Patrick Ball, late of the American Association for the Advancement of Science, is essentially a geek for justice. He and his colleagues across the world have collected, organized, and statistically interpreted the evidence of some of the gravest human rights violations of the past twenty years, from South Africa to Guatemala, from Indonesia to Chad. Their results have supported the work of truth commissions and prosecutions alike: Ball himself appeared as a witness in the trial of Slobodan Milosevic.
Their task is to distill fact from anecdote, building from many sources a reliable record of exactly who did what, when, to whom. In El Salvador, for instance, careful cross-referencing of a mass of data was able to pinpoint the individual army officers most responsible for atrocities. Forced into retirement after the end of the civil war, some of the officers brought suit to overturn the statistical conclusions. As Ball told the story to a convention of hacker-activists: “So we went into court with what lawyers go into court with. That is, dozens of cases on paper. But we also went in with diskettes with my code, and we gave it to the judge and said, ‘Here’s how it was done.’ . . . They backed off and withdrew their suits. . . . The reason it worked? Big data.”
Big data and Bayesian correlations are also helping the police with their inquiries. Kim Rossmo was a beat constable in the rougher parts of Vancouver. He could feel the subtle sensory shifts that mark one block as different from another: the unseen borders between this community, this power center, and the next. He drew the obvious, literary parallel between the city and a predator’s environment, but took it further: the repeat criminal is not just a raptor or stalker, moving at will through a mass of docile herbivores. He (for it is mostly “he”) also has the constraints of the predator: the desire to hunt with least effort and most certainty, the habitual association with a few places, the established cat-paths between them. Studying for his Ph.D. at Simon Fraser University, Rossmo took these ideas and began developing a set of algorithms to deduce the criminal’s home territory from the geographical distribution of crime scenes:Criminals will tend to commit their crimes fairly close to where they live. Now, there are variations; older offenders will travel further than younger offenders, bank robbers will travel further than burglars, whites will travel further than blacks . . . but the point is that the same patterns of behavior that McDonald’s will study when they’re trying to determine where to place a new restaurant, or that a government may look at in terms of the optimal location for a new hospital or fire station, also apply to criminals.
There is one important difference, though, and that’s what we call a “buffer zone”; if you get too close to the offender’s home the probability of criminal activity goes down. And so at some point where the desire for anonymity and the desire to operate in one’s comfort zone balance, that’s your area of peak probability.
Rossmo went on to be head of research for the Police Federation in Washington, D.C. His geographical profiling software is being used by police forces in several countries to deduce patterns from the scatter of facts across a map, correlating geographical data with the results from other investigative techniques. Its success depends on the knowledge that behavior is never truly random. Crime may be unpredictable, but criminals are not.
“You come into a room, it’s full of blood; there’s someone there with a knife sticking out of him. What you should not do is make a hypothesis. That’s, I think, the greatest source of miscarriages of justice.” Jeroen Keppens unconsciously echoes the Talmud. “Instead, you look, say, at this pattern of blood on the wall; is it a drip or a spray? If you think it’s a spray, what direction did it come from? You make micro-hypotheses, plausible explanations for each piece of evidence as you see it.”
Keppens is not someone with whom you would expect to discuss blood splatter: he is a soft-spoken, tentative, courteous young Dutchman, with nothing of the mean streets about him. And yet, as an artificial-intelligence expert, he is building computerized decision-support systems to help the police reason their way through the goriest cases:If I attack you, there will be some transfer, maybe of fibers from my jumper, and there will be existing data to say how rare this fiber is, how much we would expect to be transferred, what rate it would fall off afterwards, and so on. All these generate probabilities we can assign to the fact the fibers are found on you. For other things like mixtures of body fluids the chain of inference is more complex, but they still let you build a Bayesian net, connecting possible scenarios with the evidence.
What we are doing is supporting abductive reasoning: deductive says, “I think I know what happened; is this evidence consistent?” and inductive asks, “Based on this evidence, do I think this scenario fits?”—but abductive goes further: “I have this evidence; does it suggest any plausible explanations? What other evidence would these explanations generate? What should I look for that would distinguish between these explanations?” You build out from what you see—and as forensic science becomes more complex, it’s harder to be sure what the conclusion is. So putting in real numbers and doing formal inference calculations can be useful.
But how can a system where the probabilities are based on opinions—even those of forensic experts—be reduced to numbers? Isn’t it guesswork dressed up as science? “I think scientists do understand the degree of uncertainty in any system; there might be an argument for using ‘fuzzy’ sets representing words—‘very likely,’ ‘quite likely,’ ‘very unlikely’—rather than precise numbers. But the point is that right now, experts appear in court and use these same words without the calculations—it’s off the top of their head—whereas Bayesian probability, even with the range of uncertainty in the data, can produce very strong likelihood ratios.”
This is not the beginning of push-button justice, but computers can spread the instincts of the expert more widely, giving every police force the same informed sense of likelihood. Keppens says: “In a small town or remote region, the first person on the scene of a major crime, the uniformed officer, has probably never seen anything like this before. The decisions made in the first five minutes are crucial—sometimes they make the difference between it being recognized as a crime or being ignored. Those are the kinds of decisions we want to support.”
