Clean scene or snitch glitch?

My criticism of the New Zealand Supreme Court's majority decision in Morgan v R can now be assessed in the light of a correct approach by the Supreme Court of Canada to the use of a jailhouse informer's evidence in R v Hurley [2010] SCC 18 (14 May 2010).

In Hurley the cell mate claimed that the accused had confessed to him that he had cleaned the crime scene in an effort to remove evidence of his participation; in particular, he had attempted to remove all traces of his DNA.

One of the requirements of the warning that must be given to the jury in such cases (this in Canada is called the Vetrovec warning, after R v Vetrovec [1982] 1 SCR 811) applies where reasons to doubt the reliability of a witness may not be self-evident to the jury:

"something in the nature of confirmatory evidence should be found before the finder of fact relies upon the evidence of a witness whose testimony occupies a central position in the purported demonstration of guilt and yet may be suspect by reason of the witness being an accomplice or complainant or of disreputable character..." (Vetrovec, per Dickson J, para 42).

See also my discussion of R v Khela [2009] SCC 4.

In the circumstances in Hurley, supporting evidence for the cell mate could only be evidence that the crime scene had been cleaned. There was no issue that the accused had been at the scene, but he gave an innocent explanation for that, so presence of some of his DNA would not be conclusive. This is encapsulated in para 7 of Hurley (Mr Niemi is the cell mate):

"The evidence about the attempts to clean the room was unquestionably significant to the Crown's case. In his closing address, Crown counsel submitted to the jury that, on the evidence before the court, it was "clear that there was an attempt to clean the room" ... . The evidence of attempts to clean the room could be taken as independent evidence tending to support Mr. Niemi's evidence about his conversation with Mr. Hurley. In turn, Mr. Niemi's evidence supported the Crown's case that Mr. Hurley had not only been there, but also that he was the killer. Thus anything that tended to rebut the room cleaning theory tended to weaken the independent evidence that could be seen as supporting Mr. Niemi's version of the conversation with Mr. Hurley and, consequently, weakened the Crown's circumstantial case."

There was fresh evidence concerning the finding of Mr Hurley's DNA at the crime scene – this evidence was not available to the Court of Appeal – and the Supreme Court ordered a new trial. This evidence made it unnecessary for the Court to address the seriousness of criticism of the trial judge's Vetrovec warning, which was that the jury had not been told sufficiently why they might have reservations about the credibility of the cell mate's evidence of the accused's confession.

Hurley brings into focus the need for precision as to what evidence is capable of supporting, or undermining, the reliability of the contested evidence. Whereas in Morgan the New Zealand court majority judgment was imprecise on this (and precision would have revealed that there was no such supporting evidence), in Hurley the Canadian court highlights the supposedly supportive evidence and concludes that a second jury might not think it was as strong, in the light of the fresh evidence, as it had seemed.

Confidentiality and clear decision modelling

Anyone interested in claims to privilege by the media, especially claims in resistance to the execution of search warrants, will need to read R v National Post
[2010] SCC 16 (7 May 2010).

I mention it here only in passing, for it would otherwise require lengthy exposition. The point that will attract the attention of readers familiar with R v Grant and the questionable application of that decision in R v Harrison (see my blogs for 18.7.09, and 19.7.09) is the two-dimensional nature of the conceptual model used in National Post, in contrast to the three dimensional model suggested in Grant.

Grant concerned the admissibility of improperly obtained evidence, and there is an obvious analogy between that and the issue in National Post. The link is whether seizure of the contested evidence would be improper and the consequences if it was.

National Post adopts the Wigmore criteria for deciding whether a claim to privilege should be permitted:

"[53] The "Wigmore criteria" consist of four elements which may be expressed for present purposes as follows. First, the communication must originate in a confidence that the identity of the informant will not be disclosed. Second, the confidence must be essential to the relationship in which the communication arises. Third, the relationship must be one which should be "sedulously fostered" in the public good ("Sedulous[ly]" being defined in the New Shorter Oxford English Dictionary on Historical Principles (6th ed. 2007), vol. 2, at p. 2755, as "diligent[ly] . . . deliberately and consciously."). Finally, if all of these requirements are met, the court must consider whether in the instant case the public interest served by protecting the identity of the informant from disclosure outweighs the public interest in getting at the truth. See Wigmore on Evidence (McNaughton Rev. 1961), vol. 8, at § 2285; Sopinka, Lederman and Bryant: The Law of Evidence in Canada (3rd ed. 2009), at paras. 14.19 et seq.; D. M. Paciocco and L. Stuesser, The Law of Evidence (3rd ed. 2002), at pp. 163-67. Further, as Lamer C.J. commented in Gruenke:    

"This is not to say that the Wigmore criteria are now "carved in stone", but rather that these considerations provide a general framework within which policy considerations and the requirements of fact-finding can be weighed and balanced on the basis of their relative importance in the particular case before the court. [p. 290]""

The fourth step is the weighing exercise that here is two-dimensional: the two public interests are weighed against each other.

"[58] The fourth Wigmore criterion does most of the work. Having established the value to the public of the relationship in question, the court must weigh against its protection any countervailing public interest such as the investigation of a particular crime (or national security, or public safety or some other public good)."

...

"[61] The weighing up will include (but of course is not restricted to) the nature and seriousness of the offence under investigation, and the probative value of the evidence sought to be obtained, measured against the public interest in respecting the journalist's promise of confidentiality. ...".

Notable here, where we are thinking of an analogy with the balancing involved in deciding the admissibility of improperly obtained evidence, is the linking of seriousness of the offence and the probative value of the evidence. Those are matters that could properly be imagined as belonging on a y-axis. Against them is the public interest against admission, and in the context of improperly obtained evidence this would be measured on an x-axis. These two axes would indicate two boundaries of a field of balance points, and the field would be divided, in a way reflecting policy, into areas of admission and exclusion.

This is a conceptually clear model, and avoidance of any suggestion of three-armed balances is welcome. The Grant model presented the difficulty of regarding as a third arm of the balance the impact of the impropriety on the accused; instead of seeing this as a third arm, it would be better to combine it with the "seriousness of the impropriety" arm, or to separate it out as a prerequisite to getting to the balancing stage in a way akin to the role of the first three of the Wigmore factors.

Beware! Teachers may even now be preparing their "compare and contrast" exam questions.

Stairway to heaven?

