Law Of Evidence - Analysing Evidence, Relevance, Probative Value And Generalisations PDF

Preview

Summary

Law Of Evidence - Analysing Evidence, Relevance, Probative Value And Generalisations lecture notes...

Description

Law Of Evidence: Analysing Evidence, Relevance, Probative Value And Generalisations Logic and probability theories are systems of norms that are relevant to how we understand the world surrounding us. They regulate the process of proving. The law relies on logic as well as on facts. The law of evidence works within a framework of logic and norms. The law puts limits on these theories. THE RELATIONSHIP BETWEEN THE LAW OF EVIDENCE AND THE PRE LEGAL THEORY OF PROOF The Constituents Of Criminal Fact Finding Legal fact finding should be understood as under two boxes: Pre Legal Norms:   

Logic Probability Common sense reasoning

Legal Norms:  

Admissibility Sufficiency

Two scholars (Theayer and Whitmore) said that legal fact finding is similar to other instances of fact finding, that of science, but everyday fact finding, such as whether to determine if a particular fact occurs, is different. Legal fact finding is when we apply norms of pre legal norms. If we were only to use pre legal norms, it would give us a bit of evidence that we would want to use if we were to determine whether the person committed the crimes. But the legal norms establish that it doesn’t matter how useful the evidence would be, it would be unfair to use it, so they disregard it. It is the law that tells us that you need to prove it beyond reasonable doubt. A DIACHRONIC (concerned with the way in which something, especially language, has developed and evolved through time) LOOK AT THE INTERACTION BETWEEN THE TWO SETS OF NORMS IN CRIMINAL FACT FINDING Blue is legal norms. Restrictions – It’s admissible, but you can only use it for this purpose Probative value isn’t a legal norm.

IMPORTANT PRE LEGAL CONCEPTS: RELEVANCE AND PROBATIVE VALUE

An example to understand the pre legal process of proof: R v ADAMS Facts: A rape victim described her attacker as in his twenties. A suspect, Denis Adams, was arrested and an identity parade was arranged. The woman failed to pick him out, and on being asked if he fitted her description replied in the negative. She had described a man in his twenties and when asked how old Adams looked, she replied about forty. Adams was thirty, seven, he had an alibi for the night in question, his girlfriend saying he had spent the night with her. The DNA was the only incriminating evidence heard by the jury, as all the other evidence pointed towards innocence. Held: The defence called an expert to the jury. The defence argued that the match probability figure put forward by the prosecution (1 in 200 million) was incorrect, and that a figure of 1 in 20 million, or perhaps even 1 in 2 million, was more appropriate. The issue of how the jury should resolve the conflicting evidence was addressed by the defence by a formal statistical method. The jury was instructed in the use of Bayes' theorem by Professor Peter Donnelly of Oxford University. The judge told the jury they could use Bayes's theorem if they wished. Adams was convicted and the case went to appeal. The Appeal Court judges noted that the original trial judge did not direct the jury as to what to do if they did not wish to use Bayes's theorem and ordered a retrial. At the retrial the defence team again wanted to instruct the new jury in the use of Bayes's theorem (though Prof. Donnelly had doubts about the practicality of the approach). The judge asked that the statistical experts from both sides work together to produce a workable method of implementing Bayes's theorem for use in a courtroom, should the jury wish to use it. A questionnaire was produced which asked a series of questions such as: "If he were the attacker, what's the chance that she would say he looked nothing like the attacker?" "If he wasn't the attacker what's the chance that she would say he looked nothing like the attacker?" These questions were intended to allow the Bayes factors of the various pieces of evidence to be assessed. The questionnaires had boxes where jurors could put their assessments and a formula to enable them to produce the overall odds of guilt or innocence. Adams was convicted once again and again an appeal was made to the Court of Appeal. The appeal was unsuccessful but the Appeal Court ruling was highly critical of the appropriateness of Bayes's theorem in the courtroom. Bayes’ theorem provides a justification for our common sense. The theorem works on the basis of common sense. WHAT IS THE PROBANDUM (FACT TO PROVE) IN CRIMINAL TRIALS? In ADAMS, the only problematic element was identity. That was the fact to prove. WHEN IS EVIDENCE RELEVANT TO THE PROBANDUM? COMMON SENSE: “Evidence is relevant if it is logically probative (proof or evidence) or disapprobative of some matter which requires proof. Relevant (logically probative or disapprobative) evidence is evidence which makes the matter which requires proof more or less probable.” - Lord Simon in DPP v KILBOURNE. LOGIC / PROBABILITY THEORY: An item of evidence (E) is relevant to an hypothesis (H) if, and only if: If the two ratios are equal to one, the ratio is irrelevant.

