 Good evening everybody and welcome to this year's The Hate It a Nancy Lecture. Who was the hated to Nancy? He was an extraordinary man who lived a very long life and did a large number of diverse things. He was a Dutch man but he ended up living in the Channel Islands ac mae'n ddechrau bod oedd i chi wedi i ddweud o'r ddodd hynny, mae'r drosio, mae'r ddechrau, y deitist, y expert ar hyn, a'r barysdy. Roeddwn i chi'n gwneud y 35, mae'r ddweud yn cael ei wneud i'r ddechrau. A dyna'r ddweudio cyfnod, ac mae creu'r ddaf i gweithio'r ffondiad ar gyfer mhawr, yr oedd ystod i'r cwestiynau yn meddwl. Fy enw i'n meddwl yma, ynddo'r faniatiaeth hwnnw'r dda'r ddechrau Hedon Delance yn ymgyrchio'n Cymru, ein ffond, ddim yn ymddangos hwnnw'r ffaniatiaeth hwnnw, a gael y gallu gyda gweithio gyda lechau sydd wedi gweld mewn meddwlol iawn. Mae'n gweithio gael yn gweithio'n gweithio'n gweithio'n gweithio'i gweithio'n gweithio'i gweithio'n gweithio'n gweithio'n gweithio'n gweithio. Jeremy Order, y Prif Weinidog yn cyfnodol cwrinifol acedemi, sydd yn y Gweithio Unifarchadeffordd yn ddefnyddio'r cyfanigol fydd yn gwneud cyffredigol i fynd wedyn yn gwneud ei gweld yn ddweud gan gweithio'r cyfanigol yma, ac mae'n gweithio'r cwmysgol cymdeithasol ar gyfer y Llywodraeth gynradd yn ystod gyda'u gweithio. Mae'r Cyngor Llywodraeth Cymru yn ddau'r cymdeithasol yma, ac mae'r cyngor Llywodraeth Cymru yn ddau'r cymdeithio'n ddau'r cymdeithio diwethaf a part of the concerns about expert evidence and the need to reform the legal rules has been fueled by the number of the shaken babies and alleged copte cases, the Clark case, the Cancrix case and certain other cases. Mae'n gweithio gael yma ar gyfer Jeremy Fudd, ond yn ysgolion y bydd yn 40 ym mwyn fydd yma, ac mae'n bwysig ar hyn o gwbl yng Nghaerffordd. Mae'n gweithio'n gweithio'n cael ei amser o'r cymdeithasio ac yn gallu gwybod ar hyn yn dweud i'r hyn i ddim yn ysgolion gweithio ar hyn oedd. Ddweud i'r ddweud, John. Mae'n ddweud i'r ddweud i'r ddweud i'r ddweud i'r ddweud. I will explain a brief word about the Law Commission because although the organisation will be very familiar to law students amongst the audience here it may not be to medical students and a professor of medicine who may also be inside the audience here. Dyna y Llyfrgell Llyfrgell, oherwydd i ddechrau y Llyfrgell Llyfrgell, maen nhw'n amser oedd y Gweithio Llyfrgell yn 1965. Mae'n cyd-degynghwm o'n bwysig o'r rhaid o'r rhaid i'r Llyfrgell. Mae'n ddechrau i'r Llyfrgell ynghwyl, yma, nid o'r rhwng o'r Llyfrgell. Mae'r ddechrau i'r llaw o'r rhaid i'r Llyfrgell o'r rhwng o'r rhwng o'r Llyfrgell, a'r ysgrifennu o'r rhwng o'r Llyfrgell, ac mae'r bwysig yn weithio i gyn deployedgo antherogion i ddfightwlol o gymryd yn bwysig yn bobyn. Dyma'r gwrs powai i'w du o bobl mae'r cy Ergebnid. Mae'n ei ddechrau i fynd y cwestiwn gael gweithio o'r cyffredin â'r ysgwrddau, a'n gweithio i unedig yw ddechrau fyrddau. Yn y bydd y mae'n i gweithio i gynnwys ymddiol, roedd y sygu cyfnodau i gael cwmwysgwymu sydd wedi gwneud y dymlo y gwahol o'r hyffredin o'r law ac oes y cyfnodau cyfnodol cyd-radw yn cael ei rhaid iddo. A wnaeth ar gael cyd-radw wedi gwneud ein護ch gyda'r bydion peth yn cymwylltew hwnnw ond ond ond rhaid yn cyfnod. Nghymbeidio ar y ddechrau, diwethaf i gyrffffordi, cyfnod, gwpwsio yn y cyfnod i gael. ac yn ddildig, maen nhw'n mynd i wneud hynny diwylliant yn dweud a blynyddiant y cyfnodd gylau. Ac yn cael ei blynedd ac am hollwch ydi ynddo, mae'r 80 reffléidau, a'r hollwch sy'n ffordd o'r oedd y ddechrau gwahanol sydd gennymu'r gweithreddau. Mynd yw'r unrhyw Ie'w dda'r byd算 i chi'n cyfnodd efallai i wneud y byddiau isi tynnu. Rhaid, mae'n gweithio i gyd o'r daud i gweithio'r refflect o'r hollwch... yn y rhan o'r cyfnod, y cyfnod y Llywodraeth ym Mhwyl. Mae'n ddaeth i'r parwmysgol yng Nghymru, i'r ddraffydd, ond, y byd yw'r cyfrifiad o'r parwmysgol i'r cyfrifiad, ond mae'n ddod yn y proses. Ond ystod, mae'n rhan o'r cyfrifiadau, ond mae'n rhan o'r cyfrifiadau, mae'n rhan o'r cyfrifiadau, mae'n rhan o'r cyfrifiadau, ac mae'n rhan o'r cyfrifiadau, a'n rhan o'r idea o'r parwmysgol i'r parwmysgol ac yn ddod o'r cyfrifiadau, yn ddaeth i'n gwneud o'r proses. Ond yn rhan o'r ddod, mae'n ddod o fe allans lleoedd o'r cyfrifiadau a'r ddiddordeb llyfa gan gymhreun iaith y Prifysgol. Ondabyn ni'n rhoi arweiniwch i'r lleoedd unol o'r parwmysgol i'n mynd i gynnau'r tîm a'u digud i ddod yw bod y byddai'r projekty nid oes yn y gydag arfer, ..would you like to look at it, how did that come about? I'm going to start off with what might seem like our own peculiar place, which is not it, ..and, for those of you who haven't heard the场on accusations of data manipulation.. ..at the University of East Asia kinematic research, for those of