Disagreement over whether sufficient reasons had been given for a judge-alone decision occurred in the Supreme Court of Canada in R v Szczerbaniwicz [2010] SCC 15 (6 May 2010). The Court had given guidance on sufficient explanation for judge-alone decisions in R v REM (blogged here 3.10.08).

To prevent his wife from damaging his Masters diploma which had been hanging on the wall of a stairway, the appellant (let's call him D) pushed her. She fell on the stairs and received bruising.

D relied on the justified use of force in defence of his property: s 39(1) Criminal Code:

"39. (1) Every one who is in peaceable possession of personal property under a claim of right, and every one acting under his authority, is protected from criminal responsibility for defending that possession, even against a person entitled by law to possession of it, if he uses no more force than is necessary."

This was held to import a requirement of reasonableness (majority per Abella J para 17; minority per Binnie J para 29, 33). The prosecution must prove D did not use reasonable force and that he did not believe on reasonable grounds that he was using reasonable force.

The minority (Binnie and Fish JJ) did not suggest that the verdict had been wrong, but held that sufficient reasons for it had not been given. Of concern was absence of explanation for how the pushing or shoving was itself an excessive amount of force, either objectively or subjectively (para 26, quoting the dissenting judge in the court below). Yes, the wife fell on the stairs and received bruising, but it is wrong to reason backwards by saying that because the injuries were serious the force used was excessive (para 39). Also, it was wrong to rely on D's apparent admission that he had overreacted, because he may have lost his self-control but still used reasonable force (para 40). D can't be expected to "weigh to a nicety" the amount of force needed in a quick response situation (para 35). The fact findings were unclear as to where the diploma was when D pushed his wife, so one could not say "why" the judge concluded the force used was unreasonable (para 43).

It helps in this sort of case to return to basics and separate the actus reus and the mens rea components of the offence in the light of the claimed defence. The actus reus is the use of unreasonable force. The only evidence of unreasonableness was the injuries. They resulted from the victim's loss of balance on the stairway, so the actus reus question is whether the force used, being sufficient to cause loss of balance, was reasonable. Only if the actus reus component is satisfied do we need to consider the mens rea component. The mens rea component is either knowing that the force used was unreasonable, or unreasonably thinking it was reasonable. Those issues would require fact findings on, for actus reus, the nature of the stairway (steepness, length, availability of handrails) and the likelihood of the victim losing her balance (which way was she facing and leaning compared to the direction of the push, what was she doing at the moment she was pushed, did she have any warning, should she have foreseen the danger?), and for mens rea, what risk to his property did D reasonably apprehend – where was the diploma and what was about to happen to it, what alternative course should he reasonably have been aware of – would she have stopped if warned, what risk of injury to his wife should he reasonably have foreseen – what force did D intend to use and what was his purpose in using it?

The majority found the trial judge's decision to be "eminently justified" (para 23), stressing the finding that D had lost his temper.

This is one of those cases where the correct verdict may well have been arrived at, and the SCC majority's judgment would, on its face, appear routine, but where closer analysis by the dissenters leaves us in some doubt as to the rigor with which R v REM has been applied.

Liability in international criminal law – exclusion from refugee status

International criminal law may contain wider complicity provisions than exist in domestic law. Different standards for decision making can also exist. These points are illustrated by JS(Sri Lanka) v Secretary of State for the Home Department [2010] UKSC 15 (17 March 2010).

"serious reasons for considering", Lord Brown, with whom the other Justices agreed, held (para 39)"

"...obviously imports a higher test ... than would, say, an expression like "reasonable grounds for suspecting". "Considering" approximates rather to "believing" than to "suspecting". I am inclined to agree with what Sedley LJ said in Yasser Al-Sirri v Secretary of State for the Home Department [2009] EWCA Civ 222, para 33: "[the phrase used] sets a standard above mere suspicion. Beyond this, it is a mistake to try to paraphrase the straightforward language of the Convention: it has to be treated as meaning what it says."

The Convention under consideration here, the Refugee Convention, excludes from refugee status any one in respect of whom (Art 1F(a)):

"there are serious reasons for considering that (a) he has committed a crime against peace, a war crime, or a crime against humanity, as defined in the international instruments drawn up to make provision in respect of such crimes".

The focus here is on the acts of the applicant for refugee status. The Court stressed that mere membership of an organisation that commits war crimes is not enough to establish serious reasons for considering that he has committed such crimes.

The wide definition of complicity in international law must be applied, and here Articles 12(3) and 25(3) of the Rome Statute of the International Criminal Court were relevant. In summary (38), extended complicity here includes

" ... wider concepts of common design, such as the accomplishment of an organisation's purpose by whatever means are necessary including the commission of war crimes. Put simply, I would hold an accused disqualified under article 1F if there are serious reasons for considering him voluntarily to have contributed in a significant way to the organisation's ability to pursue its purpose of committing war crimes, aware that his assistance will in fact further that purpose."

The Court held that it is wrong to try to place the relevant organisation at some point on a continuum ranging from the (in my terms) relatively benign to the relatively malevolent, as was done in Gurung v Secretary of State for the Home Department [2003] Imm AR 115. There are (32) too many variable factors in war crimes cases, and the proposed continuum included irrelevant considerations:

" ... Whether the organisation in question is promoting government which would be "authoritarian in character" or is intent on establishing "a parliamentary, democratic mode of government" is quite simply nothing to the point in deciding whether or not somebody is guilty of war crimes. War crimes are war crimes however benevolent and estimable may be the long-term aims of those concerned. And actions which would not otherwise constitute war crimes do not become so merely because they are taken pursuant to policies abhorrent to western liberal democracies."

Instead, the focus is on what the individual applicant for refugee status actually did and actually believed.

This leads me to the question (in criminal law, including the law of evidence) of when conceptual modelling may appropriately involve the use of a continuum in analysis. One can imagine that a Canadian court might have been tempted to apply a "decision tree" type of approach to classifying terrorist organisations, by analogy to its approach in the unrelated field of improperly obtained evidence in R v Grant, discussed here on 18 July 2009 (twice) and on 19 July 2009. That case came down to what I think is one-dimensional (linear) continuum analysis. Continua in two dimensions can assist in admissibility decisions where the degree of official impropriety must be balanced against the public interest in admitting improperly obtained evidence. In rare cases the actus reus may be held to exist after a balancing of rights, for example disorderly behaviour: Brooker v R, discussed here on 4 May 2007. But in cases like JS(Sri Lanka) the focus on the applicant's acts and beliefs makes weighing inappropriate. Either there is an actus reus and mens rea, or there is not. They are either established to the necessary standard or they are not. There are no variables to measure and weigh against each other.