Richard Leary holds the bridge between the theoretical, academic side of decision-support systems and the real world of the investigation room, walls covered with Post-it notes and desks with cups of cold coffee. Until recently he was a senior detective in the West Midlands police force, Britain’s largest outside London; and he preserves the policeman’s combination of closeness and distance, speaking openly but with deliberation, taking care to enforce his credibility.
He is describing FLINTS, the computerized system he created for helping detect high-volume crimes: “It’s a systematized art form, sort of a bastardization of Wigmore, DNA profiling, and a little chaos theory.” The system works by generating queries, prompting the investigator to seek out patterns or connections among the evidence held in separate databases: DNA, fingerprints, footwear, tool marks. By supporting several hypotheses and determining the evidence necessary to confirm or disprove each, it helps point out the obvious-in-retrospect but obscure links between people and crimes that may lie hidden in a mass of data. Leary says:Abductive reasoning requires the investigator to think carefully about what he’s doing with the facts. Usually, there’s plenty of information—that’s not what we’re short of. The real questions are, “Where does this information come from, how can we use it to formulate evidence, and then how do you use evidence to formulate arguments, then take those arguments from the context of investigation to the context of the court, all while still preserving the original context?” If you’re going to do that, you can’t just rely on gut instinct or on flashy technology—an investigator has to think in a methodological fashion, to have a conscious, logical system of assembling information into arguments.
The advantage of the Wigmore approach to systematizing investigation is simple: it points out gaps in a
vailable evidence. “All too often, an investigator only seeks new evidence to try and firm up a currently favored theory, rather than to discriminate between credible alternatives. Missing evidence and intelligence are often not considered in themselves. But you’re actually trying to eradicate doubt—that’s the purpose of investigation—and one of the best ways to do that is to look at the data you do have and then see what’s missing in the light of each hypothesis; then engage in searching for that new data. It’s common sense, but it’s not common practice.”
The reason, according to Leary, is police investigation training, which is usually based on the law rather than logic. “For example, the national senior investigating officers’ course, right now, concentrates on murder. Well, I don’t understand what’s the special logic of investigating murder as opposed to the logic of burglary or the logic of illegal drug trafficking. They are assuming that investigation is driven by law rather than by science and logic; their training material is subject-specific, not about developing methods of thinking.”
Do computerized reasoning systems provide the answer? Only up to a point: “Data is collected according to a fixed objective and it’s arranged and presented systematically, so there’s a golden thread of logic running through it. But data is never perfect; for high-volume crimes, for example, a lot is simply not reported. And when you have these very visual ways of presenting data—crime hot spots on a map—they are very persuasive. The visual pattern can skew the analyst’s judgment. You can never just abdicate human responsibility to a machine.”
Of all the deified virtues of the ancient world—Wisdom, Domesticity, Revenge—we set up statues only to Justice. She alone remains a goddess, perhaps because justice is so hard to define as a human quality. It should be eternal yet contemporary; absolute in law, yet relative to the case. We understand that legal proof—whether by the preponderance of evidence or beyond a reasonable doubt—is a matter of probability; but that the choice, once made, goes by forever ’twixt the darkness and the light.
Trial by jury represents our attempt to wring error out of judgment. We hope that, as with scientific observation, averaging the views of twelve citizens will produce a more accurate result than asking just one. This assumes, though, a normal distibution of juror prejudice around some ideal, shared opinion—but prosecutors and defenders alike try to skew that distribution through juror selection.
Different lawyers swear by different systems: one considers the Irish likely to feel sorry for the accused, another thinks they have too many relatives in law enforcement. The great defender Clarence Darrow sought out Congregationalists and Jews, but strove to purge his juries of Presbyterians. Everyone agrees that suburban homeowners convict: they fear crime, worship property, and haven’t suffered enough. It is an axiom that if your client is likely to be found guilty, you must try to get a cantankerous old woman on the jury, who will enjoy resisting the eleven others.
Modern methods of jury packing began, surprisingly, with the trials of antiwar activists in the 1970s, when volunteers from social sciences departments profiled the various shades of opinion in the towns where cases were coming to trial, giving their advocates the ability to shape the jury through summary objection. Their success inspired a whole industry: one of the professors, Donald Vinson, opened a $25 million firm to offer the benefits of social profiling to the wider world. He claimed that if you ordered his full range of services you could be 96 percent sure of the verdict—which is the kind of justice worth buying.
Law is what lies beneath, but it also means fairness—“giving someone his law” used to be the term for a head start or a favorable handicap. The biggest legal fiction of all is the pretense that the process itself assures fairness, when it is we who should do so. It is we who take the continuum of experience and assign it to one of two doors (except in Scotland, where the third door is “not proven”—colloquially, “not guilty, but don’t do it again”).