you who haven't heard such accusations, ..we're welcome to plan it here on Earth because in fact these accusations have been much written about in the newspapers as you know... Unrhyw o'i fatigol iawn i'n ddech 사람들ol i Llywodraeth a allan i ddweud y Prifני cyrraeth a chyfnod a tŵr yn ein bodi yng ngoswm iawn. Ond y tŵr ar mwynnwys cyrraeth yn gyda hynny yn gweithio cyfan, ond mae'r ddweud yn ddullai'n ddech, ond mae'r cyfeirio yn gyda kolod arythleniadau mwyaf. Donodd i'w ddechrau i wneud o ddweud i'n ddweud i ddweud i'r lle cyrraeth a'r lle amser o'n fathafol iawn. Mae'r prysgol yn gweithio, a'r bwysig ddim yn ddigon i'r gwaith, yn llwyth i'r cymryd, ac mae'r ddim yn cael o'r byd, mae'n ddorol gwaith o'r pethau. ond mae'r ffordd yn y dwyliadau o rhan o'r dwyliadau o'r ddweud o'r ystyth i'r mhwysydd mwyaf i'r perffesion ar gyfer ystod o'r cyfrifonau. Felly mae'n ddigon ni'n gweithio yng Nghymru yn mynd i'r gwybod hynny? y gallwch chi'n ddweud i gyfnodd. Mae'r byw'r ysgrifennu sydd y Comitee Ffant yw'r problematig o'r ffordd yw'r ffordd o'r ffordd. Felly, y gallwch chi'n bwysig o'r statysgwysu yn cyfnodd mewn data mewn. Yn cyfnodd mewn data, rydyn ni'n gwybod i'r ffordd yw'r lawid yn ystafell, yn cyfnodd, ond nid o'r gwaith yng ngyfnidol, er mwyn yno'r ddeuedd o'r gwaith ymlaen nhw'n bydd ymlaen nhw'n gwneud o'r fudio'n gyrfaen fel y dda'u fel y ddweud o'r lŷn yn bach o'r ddweud o'r rhigor, ond oes o'r cymdeilio'r cymdeilio, oherwydd o'n bach o'r cymdeilio'n gyrfa, oherwydd o'r cymdeilio'n cymdeilio'n cymdeilio. O ran ydych chi'n gweld? Felly, y clynydd y gallu cyfgleddau y Llywodraeth Llywy oherwydd y clywed mewn oeddiadol iawn a ulecaeth. Yma hwn yn oed. Mae hwyddi'n ceisio yn y ffrind ohono. A oedd y clywed yn gweld cyfgleddau. Mae hwn wedyn ei wneud yn y Prifysgol Gweithgaeth, oherwydd ylltyniad ar y bryd o'u cyfrwyr hyn yn rhaid. Er bod y cael ei'r gweld iawn i'r find gweld hynny oherwydd unrhyw mwyaf ar hynny yn ei hynny. Mae'n gwybod yn gweld yn gwybod, a'r gweithio'r ffordd o'r dda, mae yn dda i gael arall. Ond yw'r wych yn cael ei ddweud, a'r ddweud o'r ddweud o'r gweithio'r dda. Mae'n ddweud, ond mae'n ddweud, ond mae'n ddweud. Mae'r ddweud'r ddweud o'r potod, yn y ffwrdd, yn y ffordd o'r ddweud, o'r cyffredin iawn, dda i'n ddweud y gweithio ar y ffordd. Felly, os ydych yn cael y cyfnod bwysig wedi'u gwirio dod o'r hyfforddiadau hyfforddiadau hyfforddiadau, yn cyfryd i'r cyfnod o'r hunain, mae hynny ymwysig yn ei ddod yn ei ffordd, felly mae'r gwirio'n ei wneud i'w gwirio'n ei wneud i'w gwaith. Mae ymddwys i'r cyfnodau a'r hyfforddiadau sy'n cael ei ffyrdd yn cael ei bwysig, ac mae'n bwysig i'r gwirio sydd fwyllgor cyfnodau sy'n cyfnodol, felly mae'r cyhoedd yn ymdfaen i'w ddweud y ddweud y ddweud y ddweud ac, yw ddweud y pob datblygu o'r cyhoedd yn dda'r ddweud yn y ddweud y ddweud y ddweud, ddweud am bwysig, a'r ddweud y ddweud eu ddweud y ddweud. Byddwch chi'n ddweud y ddweud o'r cyhoedd a'i bwysig, neu'r ddweud o'r ddweud, i gyrraedd yr arbennigol yn cael ei maelodau, Oherwydd y bydd yn yma yn ymdindigol, yna'n gweithio'n ddefnyddio'r gweithio'i gwneud. Ymdindigol ar gael o'r gwneud yma yn ymdindigol i gyflwyno ac os yw'r gwneud, ydych chi'n gwneud, yw'r amdain ac yn gweithio. Yn ymddindigol, dyma'n gweithio'n gweithio, ond mae'n ddweud yma yn ymddindigol, answered on climate change and for your grasp of statistics were a bit hasty, even though of course, a these in some parts of your investigations, you may rely on some statistical evidence. And for all experts of course, other than in a rather narrow disciplines, it will inevitably be the case that you do rely, at least to some extent, on assumptions about what other people are saying in their areas of expertise. I work on the criminal law. I recommend changes of the criminal law. I assume in all of that, of course, that a change of the criminal law may have some effect of a deterrent kind, preventing people from committing offences they might otherwise have gone on to commit or committing greater numbers and so on. So I'm making assumptions about human nature and about which I know no more than the average person. I'm not an expert on that particularly, I'm an expert on the criminal law. I may make assumptions on which my work at least in part relies, I hope not wrong. But that's just an illustration of how we can all, however expert we may be, be forced to rely explicitly or implicitly on data that relates or expertise that relates to another field of expertise. It might have been right perhaps for Baron for Hayden de Lacey, who seems to have had this extraordinary facility across such a wide range. I think John Ambit has mentioned that he spoke about five languages, although being a Dutchman perhaps perfectly normal, but nonetheless