Fundamentals of secondary liability and wilful blindness

A useful reminder of the meanings of aiding, abetting, and wilful blindness has been given by the Supreme Court of Canada in R v Briscoe [2010] SCC 13 (8 April 2010). Charron J, for the Court, noted that the actus reus of aiding is assisting or helping the principal offender, and abetting is encouraging, instigating, promoting or procuring the commission of the offence (para 14).

The definition of secondary liability in s 21(1)(b) and (c) of the Criminal Code will be familiar everywhere:

"21. (1) Every one is a party to an offence who

(a) actually commits it;

(b) does or omits to do anything for the purpose of aiding any person to commit it; or

(c) abets any person in committing it."

The mens rea element has two components: purpose or intention of assisting (which does not require a desire that the offence should be committed), and knowledge of the principal offender's intention to commit the offence (para 16, 17).

In some jurisdictions inciting, counselling, and procuring are included with aiding and abetting, but in Canada counselling is dealt with in s 22 of the Code and is defined there to include procuring, soliciting or inciting.

Wilful blindness is a substitute for actual knowledge. It is not the same as recklessness. It is a subjective state of mind: seeing the need for further inquiries but deliberately choosing not to make them. Wilful blindness could more aptly be called "deliberate ignorance" (para 20, 21, 24).

The Court in Briscoe approved Glanville Williams's description, in Criminal Law: The General Part (2nd ed, 1961) at p 159, of wilful blindness being equivalent to knowledge. This could be a little confusing, because if it is equivalent to knowledge, and if recklessness is the unreasonable taking of a known risk, then wilful blindness should constitute the knowledge required for recklessness. We are left clear on what wilful blindness is, but confused about why its distinction from recklessness is important. Williams was, as the passsage quoted at para 23 of Briscoe makes clear, concerned to distinguish wilful blindness from civil negligence (in relation to the failure to perceive what a reasonable person would have perceived). A negligent failure to be suspicious could not amount to wilful blindness, but - I suggest - recklessness involves an actual suspicion and could be wilful blindness. Acting in the face of a suspicion is not far removed from acting while not wanting to know if the suspicion is true.

These considerations prompt us to ask whether the mens rea for a secondary party should necessarily be stricter than that for a principal offender. However, no new law was established in Briscoe. The alleged secondary party had to know that the principal had the necessary state of mind to be guilty of the offence, and here the trial judge should have considered whether Mr Briscoe had been wilfully blind so as to have this knowledge imputed to him. The appeal was dismissed, so that the Court of Appeal's order for a retrial stood.

Notional even-handedness?

Who can more easily adduce hearsay evidence – the prosecution or the defence? Or are they equal?

In Morgan v R (criticised here recently) the majority of the Supreme Court held that the Crown could adduce evidence that the accused when in a cell had confessed to his cell-mate who was now a hostile witness. This hostility was the route through which the evidence came in, but in reality the position was the same as if a reliable witness had been called to say that the cell-mate had previously said that the accused had confessed. This is, in effect, like the use of a "notional witness" who gives hearsay evidence.

One would think, by analogy, that the defence could call a reliable witness to say that another person – not the present accused - had confessed. But in R v Key [2010] NZCA 115 (31 March 2010) the Court of Appeal refused to allow the defence to ask a police officer whether he had made a note of his conversation with the mother of a witness, who said that the witness had confessed to her. The witness in evidence had denied making such a confession to his mother, and the mother was not available to testify. The reason the evidence from the police officer was not admissible was because there was insufficient assurance of reliability as is required by s 18(1)(a) Evidence Act 2006.

It seems odd that there was insufficient assurance of reliability where (1) a mother told the police what her son had said; (2) where the son had spoken against self-interest; (3) where a police officer investigating the case noted the conversation. Where were the sources of error in these against-interest communications? Furthermore, in contrast to Morgan, in Key the confession contained information that only a guilty person would know and that was corroborated by other witnesses.

I was not involved in either case, and it may be that on careful scrutiny of the evidence Key can be justified. But these are the sorts of cases that make observers most anxious about the even-handedness of the application of the law.

Slaves to the system?

For refusal of counsel's application for leave to withdraw from acting, see R v Cunningham [2010] SCC 10 (26 March 2006).

Current purpose or original meaning: interpreting the US Constitution

Justices Breyer and Scalia debate Constitutional interpretation! Video clip available here.

Possession and internet files

The meaning of possession of electronic files has been established by the Supreme Court of Canada in R v Morelli [2010] SCC 8. This case concerned allegedly pornographic internet files. Central to the requirements for possession is the concept of control.

The Court split 4-3. The minority made much more use of case law than did the majority, and the majority relied more on fundamental principles. The majority held that knowledge is a requirement for control, and that control is not constituted by the mere viewing of an internet file. Control requires knowingly acquiring the underlying data file and storing it in a place within one's control. The mere automatic caching of an internet file does not of itself amount to storing it unless that caching occurred with knowledge that the file was retained.

The minority regarded control as existing when the accused willingly took or maintained the use or benefit of the material, so that it did not need to be on his computer's hard drive.

In this case a search warrant had been obtained without sufficient grounds (as held by the majority) and the evidence had been obtained improperly. Applying R v Grant [2009] SCC 32 the majority held the evidence to be inadmissible.

Evidence of fingerprint matches

For an interesting article on how fingerprint evidence should be given, see New Scientist, 22 March 2010 , "Fingerprint evidence to harden up at last" by Linda Geddes.

The Waihopai spy base case

In this site, which is aimed at the decisions of leading appellate courts, I do not usually comment on trials. However, today our little nation is abuzz with discussion of the implications of what the media call the Waihopai spy base case.

Some people damaged some big plastic domes that cover secret communications facilities. I am vague on this description, but for me the most interesting thing about this case is the revelation that the big white domes can be slashed with a knife to expose the underlying things. Aerials and whatnot. The people were put on trial and yesterday were acquitted by a jury of offences like burglary and wilful damage.

What was their defence? That is not really the point. What matters is that a jury decided the case, and acquitted when a judge almost certainly would have convicted.