When we are equally likely to find evidence that the defendant is innocent or guilty, then the numerator and denominator equal one. ADAMS case: In this case, since it’s rape, there is very likely that there is semen left in the vagina. It is likely to match the offender. The numerator probability is high, but the denominator value low. It is very unlikely that if the defendant is not the person who left the semen, the DNA of the defendant will match someone else. The ratio is not equal to the relevance of the evidence. These two definitions of common sense and logic / probability are exactly the same. Logic and probability are very in line with common sense. You can use either two definitions. AND WHAT IS THE PROBATIVE VALUE OF THE EVIDENCE WITH RESPECT TO THE PROBANDUM? What tells us that evidence is proof of a fact? Probative value can be defined as the strength of the evidence. COMMON SENSE: The probative value of the evidence with respect to the probandum is the extent to which the evidence makes the probandum more or less probable. -

To what extent does it make guilt more or less probable?

For example, ADAMS. If you come across the evidence of the DNA match, the case changes considerably. The guilt increases. Whereas if the defendant was spotted in the vicinity, it might not increase the guilt because there could be an innocent explanation such as they were out to buy groceries. LOGIC / PROBABILITY THEORY: The probative value of an item of evidence (E) with respect to an hypothesis (H) is given by the value of:

MAGNITUDE: The greater the difference between numerator and denominator, the higher the probative value of the evidence. DIRECTION: If the numerator is higher, the resulting probability of guilt is higher, if the denominator is higher, the resulting probability of guilt is lower. Therefore, this definition is much more advantageous. ADAMS: The need to recognise the face is very important considering the crime. The importance of the evidence is high. The hypothesis (the denominator) is much higher than the numerator, therefore showing that the defendant might be innocent (although the DNA evidence changes this). BUT HOW ARE WE TO COMBINE MORE ITEMS OF EVIDENCE? They’re usually more than one piece of evidence. So how is it we determine what the probative value is of multiple items?

Bayes’ Theorem: A logical method of combining the probative value of multiple items of evidence.

Plugging in the values of the evidence in ADAMS:

It is not about the amount of evidence you have, but the property value of them. So, although Adams had an alibi and couldn’t be identified, the property value of the DNA is much too great. WHAT DOES BAYES’ THEOREM TEACH US ABOUT REASONING WITH MULTIPLE ITEMS OF EVIDENCE? UPDATING: The probability of a fact is subject to change each time a new item of relevant evidence is considered. COMBINATION: The posterior probability of a fact is a function of the probative value of all the items of relevant evidence considered. The higher the probative value of an item of evidence, the more the posterior probability will change, the direction of the change depending on what the evidence is probative of (innocence or guilt). GENERALISATIONS AND THE CHALLENGES OF CLASSIFICATION Generalisations As The Glue That Sticks Facts Together A generalisation is, roughly, a statement to the effect that (no, few, some, most, all) facts (things / people) falling into a certain class have a particular property. EXAMPLE: All men are mortal. Legally Relevant Generalisations: “DNA match is strong evidence of guilt.” “A victim’s failure to identify the defendant is indicative of the defendant’s innocence.” “A corroborated alibi is stronger evidence of innocence than it is of guilt.” These generalisations are reflected in the likelihood ratios used in ADAMS. How likely is it, if the defendant is guilty, we’ll find a DNA match?

Questions like these are answered by looking at statistical facts. WHERE DO GENERALISATIONS COME FROM AND HOW DO WE USE THEM?