he has this extraordinary facility and perhaps to him this would not have been a problem. But for most of us in this world in which we increasingly specialise, it is a problem. Now anticipating that a problem of this sort would loom large, I commissioned as part of our consultation paper a very brief paper from Professor David Hand, who's the head of the Royal Society of Statistics and Professor of the Imperial College. He very came up with some guides as to what one should do when one has to rely on statistics in particular. And how one should present one's results, where that kind of reliance is involved. And as you can see here, as you likely said, statistics are commonly used to match data to theories. That was certainly true in the UEA case. And so the possible sources of error must be clear for positive results, what percentage, what proportion. Theories must not predict too wide a range of possible outcomes. Perhaps a problem with astrology, for example. What was infinitely variable in the range of possible outcomes. Although not an expert discipline for other reasons, but nonetheless. Any missing or imperfect data must be accounted for. Professor Hand was of the view that even Newton, for example, used to present rather suspiciously perfect data with apparently no false negatives or false positives. And in his view, there's been in Newton's work a certain element of illegitimate clearing up of what all might be regarded as messy data. But nonetheless, it doesn't invalidate the theory, of course. But it just, what should have reasons to be suspicious in Professor Hand's view, no doubt rightly, where data appear to be absolutely perfect and there are no false positives, false negatives and so on. And so those imperfect or missing data must be accounted for. We'll come across a bunch of cases where a number of these problems on the bullet points above there have led to the evidence being undermined and criminal convictions being overturned. And of course, at the bottom there, possible sources of bias, but that's pretty self-evident. It's a possible source of unreliability. Now I'm going to get a little bit more into the legal elements. And what I've done here in a very rough and ready way, really, is to distinguish different ways in which one might fall down, if you like, or fall short of expectations, really, when one is exposed to the full glare of the criminal trial or if one's been under undue pressure from one side or the other to present statistics in a way that supports certain conclusions rather than others. Now, there are three ways, I've divided up, this is really just something that occurred at the top of my head, it may not work terribly well, but what I've done is distinguished three kinds of problem or issue that may crop up in relation to expert evidence given in criminal trials. One, of course, is subjective honesty. I mean, there's always a risk that someone will present data in a dishonest manner, but that's relatively small one, I hope, as a problem. It's always possible. Now secondly, there's what I call rather obscurely subjective, objective things. Now what does that mean? Well, what that means is where you're not being dishonest. You honestly, subjectively believe that the data supports a particular conclusion, but there is a dimension of objective fairness to what you're doing. In other words, there is an aspect in which it could only be fair to present data you honestly believe supports an conclusion if you put forward all the data, even the bit that doesn't support you particularly well or the bit you can't explain or something like that. That's the objective element in being fair when you present your evidence. It's not enough that you think of the data support what you're saying that you must have regard to the objective element of fairness, too. And finally, there is the purely objective question of reliability. How, well, whether you're purely objective may be over the question, I don't want to put too much stress on that, but there is a basic question, if you like, about how reliable certain kinds of evidence is. Irrespective of what you may think, or what may be fair to the other person in terms of what you reveal about your work, there is just the basic question of objective reliability. So there are those three dimensions, all of which may come into play in one way or another in criminal, or indeed for that matter in civil cases. A lot of what I'm saying here will not be of special relevance to criminal cases. It will be relevant to civil cases, too. I'll come back to that right at the end. But obviously, from my point of view, what's important about it is how it impacts on criminal cases. Now, well, for what purposes need scientific evidence or other kinds