The case illustrates the role of juries in bringing community standards to the law. The common law's origin is the customs of the realm, and those customs were based on what people thought was reasonable. If the community thought that what a person believed and what he did were reasonable, then he would not be held to be acting illegally. You can apply the same criteria today to the criminal law. Although offences are now statutory, their definitions will always - except for some minor offences of absolute liability - be such as to permit acquittal of people who have held reasonable beliefs and who have acted reasonably in pursuit of those beliefs. Sometimes the law is more generous, allowing a defence if the belief was unreasonably held as long as it was honestly held (for example, self defence).

The best arbiter of what a community believes to be reasonable is a jury. This is recognised by courts (see the discussion by Heydon J in AK v Western Australia [2008] HCA 8 at paras 90-98) and by law reform bodies (see the New Zealand Law Commission report "Juries in Criminal Trials" para 78). There are limits to the extent to which juries should be admired (as I have suggested). Even so, if you say the jury was wrong in this case, you have to show that community values are such that what the defendants did was, in the context of what they reasonably believed, unreasonable.

We disagree, but we won’t say why

The majority in Morgan v R [2010] NZSC 23 (16 March 2010) do not explain why they reach the opposite conclusion to that of the dissenting Chief Justice. All Justices agreed that the issue came down to whether there was unfairness in admitting the contested evidence. The Chief Justice explained in detail why she thought there was unfairness. In the absence of a detailed response by the majority we are left feeling some disquiet about the soundness of the majority's conclusion.

Here the contested evidence was a statement previously made by a witness who was now hostile to the extent that he claimed not to remember the events in issue, and the making of any previous statement about them.

In contrast to the earlier position, under the Evidence Act 2006[NZ], any statement that is admissible is admissible to prove the truth of what it asserts. Where a witness is declared hostile, a relevant previous statement by that witness can become admissible. It may be elicited in cross-examination by the party that called the witness: s 94, or the Judge may permit such cross-examination even if the witness is not hostile: s 89. If the witness is hostile to the extent of refusing to be cross-examined, it is arguable that he is not a "witness" within the definition of that term, and his prior statement is hearsay. But even if it is not hearsay, fairness should require reasonable assurance that the statement is reliable when the s 8 discretion to exclude it is being considered.

The matter comes down to this (as expressed in the majority judgment):

"[41] Parliament has legislated to make previous statements of a hostile witness admissible as proof of their contents without adoption, presumably on the basis that the witness will be subject to cross-examination. The reality of that premise may differ from case to case. Parliament's policy decision should not be undermined by too ready a resort to s 8. It certainly should not be undermined on any generic basis. The ultimate question will always be whether the evidence is unfairly prejudicial in all the particular circumstances of the case, of which opportunity for realistic cross-examination will always be important."

The majority considered that in this case the witness had implicitly adopted his previous statement when he was cross-examined by the prosecutor. But this was on the basis of the last two questions in the critical passage, whereas the Chief Justice put these in a larger context and concluded that the witness had consistently refused to acknowledge the previous statement. Because she gave more detailed reasons, she is more convincing on this point.

The majority held that difficulties that might have arisen if defence counsel had attempted to cross-examine the witness on the prior statement were not such as to give rise to unfairness. No elaboration of that was offered. So we don't know why it was fair to admit the statement. The Chief Justice explained her conclusion by pointing out that in this case there was no other evidence sufficient to provide a reasonable assurance that the prior statement was reliable. The majority alluded to "a number of circumstantial features pointing to Mr Morgan's involvement, established elsewhere in the Crown's evidence" (para 45).

Still, the case is authority for, or illustrates, the following points: statements are admissible for their truth; admissibility is subject to s 8; there is a need for particular care to avoid unfairness when there can be no cross-examination of a hostile witness; the judge should hear evidence in voir dire before ruling on admissibility and should include an exploration of the witness's willingness to be cross-examined.


A significant omission from the factual narrative is whether the witness, a cell-mate of the accused, knew what the accused was charged with at the time he reported to the police that the accused had confessed. The Court of Appeal judgment summarises the facts ([2008] NZCA 537):

"[27] Mr Roskam shared a cell with the appellant shortly after the appellant was remanded in custody. Mr Roskam said in a statement to the police that he and the appellant had been watching a programme called “Police 10-7” which featured the robbery at issue in the present case. The appellant became very excited, and subsequently (some days later) told Mr Roskam that he had committed the robbery."

In the absence of this important fact, I will assume that the cell-mate knew, at the time he reported the confession, what the accused was charged with.

On that basis, the majority judgment contains a logical fallacy. It is fundamental but easily overlooked:

“[45]  One of the matters that will have significance in the s 8 assessment is the view that the Judge forms of the inherent reliability or otherwise of the statement in issue.  In this case Mr Roskam’s statement that Mr Morgan had confessed to him, in an apparently spontaneous manner, his involvement in the robbery was substantially supported by a number of circumstantial features pointing to Mr Morgan’s involvement, established elsewhere in the Crown’s evidence.”

This was not a case where the alleged confession contained details that only the offender could have known. It amounted simply to the accused, watching a news item about the robbery on television in a cell with Mr Roskam, allegedly yelling out “yeah, yeah!”, getting excited, and saying it was him (see para 8, the passage from the prosecutor’s cross-examination of Mr Roskam). Therefore, the “number of circumstantial features” that the majority say supported the reliability of the statement in issue (that is, Mr Roskam’s statement that the accused confessed) were not in the alleged confession but were indicative of the accused’s guilt. But here the majority confuse proof that the alleged confession had been made with proof that the alleged confession was true. The latter is not a substitute for the former. Just because a person may be guilty does not mean he is likely to have confessed.

The question is, "does this evidence, the confession, exist?" The majority say that if the confession existed it would fit with other evidence against the accused, therefore the confession existed. You can see the circularity. If the confession did exist, then the other evidence would indeed be useful in assessing whether it was true; but we have not got to that stage yet. To decide whether the confession existed, we would have to assess the credibility of the witness and that would include examining his honesty and whether there was any incentive for him to lie. He was endeavouring to retract his evidence, and that would be a relevant factor too. If the evidence of the existence of the confession was of sufficient value to have a tendency to prove that the confession existed, then it would be relevant evidence. Then its probative value would have to be assessed, and weighed against its unfairly prejudicial effect. The probative value could be assessed in Bayesean terms as a likelihood ratio: the probability of the witness saying that the confession existed, on the assumption that the accused was guilty, compared with the probability of him saying it existed, given that the accused was not guilty. If the witness had low credibility, he might have been almost equally likely to say that the confession existed, whether or not the accused was guilty. The likelihood ratio, or the probative value of his claim that the confession existed, would be very small. The risk of unfair prejudice would outweigh the probative value, because the jury would be likely to attach too much weight to the witness's claim that the confession existed.