THE PROBLEM WITH CLASSIFYING INDIVIDUALS US v SHONUBI Facts: The defendant, Mr Shonubi, is a Nigerian citizen, resident of the US. He works as a toll collector. Shonubi is convicted for smuggling heroin into the US through JFK airport on a particular occasion. It is established that he has smuggled heroin also on previous occasions, even though he has not been convicted for these conducts. According to US law the sentence should depend on the total quantity of heroin involved (also on the amount of heroin smuggled on previous occasions). In the absence of evidence on the amounts previously smuggled by Shonubi, the judge resorts to statistical evidence (generalisations) concerning the amount of heroin usually smuggled on individual trips by Nigerian drug smugglers arrested in JFK. Held: The court of appeal strikes down the sentencing decision, arguing that the statistical evidence cannot be used because it is not evidence specific to Shonubi’s case. They do not explain why and what the evidence specific is supposed to be. WHAT LIES BEHIND THE REQUIREMENT OF SPECIFIC EVIDENCE? Colyvan: The case of SHONUBI raises the reference class problem. Each of us belongs to a number of classes. Different traits allow you to classify in a number of infinite ways. Because there are so many classes, it is virtually impossible to identify the class that would give us the answer as to how much drugs Shonubi smuggled.    

Class of Nigerian citizens. Class of heroin smugglers. Class of residents of the USA. Class of toll collectors.

The only specific evidence we can use is evidence of Shonubi’s prior trips. Since this doesn’t exist, he cannot be convicted for the drugs he smuggled in the last trips and only on the current one. THE REFERENCE CLASS PROBLEM: UBIQUITOUS (PRESENT) BUT WITH A LOGICAL SOLUTION 

Every inference involves generalisations (therefore, classification).

    

Thus, the problem is ubiquitous (present, appearing, or found everywhere). In fact, not even the inference suggested by Colyvan escapes the reference class problem. It derives, that if the problem had no solution we would not be able to draw correctly any inference whatsoever. But we do draw inferences. Importantly, we often get them right and we don’t seem to draw them randomly. Therefore, the problem seems to admit of a logical solution. What is the solution? Consider toll collectors verses drug smugglers.

Not all reference classes are useful or relevant. BUT IS THERE ALSO A MORAL PRONG OF THE REFERENCE CLASS PROBLEM? Epistemic Prong: Related to the accuracy of the finding. Accuracy of inference – truth finding so far. Moral Prong: This has to do not with the accuracy, but to do with finding a reference class that is morally appropriate. That would render our inference morally appropriate. For example, in SHONUBI, some might find that it was morally wrong the inference the judge used. He decided to judge someone’s behaviour by looking at how other people behave. Did what the judge do in SHONUBI, violate a moral rule on how we should draw inferences? Even if his inference was correct. IS THIS A SENSIBLE MORAL RULE CONCERNING OUR INFERENTIAL REASONING? When we are in the process of drawing an inference about the behaviour or the traits of an individual such as a witness or the defendant, morality demands that we refrain from using evidence relating to people other than the individual at issue. But if we couldn’t use it, we would be unable to draw many inferences concerning individuals, starting with inferences that matter for criminal trials such as witness reliability. We often use it and in many of these cases our lines of reasoning are not morally problematic. For example, judging attendance by looking at the collective attendance of lectures. However, using this evidence is not equivalent to holding individuals responsible for the conduct of other people. WHAT ABOUT THIS, INSTEAD? Morality demands that we refrain from using reference classes built upon socially problematic traits such as race, nationality and social class. The risk of using these traits is that of producing or strengthening (in particular, negative) stereotypes. Another related risk is that of giving the false impression that someone is convicted and sentenced for the socially problematic traits that define her, rather than for what she has done. Was this an issue in SHONUBI? Arguably, yes. It may have strengthened stereotypes against Nigerians. REMEMBER THE FUNCTION OF EVIDENCE LAW WITH RESPECT TO THE PROCESS OF PROOF

If we accept that the second reason is a reasonable demand, we are bound to recognise that there are reasons for the law to interfere with the pre legal norms in order to safeguard particular values, even if this interference reduces the accuracy of fact finding. But whether there is, in fact, a moral prong of the reference class problem and whether the law should recognise this prong are still open questions....