.
uscourts.gov/forms/law-enforcement-grand-jury-and-prosecution-forms/
search-and-seizure-warrant.
18. Jean-Paul Sartre, La Nausée (Paris: Gallimard, 1947), 59–60. My translation.
See also Gérard Genette, “Vraisemblance et motivation” [“Likelihood and
Motivation”] in Figures II (Paris: Editions du Seuil, 1969), 71–100.
19. Roland Barthes, “Introduction to the Structural Analysis of Narrative” (1966), in
The Barthes Reader, ed. Susan Sontag (New York: Hill and Wang, 1982), 266.
20. See Jerome Bruner, “The Narrative Construction of Reality,” Critical Inquiry 18
(1991), 6.
21. Alan Dershowitz makes an argument against narrative construction as
creating juror expectations of a tidy narrative whereas life is largely chaotic: see
Dershowitz, “Life Is Not a Dramatic Narrative,” in Brooks and Gewirtz, Law’s
Stories, 99–105. But such construction seems to me inevitably part of our
understanding of events. See also the comment on the recent sensational
television documentary The Jinx (containing the apparent confession to
murders by Robert A. Durst) by documentarian Joe Berlinger: “Real life doesn’t
necessarily mirror the arc of scripted drama, and yet there has been this push in
television to bring those two kinds of storytelling together”; Jonathan Mahler,
“Irresistible TV, but Durst Film Tests Ethics Too,” New York Times, March 16,
2015. Will TV crime drama in turn inflect legal procedure?
22. Bruner, “Narrative Construction of Reality,” 8.
23. Although mentions of narrative integrity as such are rare in jurisprudence, the
idea that parties should be able to tell and juries to hear a complete story is more
commonly iterated. See, e.g., Lakeside v. Oregon, 435 U.S. 333, 339–41 (1978)
(cited by Souter in Old Chief for the proposition that a nontestifying defendant
may demand that a jury be instructed not to draw any adverse inference from his
silence based on the propensity of jurors to fill in the gaps of the story if certain
evidence is withheld); United States v. Pepin, 514 F.3d 193, 207–208 (2d Cir.
2008) (citing Old Chief for the premise that the lower court abused its discretion
in excluding evidence that the defendant dismembered the body after committing
homicide because these details “form[ed] part of the res gestae, the narrative

[ 106 ] Methods
07
1

that the government rightly seeks to tell at the guilt phase of a trial”); United
States v. Hill, 249 F.3d 707, 713 (8th Cir. 2001) (citing Old Chief to hold that past
criminal evidence can constitute the “story of defendant’s criminal behavior” as
long as it goes to the intent to commit the present crime); United States v. Rezaq,
134 F.3d 1121, 1142 (D.C. Cir. 1998) (applying narrative integrity analysis of Old
Chief to stand for the right of defendants generally to receive evidence in the
form of actual classified governmental documents rather than a summary). See
also United States v. Saunders, 209 Fed. App’x 778, 782 (10th Cir. 2006); United
States v. Gartmon, 146 F.3d 1015, 1021 (D.C. Cir. 1998); United States v. Inserra,
34 F.3d 89–90 (2d Cir. 1994); United States v. Fortenberry, 971 F.2d 717, 721
(11th Cir. 1992); United States v. Storm, 915 F. Supp. 2d 1196 (D. Or. 2012) (citing
Old Chief); Michael S. Pardo, “Juridical Proof, Evidence, and Pragmatic Meaning:
Toward Evidentiary Holism,” Northwestern University Law Review 95 (2000): 399–
442; Nancy Pennington and Reid Hastie, “A Cognitive Model of Juror Decision
Making: The Story Model,” Cardozo Law Review 13 (1991): 519–58; Philip Meyer,
Storytelling for Lawyers (Oxford: Oxford University Press, 2014).
24. One might find some similarities between Souter’s dicta on narrative integrity and
the common law doctrine of res gestae, which has been defined in various ways,
most recently as the “inextricable intertwining” of evidence of one crime with
the evidence of other crimes, and which was laid to rest in federal jurisprudence
by the Seventh Circuit prior to Souter’s opinion. Stephens v. Miller, 13 F.2d 998
(7th Cir. 1994). State courts remain split, and use unusually vivid and vehement
language, on whether the common-law concept of res gestae can be reconciled
with state rules of evidence. Compare State v. Kemp, 948 A.2d 636, 652 (N.J.
2008) (“Res gestae is the moldy cardboard box in the basement, whose contents
no longer have any utility but which we nevertheless fear discarding. The time
has come for us to rid our evidence rules of this ancient doctrine that no longer
has any contemporary relevance.”) with State v. Nelson, 791 N.W.2d 414, 423–24
(Iowa 2010) (“Therefore, under this narrow interpretation of Iowa’s inextricably
intertwined doctrine that completes the story of the crime, we must analyze the
State’s argument that the evidence the defendant was a drug dealer is not evidence
of other crimes, wrongs, or acts but is, in fact, intrinsic evidence completing
the story of the charged crime of murder in the first degree.”). See also Edward
J. Imwinkelried, “The Second Coming of Res Gestae: A Procedural Approach to
Untangling the ‘Inextricably Intertwined’ Theory for Admitting Evidence of an
Accused’s Uncharged Misconduct,” Catholic University Law Review 59 (2010): 719,
722–23, 727–28, which argues that res gestae, and a recognition of the concept
of narrative integrity introduced by Old Chief, has been revived through a use of
Federal Rule of Evidence 404(b) and is recognized by most circuit courts.
25. Old Chief v. United States, 183.
26. Ibid., 187.
27. Ibid., 187.
28. Ibid., 189.
29. Ibid., 188.
30. Ibid., 187–88.
31. Ibid., 191.
32. See Jeannie Suk, “The Look in His Eyes,” Harvard Law School Public Law and
Legal Theory Working Paper Series no. 10–13; reprinted in Criminal Law Stories,
ed. Donna K. Coker and Robert Weisberg, 171–211 (New York: Foundation
Press, 2012).

R E T R O S P E C T I V E P R OP H E C I E S [ 107 ]
0
8
1

33. I note that attorneys are now posting filings in civil suits on the Internet in an
attempt to “control the narrative.” See “Lawsuits’ Lurid Details Draw an Online
Crowd,” New York Times, February 23, 2015, http://nyti.ms/17L2Lgn.
34. See Mickens v. Taylor, 535 U.S. 162 (2002).
35. See Mickens v. Taylor, Petition for Writ of Certiorari to the United States Court
of Appeals for the Fourth Circuit, No. 00-9285. I speak of this petition as the
work of Anthony Amsterdam because I had the good fortune to work with him
in his Persuasion Institute at the time he was writing it.
36. Joseph Conrad, “Preface” to The Nigger of the Narcissus (New York: Doubleday
Page, 1897), 3.

[ 108 ] Methods