After conscientiously advising care in the application of s 8, the majority misapply it to the probative value of the confession instead of to whether the confession existed. It is difficult to accept that a judgment flawed in this way can enhance the standing of the Court.

Of course, if Mr Roskam did not know what the accused was charged with, the other evidence of guilt would support his credibility. If that fact had existed, however, it would have been highlighted in the judgments.

 

People interested in the real politik of law will note the distancing between the Chief Justice and Wilson J, and the apparent collegiality between Wilson J and the other judges whose judgment he delivered (Blanchard, Tipping and McGrath JJ). Ho hum. And flash forward! "I think, but dare not speak."

Small, but perfectly formed – judicial acceptance of LCN DNA analysis

Updating the forensic use of LCN DNA analysis: R v Wallace [2010] NZCA 46 surveys developments and refers to R v Reed & Garmson (2009) EWCA Crim 2698. Brief mention of Bayes' Theorem, led the Court in Wallace to the following "insight" (para 87):

"As a matter of logic, what all this means for criminal trials is that with low level DNA profiling it is not so much a question of the particular technique or method. This can, if rather painfully, be explained to judges and jurors. It is the philosophy of the approach that is important: the uncertainty of the method of transfer does not diminish the fact of the matching DNA profile for the former does not invalidate the latter. The method of transfer however is of great moment for the court, whilst it does not necessarily concern the scientist."

That is to say, a match is a match is a match, but the court needs to know why there is a match.

The Reed & Garmson dicta were summarised in Wallace (para 97):

"Three points can be made about this decision. First, it is apparent that the Court of Appeal of England and Wales accepted that at least under defined conditions Low Template DNA can form admissible profiles, capable of reliable interpretation. Indeed, during the course of the Reed & Garmson appeals both appellants abandoned their general attack on the reliability of Low Template evidence using the LCN process. Second, the Court of Appeal of England and Wales prescribed what might be called a "safety" floor limit. Above the range described by that Court as being a stochastic threshold of 100 and 200 picograms results can be so produced which are capable of reliable interpretation. Third, in the view of that Court a challenge to the validity of the method of analysing DNA by the LCN process should no longer be permitted where the threshold limit is exceeded." [the Court's emphases]

There may be some circumstances in which whether there is a match can be an issue, but these are limited (on current understanding) to cases where the amount of DNA from the crime scene was so small – around 100 to 200 picograms (1 picogram = a millionth of a millionth of a gram) that random errors (called stochastic effects) could cast doubt on whether a match existed (applying Reed & Garmson).

In Wallace it was argued for the appellant that it was incorrect to regard the relevant sample of DNA from the scene as coming from a mixed source, and therefore it was wrong to conclude that it did not exclude the accused. This was an issue of interpretation and the evidence supporting the appellant's view was not in this case sufficient to cast doubt on the correctness of the verdict. The Court emphasised that DNA evidence must always be evaluated in the context of the other evidence in the case. In Wallace the DNA evidence which was the focus of this appeal was not significant in that context.

I should add that I have summarised the facts in Wallace a little inaccurately in the interests of simplicity. The DNA just referred to was actually alleged to have been from the victim, and was found on a metal bar, allegedly the murder weapon, which the prosecution claimed to link to the accused.

Another sample of DNA was from the victim’s shoelace, and was allegedly from the accused. However this latter DNA was (para 33, 38) thirteen times more likely to have come from the accused or a paternal relative as from any randomly chosen New Zealand male. The trial court, as emphasised by the Court of Appeal, made it clear to the jury that this evidence did not prove the identity of the accused as the murderer, but it was merely consistent with the prosecution case. In reality – and contrary to the prosecution’s disavowal at trial - this DNA evidence was highly probative of guilt.

To say that the DNA was thirteen times more likely to have come from the accused (or a paternal relative) than from a randomly selected male means that if the likelihood of a random match is x, the likelihood of a match with the accused or such relative is 13x. The Bayes likelihood ratio is 13. Despite the protestations to the contrary at trial, this evidence was highly probative of guilt.

The appellate court needed to be satisfied that the other evidence in the case proved beyond reasonable doubt that the accused was the murderer. This is because, in Baysean terms, if the prior probability of guilt was only 0.90 (let’s assume this is not proof beyond reasonable doubt), the DNA on the victim’s shoelace would have increased that to a probability of guilt of 0.99. That is, this DNA evidence would have removed a reasonable doubt if the other evidence only proved guilt to a probability of 0.90. It would have been highly probative and important to the prosecution case, and errors in its interpretation would have been significant.  

Criminal Procedure Simplification

For my comments on aspects of the New Zealand Law Commission's Criminal Procedure (Simplification) Project – Reforming Criminal Procedure discussion paper, click here.

When fallacies don’t matter

The United States Supreme Court has discussed the prosecutor's fallacy in McDaniel v Brown [2010] USSC No 08-559, 11 January 2010. After several different explanations of what this fallacy is, the Court held that in this particular case the probabilities were such that even on the numbers most favourable to the accused (Mr Brown) the jury had not been wrong to convict him. The case concerned DNA so of course – to put it in Bayesean (or, if you like, Bayesian) terms which are clearer than those used by the Court - in the absence of an innocent explanation for his DNA being at the scene the denominator of the likelihood ratio was small, the ratio large, and the probative value of the evidence large.

This illustrates the point that errors in reasoning do not always matter. Hopefully this will not encourage expert witnesses to avoid accurate thinking.

Overview (4) – Reasoning with uncertainty

We can be sure about very little.

"'We reason rashly and at random,' says Timaeus in Plato, 'because our judgements, like ourselves, have in them a large element of chance.'"

Montaigne, Essays, Book One: Chapter 50, above p 130.

Uncertainty can increase the closer something is scrutinised. Our quotidian lives would be impossibly disrupted if we indulged in a Plato-like thoughtfulness, so we normally come relatively untroubled to our settled beliefs. Forensic fact-finding is fraught with difficulties that it is necessary to ignore if decisions are to be reached. There are even studies that suggest that the demeanour of a witness is not a reliable guide to his truthfulness; you may as well toss a coin. Is the only instruction a judge can sensibly give a jury the familiar exhortation to "decide what evidence you accept and what you reject" and "use your common sense"?