of evidence be sufficiently reliable? Well, I think you need to distinguish a number of things. One, for example, is scholarly speculation, although in fact I could have just put a speculation in my case, the scholarly bit is important here, because whatever flaws may have been found in, for example, just to use that example, the UEA, Climate Change Research Institute evidence, whatever flaws there might have been, it seems to me it was easily accurate enough for the basis of a very leech scholarly speculation or political debate or something like that. It was sufficiently reliable, even though the data were messy and various other things about it, for that kind of purpose. And the degree of reliability that something has to have in order to make a contribution of some kind to a scholarly debate is not very high, it seems to me. I mean, obviously, it can't be wholly unreliable, otherwise why are we putting it forward, but it doesn't have to meet a very high criteria. I mean, that just seems to me to be a matter of common sense. But again, it's something that has caused difficulty in criminal cases and civil cases, because, on occasion, articles written with scholarly speculation in mind have been used in other contexts and their conclusions spun out across a larger range of activity, where really they don't necessarily belong. Then there's civil litigation, of course, where lots can serve whether something is provable on the balance of probabilities and it may be that some evidence is appropriate for that purpose in a way it would not be in a criminal trial. So it might be, for example, that evidence of the way that, and I'll come back to the example evidence, the way that somebody walks, the fact that they've got a distinctive kind of walk or get might be sufficient for the person of identification in a civil case, but might not be sufficiently up to it for the purposes of a criminal case. And I hope I won't forget to come back to that example because it's quite interesting one a little bit later on. And finally, of course, we then have criminal litigation and there are two bits to this, really. Criminal litigation, as you know, it all involves proof beyond reasonable doubt and so your scientific evidence, if it's going to form a part of that, you might think has to be, in terms of objective reliability, extremely highly reliable. But of course, actually, the position is more complicated than that. It certainly has to have an extremely high degree of liability. If you are, to use the common phrase, a prosecution's star witness, if effectively the whole case will fall apart where it's not for you and your evidence. And of course, that's not uncommon in cases where, for example, DNA evidence is conserved. I'll come back again to that. But if you're, again, if your evidence is only supporting evidence or suggestive, it supports the prosecution's case but it's not the main bit of it, well then maybe the objective reliability of it does not have to be quite as high as that of all that. So long as it's made clear, obviously, it's a period of court, what degree of reliability it does have. Again, things may be different on the defence side. If we turn things around, all the defence has to do, of course, as you know, is throw a bit of sand into the prosecution's face. All you have to do is to cast a bit of reasonable doubt on the conclusion, and it might be that a scientific theory of some kind, although perhaps a little bit coming from left field, is sufficient to do that. So it's not always the case that when we're thinking about scientific evidence, we're thinking about, although even this is a dangerous example, DNA evidence, which is thought to be proof, beyond proof, really, of extremely high degree of reliability, not all scientific evidence is like that, but it doesn't matter, necessarily, for the purpose of the criminal trial. It depends what you're trying to do in that trial. Are you open? Well, there are some controversial cases that a number of you may know about here, and I don't really want to chew over the fact in relation to those in the course of my talk, although you may want to come back to the question, and that's fine. Certainly one of the expert, prosecution experts in the Sally Clark case, you'll remember, which was a cop death case, where there was more than one cop death in the same family, and the implication was that for that reason, it was not actually cop death, and there was an implication of criminal intent. Now, one of the prosecution experts failed to disclose a test result for one of the two infants involved. Now, at the very least, I suggest, that transgresses what I called the subject of object development of fence. That expert might have thought that that doesn't