What, therefore, do we need to assure that justice is “what usually happens”? British law has long relied on the idea of the “reasonable person,” defined, at the turn of the twentieth century, as “the man on the Clapham omnibus.” The image was carefully chosen: Clapham, south of the Thames, was a bulwark of lower-middle-class respectability. The man on the omnibus would be jolting home from a clerical job in the City. His opinions would be moderate, his expectations mildly optimistic, and his indulgences quiet and frugal. He would read newspapers but not appear in them.
On the bus to Clapham today you will sit between an investment banker, who is going to triple-lock herself into her bleak apartment before microwaving some Lean Cuisine; and an unemployed Muslim youth who is going to hang out on the streets with his friends before being stopped and searched by the police for the ninth time in the past six months. Either of them (for Britain does not have summary objection) could end up on the jury that tries your case. Which of them is the “reasonable person”?
Most studies of jury deliberation suggest that jurors make decisions intuitively, swiftly, and in relation to themselves. They create a story that revolves around personalities rather than evidence. The author of “He’s a cold-hearted killer who planned this all ahead of time” has heard the same facts as the man who sees “Basically a nice guy who got into a situation that was too much for him.” Once a story has taken shape, a juror assesses evidence in terms of it. The twelve jurors will each fill in gaps with causes and motivations supplied from their own experience: in effect, they create their own probability. This is where Aristotle’s model breaks down and rhetoric earns its bad reputation, as the smart lawyer plays to the assumptions of each type, not to a shared sense of law or likelihood.
As jurors, though, our job is to decide on the facts; so why aren’t we allowed the tools to do so? Why are there laws of evidence restricting what can come to court? Why are we not given a briefing on legal issues by the judge in plain language before the trial begins? Why, instead of being a randomly chosen jury of peers, are we selected from a sociological shopping list? And why—since many true conclusions are counterintuitive—are we denied instruction in the forms of probability that let us refine our intuition? How else, in this complex and unpredictable world, will we know what usually happens?
Certainty is the song the sirens sang; but probability is a tune we all could learn without danger—not as a substitute for common sense, but as a check on it. Law is a succession of likelihood problems, no two the same, with every possible flaw of evidence and presentation. Yet we need not depend solely on raw instinct, rhetoric, or prejudice to solve them: we have methods to refine our opinions and bring the likely out of the mass of possibility. As the great lawyer Robert Ingersoll warned a jury in 1891: “Naturalness, and above all, probability, is the test of truth. Probability is the torch that every juryman should hold, and by the light of that torch he should march to his verdict. Probability!”
9
Predicting
I had a dream, which was not all a dream.
The bright sun was extinguish’d, and the stars
Did wander darkling in the eternal space,
Rayless, and pathless, and the icy earth
Swung blind and blackening in the moonless air;
Morn came and went—and came, and brought no day.
—from Darkness, written by Byron in the nonexistent summer of 1816 that
followed the eruption of Tambora, in the East Indies
Today the rain falls in a series of receding planes; yesterday the sun saw everything, blazing from a porcelain-blue sky touched up with stratocirrus by the best Italian ceiling painters. This window has framed afternoons when the blind fog wiped its dripping nose against the glass; roaring nights when a runaway westerly bucketed by; and stunned, reverent mornings of first snow with crows speckled calligraphically across the silent fields.
When the English begin all conversations with a discussion of the weather, it is a way of gossiping, without vulgarity, about the most dynamic personalit
y they know. The weather is an ever-present but capricious lover, alternating moments of heart-lifting generosity with flashes of devastating temper. Earth is potential, weather is action; we propitiate the goddess—we watch out for the god.
Now, when Zeus has brought to completion sixty more winter
Days, after the sun has turned in its course, the star
Arcturus, leaving the sacred stream of the ocean,
First begins to rise and shine at the edges of the evening.
The lines are from Hesiod’s Works and Days, a poetic compendium of useful rural knowledge. A very old farmer’s almanac, it tells the days to begin planting, pruning, and threshing, and explains how evil came into the world. Hesiod, himself a Boeotian shepherd, believed entirely in the interrelatedness of these things—yet any farmer can tell you that one year is very different from another. Less piously accepting minds began to wonder: was Zeus responsible for the consistency or the inconsistency? Theophrastus, Aristotle’s student and successor, wrote extensively about the wind, thunder, and lightning, but denied their divine origin: capriciousness could not be the work of God, the fountain of order.
If thunderbolts originate in God, why do they mostly occur during spring or in high places, but not during winter or summer or in low places? In addition: why do thunderbolts fall on uninhabited mountains, on seas, on trees, and on irrational living beings? God is not angry with those!
The classical world’s ambivalence about the weather reflected a familiar division between the comforts of universal divine causality and the uneasiness of scientific doubt. Opting for comfort was the poet Aratus, whose Phenomena—a cobbling together of borrowed astronomy, conventional piety, and folk weather-lore—was the most copied work after the Iliad and the Odyssey. In the Phenomena, God is everywhere: “red sky at night, shepherd’s delight,” for instance, is one of His assurances—not, perhaps, as strong a covenant as the rainbow, but still useful.
Chances Are Page 23