Some efforts to increase the accuracy of judicial fact-finding have called in aid mathematics. This approach can be given a label of convenience, but the danger is that doing so will suggest that these efforts are just passing fads. Peter Tillers, an academic from the USA, calls them the New Evidence Scholarship, and identifies three generations of this:

"The first generation of the New Evidence Scholarship emphasized the heuristic uses of mathematical analysis of evidence; it emphasized that numbers (especially as used in probability theory) could illuminate the logic and structure of factual inference in general and of particular problems of factual inference.

"The second generation of the New Evidence Scholarship focused on mathematically-laden problems of scientific evidence (e.g., DNA evidence) and on problems of factual inference that seem tractable to statistical analysis.

"The third generation of the New Evidence Scholarship (NES) also uses mathematical argument and analysis. But this variant of NES does not require or expect consumers of mathematical analysis to do computations. Instead, NES-3rd uses mathematics and computations to develop tools for deliberation about inference, tools that do not require or expect the user of the tool to do computations."

The growing recognition of the pitfalls of ignoring the logic of mathematics, and particularly the logic of probabilities, coupled with appreciation of the impracticality of requiring judges and jurors to apply complex formulae in their deliberations, is the unifying theme of this scholarship.

Each of these applications of mathematical reasoning is exemplified by cases discussed here. Sometimes mathematics is ignored when it would have been of assistance, as in Brown v Attorney-General 6 March 2005 and also Wi v R mentioned in update to note for 4 July 2008. Refinement of DNA techniques has been accompanied by recognition of increased risks of contamination and of falsely positive results: R v Hoey 2 January 2008. Applications of Bayes' Theorem have pointed to the correct way for probabilistic evidence to be given, and also to how reasoning with conditional probabilities can be fallacious: R v Bain 8 June 2009. I have collected some observations on reasoning with probabilities in a draft paper available here.

Some criticism of the use of mathematical reasoning is advanced on the basis that, since the values of the variables are a matter for dispute, the equation containing them is of no use. More moderate criticism, of the same kind, acknowledges that the equations may have limited use, although it makes too much of claimed difficulties in calculating the probability values. The latest developments in this branch of evidence are directed at meeting the points made by such critics.

I give the last words to clever old Montaigne:

" The uncertainty of my judgement is so evenly balanced in most cases that I would willingly refer the decision to a throw of the dice; and, after giving much thought to our human weakness, I observe that even sacred history has left us examples of this custom of leaving chance and fortune to make the decision in matters of doubt: 'and the lot fell upon Matthias' [Acts I, 26]."

Montaigne, Essays, Book Two: Chapter 17, above p 216.

Overview (3) - Balancing

Application of rules is the least interesting part of judicial decision making. Vastly more intellectually absorbing is the process of reaching a decision by balancing of conflicting rights, values or interests.

"... balancing introduces order into legal thought. It requires the judge to identify the relevant values; it requires the judge to address the problem of the relative social importance; it requires judges to reveal their way of thinking to themselves, as well as to others. It facilitates self-criticism and criticism from the outside. As Judge Coffin wrote:

"Open balancing restrains the judge and minimises hidden or improper personal preference by revealing every step in the thought process; it maximises the possibility of attaining collegial consensus by responding to every relevant concern of disagreeing colleagues; and it offers a full account of the decision-making process for subsequent professional assessment and public appraisal. [See Frank M Coffin, "Judicial Balancing: The Protean Scales of Justice," 63 NYULRev 16, 25 (1988).]

" Indeed, balancing is a way of thinking; it is a conceptual mentality; it is a process that leads to decision. It requires dealing with how genuinely problematic is the situation created by conflicting values. [See Frank Michelman, "The Supreme Court 1985 Term – Foreward: Traces of Self Government," 100 Harv L Rev 4, 34 (1986).]"

Aharon Barak, The Judge in a Democracy (2006), p 173.

Relatively simple balancing can occur where one person's interest is weighed against another's. An example is the determination of whether the actus reus of disorderly behaviour includes the act of singing outside the complainant's house, on the street, in daytime, and intentionally causing the complainant, a night shift worker, to lose sleep: Brooker v R 4 May 2007. Balancing here was controversial: two judges applied balancing and two applied rights limitation; see further my note dated 4 May 2007. Balancing also occurs where the issue is media access to court proceedings: Rogers v TVNZ 19 November 2007.

A more orthodox environment for balancing is the decision whether evidence should be ruled inadmissible because it was obtained improperly. A case may require two balancing exercises: one to determine whether there had been an impropriety in the obtaining of the evidence, and, if there had, the other to determine whether that evidence is admissible: R v Singh 2 November 2007.

The exercise of anticipating the result of an admissibility decision can be approached by lawyers in different ways: by studying how judges have arrived at such decisions, or by studying the results of those decisions and identifying the pattern of precedents. A statute may tell judges what factors to take into account, and what the criterion for decision is (for example, whether exclusion of the evidence is proportionate to the impropriety (s 30(2)(b) Evidence Act 2006[NZ]). Anticipating the weight that will be given to each factor and where the balance will rest is not easy without reference to precedent, and for this reason lawyers may more easily focus on the results in similar cases.

The Supreme Court of Canada has offered an explanation of the way judges should reach their decisions in balancing cases involving improperly obtained evidence: R v Grant 18 July 2009, and has illustrated it in R v Harrison 19 July 2009. The conceptual model of the decision was described by the Chief Justice as a "decision tree". It has difficulties which are reflected in its three-dimensional structure. In New Zealand there has been a transition from the pre-Bill of Rights Act 1990 spectrum model to a period of rule application during the early days of the Bill of Rights, to a discretionary approach now enacted as the abovementioned s 30 Evidence Act 2006. The current balancing is two dimensional and precedents can easily be placed on a diagram which makes predictions of results a relatively simple task. At least, it's a simple task for people who are not afraid of diagrams.

Another form of judicial balancing occurs where a decision has to be made about whether the probative value of evidence is outweighed by the risk that the evidence will have an unfairly prejudicial effect on the proceeding. This balancing is familiar throughout the jurisdictions that have laws of evidence (not including, for example, the civil law countries of continental Europe, although a similar balancing is seen in Germany: Gafgen v Germany 3 July 2008). It is not an easy weighing exercise on which to construct a conceptual model as in each case the judge's perception of the risk of prejudice is very much a personal matter. There was unanimity that certain evidence should be excluded under this weighing exercise in the New Zealand Supreme Court in Bain v R 12 June 2009, but, before the case got to the five judges of that court, four judges had held the disputed evidence admissible. It is arguable that this weighing exercise is misconceived because it leaves room for significant risk of prejudice if the probative value is assessed as high. The decision could better be put as whether admission of the evidence would create an unacceptable risk of unfairness. That would avoid weighing altogether.

Overview (2) – Standards of proof

[I have seen that this is an oft-visited posting. For revision and updates, check the link to the introductory essay. Other entries this month are compiled there. Where a case is named with an associated date, that date refers to another posting on this site.]

The expression "standard" of proof refers to the level of proof required, and is not to be confused with the expression "burden" of proof, which refers to which side has to prove an issue. The usage is, who has the burden, and to what standard?

There are at least two standards of proof in criminal law: proof beyond reasonable doubt, and proof on the balance of probabilities. Sometimes more standards are recognised: the standard of a reasonable doubt, as where a presumption applies, rebuttable by proof to the contrary (this is sometimes called, confusingly in the light of the usage I have set out above, the evidential "burden"); and a flexible approach to the balance of probabilities, where many standards or levels of proof might be recognised as appropriate depending on the seriousness of the issue. As to proof of a reasonable doubt, this is not recognised as a standard in New Zealand (R v Hansen, 19 September 2005 and 20 February 2007) but it is in the United Kingdom (R v Lambert 19 September 2005, 12 December 2008, and s 2 Drugs Act 2005[UK] which is yet to be brought into force by a Commencement Order). Other standards are sometimes referred to on appeals, where the appellant must satisfy the court that in the absence of an error at trial there was a "real chance" of a more favourable verdict. This "real chance" is less than on the balance of probabilities (Chamberlains v Lai 11 September 2006).

How do we know when a standard has been reached? At this point Montaigne again intrudes his mischievous little pen:

"'As the scale of the balance must necessarily sink when weights are placed upon it, so must the mind yield to clear proof' [Cicero, Academica, II, 12]. The emptier a mind is, and the less counterpoise it has, the more easily it sinks under the weight of the first argument. That is why children, the common people, women, and the sick are particularly apt to be led by the ears."

Montaigne, Essays, Book One: Chapter 27, above pp 86 – 87.

Naughty Michel.

Courts are notoriously unwilling to assist juries by elaborating on the meaning of "beyond reasonable doubt". This is shameful. It is one of the law's little dishonesties that favour the prosecution. It is said that it is no use likening the standard to a given probability of proof, because people disagree on what the probabilities are. This is daft. The variation in what people think proof beyond reasonable doubt equates to in probabilities is astonishing (R v Wanhalla 25 August 2006). If people have not even a general idea of what beyond reasonable doubt means, how can they decide whether proof meets that standard?

Aversion to mathematical analogies is one of the law's shameful inadequacies. It reflects the stage in school when pupils, at age about 14, are divided into "arts" and "science" streams. Lawyers and judges are usually from an arts background, and have been brought up to think they cannot achieve an understanding of mathematics or science. At the same time, and ironically, lawyers and judges pride themselves on their logical reasoning. It will be interesting to see what moves are made in this area by the High Court of Australia now that it has a Chief Justice (Bob French) who has a science background. (Naturally Australian counsel are alert to the need for change: my thanks to Jeremy Gans for drawing my attention to Clarke v R [2009] HCA Trans 336 (11 December 2009) in which special leave to appeal was refused by Crennan and Bell JJ.)

Areas of uncertainty exist. There is some doubt as to whether a standard of proof applies to a "risk", such as the risk of danger to the public if a person is released on bail (O v Crown Court at Harrow 31 July 2006) or on parole (Re McClean 19 July 2005). There is also doubt as to what, if any, standard of proof applies to proof of similar facts (R v Perrier and R v Holtz 14 October 2004; HML v R 26 April 2008). Sometimes the choice of standard appears inappropriate, such as the application of only the balance of probabilities to proof that evidence tendered to the court was not obtained by torture in another country (A v Secretary of State for the Home Department 9 December 2005), and there is still work to be done on what if any standard applies to the likelihood of unfair trial or of inhumane treatment in a foreign jurisdiction if the person is extradited (Gomes v Trinidad and Tobago 1 May 2009).

It seems intuitive that serious issues should require proof to a higher standard than issues that are less serious. This leads to difficulties where the law insists that a standard, usually the balance of probabilities, is fixed. The Supreme Court of Canada does not acknowledge that the standard is flexible, and it maintains that the degree of caution with which proof is assessed is the same whether the issue is serious or less so: FH v McDougall 19 October 2008. The opposite position is taken in the Strasbourg court: Saadi v Italy 11 December 2006, and this is probably the same as that taken in New Zealand: Z v Dental Complaints Assessment Committee 25 July 2008.

Another area where discussion of standard of proof has occurred is the fundamental requirement for evidence to be "relevant" before it can be admissible. Relevance usually means having a tendency to prove a matter of consequence to the determination of the proceedings. Variations on that formula occur, but the essence is "a tendency to prove". Serious discussions occur in the academic literature about what this means. What if an item of evidence has a small tendency to prove one thing, but a greater tendency to prove another? Is it relevant as evidence of the first thing? What if the item of evidence has only a very slight tendency to prove a matter in issue? In Bain v R 12 June 2009 the Supreme Court of New Zealand split on whether indistinct sounds were relevant on a recorded telephone call, the majority holding that to establish relevance it is necessary that the interpretation of the evidence advanced by its proponent be "reasonably" open or available. Further thought was given to relevance in Wi v R (see update to note for 4 July 2008) where the accused sought to adduce evidence that he had no previous convictions. This sort of problem calls for statistical analysis, but the New Zealand Supreme Court did not take that opportunity.

Overview (1) – Interpreting the proviso

This introductory part continues with my overview of important topics that are often addressed by the appellate courts in criminal cases. Where I refer to cases with accompanying dates, those are dates of blog postings on the Blogger site; on the WordPress site the dates may be a day out due to clock settings on the servers.

First overview topic ...

Interpreting the proviso – what is a "substantial miscarriage of justice"?

For around 100 years courts have dealt with criminal appeals by determining whether an error in the trial court amounted to a "substantial miscarriage of justice", but very little of that time has been spent examining what that phrase means. It has been found to be a difficult concept. Some courts have tried to achieve clarity on a case-by-case basis, but the alternatives of legislative change (in the UK since 1995 the criterion has been whether the conviction is "unsafe"), and of ignoring the recent judicial efforts by going back to fundamental statutory interpretation, have also been taken.

Courts that continue to address the meaning of substantial miscarriage of justice identify one characteristic as whether the error could have affected the outcome of the trial in a way adverse to the accused. This leads to the further question: does the appellate court ask whether the jury at trial could have reached a different conclusion, or does it ask whether it – the appellate court – would have reached that different conclusion? There is a split between courts according to which answer they give to that question.

The Privy Council has remained faithful to the traditional view that the answer is that the appellate court asks whether the jury could have reached a different conclusion. The High Court of Australia has changed its approach (Weiss, noted 16 January 2006) and, scrapping the case law and returning to principles of statutory interpretation, it decided that the appellate court asks whether it itself could have reached a different conclusion. The Supreme Court of New Zealand has reacted against the Privy Council's traditional treatment of Bain (see, for example, note for 9 July 2009) by adopting (in Matenga 20 July 2009) much of the High Court of Australia's new approach. In turn, on a subsequent appeal from New Zealand (Barlow 9 July 2009), the Privy Council has felt itself obliged to follow the new line but has emphasised that under this new approach the result in Bain would have been the same. The lingering problem for subsequent would-be appellants is how to get the Supreme Court to hear an appeal if it is based only on advocating a different interpretation of the evidence from that taken by the Court of Appeal.

Independent of the effect on the outcome of the trial, another characteristic of a substantial miscarriage of justice is that the error at trial deprived the accused of a fair hearing. The Privy Council made the right to a fair hearing absolute, and that is uncontroversial now, although before then another approach, in which the accused's right to a fair hearing could have been balanced against the rights of victims and of society, was possible. It is one thing to hold that the right to a fair hearing is absolute, but what is a "fair" hearing?

The meaning of fair hearing will need to be consistent with the place of that concept in the "substantial miscarriage of justice" model. The model that has emerged – and which may well be unintended – in Matenga is that the first question is whether the hearing was fair; if it was, the second question is whether the result of the hearing may have been different in the absence of the error. At first glance this Matenga model sounds fine: it places fairness of hearing as the primary requirement, and that looks like being an appropriate reflection of its quality of absoluteness. It requires the appeal to be allowed if the trial may have been unfair. But the difficulty is, what does fair trial mean if the trial may have been fair but also may have contained an error that amounted to a substantial miscarriage of justice? Logically, it seems odd to make the set of fair trials intersect with the set of trials involving a substantial miscarriage of justice.

A clearer concept of a fair hearing would require a different relationship between fairness and substantial miscarriage of justice. Under this model, a fair trial would be one in which there was no substantial miscarriage of justice, and that means one in which the law was properly applied to facts that had been determined impartially. Here, an impartial determination of facts presupposes an unbiased tribunal and also a tribunal that is not deflected from impartiality by errors of law. There is thus room for some errors of law, and it is usual to recognise that perfection is not always attained, but errors of law that affect the impartial determination of facts are not permitted. Nor are errors of law in applying the substantive law to properly determined facts, if such errors may have affected the result in a way adverse to the appellant. A trial may seem to have been fair but afterwards fresh evidence may cast doubt on the correctness of the verdict; in such cases there would be a substantial miscarriage of justice, albeit with the trial only having been retrospectively unfair, and the appeal would have to be allowed.

Introduction (4) – the quest for clarity

How wonderful it would be if laws could be simple. The quest for simplicity struggles against the head wind of reality: life is complicated. Laws have to be applied in individual cases, often to facts that were never anticipated by legislators. It is inevitable that even simple laws will accrue complex shades of meaning as judges adapt them to meet the requirements of justice in particular cases in accordance with an imagined legislative intent.

"The truth is that a close knowledge of the facts is essential, not because of the precedent system, but as a prerequisite to doing justice in the particular case. The facts are the fount of individual justice."

EW Thomas, The Judicial Process (2005) p 321.

If simplicity is an impossible goal, clarity is not. Clear laws are essential, no matter how complicated they may have to be. There has to be a lot of law, so at least it should be clear law. Otherwise,

"As we once suffered from crimes, so now we are suffering from laws."

Tacitus, Annals, III, XXV, quoted by Montaigne, Essays, Book Three, Chapter 13.

There is a risk that

"Anything that is divided into minute grains becomes confused."

Seneca, Letters, LXXXIX, quoted by Montaigne, ibid.

The counter-measure is the quest for clarity.

Montaigne's opinion of laws could have been written, not in the sixteenth century, but today:

"Now the laws maintain their credit, not because they are just, but because they are laws. This is the mystical basis of their authority; they have no other. And this serves them well. They are often made by fools, and more often by men who, out of hatred for equality, are lacking in equity, but always by men: vain and unstable creators. There is nothing so grossly and widely, nor so ordinarily faulty as the laws."

Montaigne, ibid.

Schopenhauer's advice to writers needs to be borne in mind by law makers – legislators and judges - as well as by law teachers, students and law commentators:

"Obscurity and vagueness of expression is always and everywhere a very bad sign: for in ninety-nine cases out of a hundred it derives from vagueness of thought, which in turn comes from an original incongruity and inconsistency in the thought itself, and thus from its falsity. If a true thought arises in a head it will immediately strive after clarity and will soon achieve it: what is clearly thought, however, easily finds the expression appropriate to it. The thoughts a man is capable of always express themselves in clear, comprehensible and unambiguous words. Those who put together difficult, obscure, involved, ambiguous discourses do not really know what they want to say: they have no more than a vague consciousness of it which is only struggling towards a thought: often, however, they also want to conceal from themselves and others that they actually have nothing to say."

Schopenhauer, "On books and writing" in Essays and Aphorisms, above pp 204-205.

Much of legal discourse requires clarity about things that are not absolute. The exercise of discretion, judgment, requires a relinquishing of faith in absolutes. Petulant old (at 42 years) Nietzsche reminds us:

"Really, why should we be forced to assume that there is an essential difference between 'true' and 'false' in the first place? Isn't it enough to assume that there are degrees of apparency and, so to speak, lighter and darker shadows and hues of appearance – different valeurs
[values], to use the language of painters?"

Nietzsche, Beyond Good and Evil, above p 35.

So, don't be afraid of complexity, but strive for clarity.

"Everything that can be thought at all can be thought clearly. Everything that can be put into words can be put clearly."

Ludwig Wittgenstein, Tractatus Logico-Philosophicus, para 4.116.