matter, that it's really unhelpful to disclose that evidence, or whatever it may be, but as a matter of objective fence, it should have been disclosed. Another expert in the Clark case faced what I referred to earlier on, that is that this expert did not allude to the fact that cop death can be affected by environmental and genetic factors, as well as whether the baby is positioned in the cot or not in the cot next to the mother, or whatever it may be. In other words, person care issues. It can be affected by a wide range of issues. Now, I'm saying that I'm not reporting to being an expert on this. This is sort of 78th hand reporting or something. The purpose of mentioning it is to give you some examples of how we can run into the kind of problems I mentioned earlier on in real life cases. That was the first point there on the bullet point. It just simply explains, I think, the fairness issue in a little bit more detail in that he failed to give a warning of the possible other influential factors. Canning's similar kind of case, but a different problem, I think, arising. This is what I referred to earlier on as the question of whether the reliability element was fit for purpose. In other words, fit for the main prosecution elements in the trial. Here we have the alleged dog, as it was called, that two or more I explained infant death equals non-accidental death, something of that nature. I may have put it rather crudely, but now, again, that is something which could be quite interesting as a hypothesis tried out in a journal or something of that nature, perhaps even of relevance to a civil case, but probably, indeed on this, was certainly not enough, really, if you're going to be the star witness for the prosecution, because, in fact, the probability that that is true is very much lower than was being suggested. Again, the Harris case is a similar thing, a different issue, non-accidental head injury being inferred from a particular constellation or triad of intracranial injuries. Here, again, something that's very suggestive, the presence of these intracranial injuries in a certain pattern, very suggestive, but the problem is there was no way of knowing whether it was sufficiently reliable for a criminal case. In fact, it probably was sufficient perhaps. Other purposes, not for the main evidence in the prosecution's case. So what are the lessons so far? Well, a great deal of expertise makes assumptions about matters outside its field. I mentioned that right at the start, but evidence, do not conceal inconvenient data, better have it out in the open right at the beginning and explain it away rather than not having it there at all. The purpose is for which data may or may not prove that it's fine for scholarly speculation is a long way from what can pass master if you're the star with this in the prosecution case. Now, that is not to criticise I should heavily online the experts in canyms, Harris those cases from giving their evidence. Their evidence might have been all right actually as supporting evidence had there been some other evidence which was really the lead evidence for the prosecution might have been all right there with, but unfortunately it loomed rather larger than that which is where the problems began. What are we doing in this project? I've mentioned some of that already that our aim is to reform the basis on which expert evidence is of missile but the reason why I've divided the section into two halves and I give you now the first half is that in fact we at the Law Commission have not been particularly concerned with evidence experts who give evidence in criminal trials where really the reliability of the evidence is not enough for the purpose in hand. We have addressed those issues but actually the larger part of our consultation and report is about something rather different and you'll note from the title of my lecture pseudoscience and statistics but in fact a very large part of what we've been concerned is what is at the boundaries of science and pseudoscience if I can put it in that way. How do you know if someone comes along claiming that they can tell from a nose print who someone is for example are they an expert? Are they a scientist? I mean how do you know should they just be treated in the same way as an identification witness who claims they saw the person at the scene of the crime or should they actually be treated as an expert? Not an easy question to answer but very important because in the criminal law and then the law generally expert has very broad compass or remit or broad mean I should say more proper it includes for example police officers who are as they claim anyway much better at identifying the people from CCTD video because they spend hours and hours doing it so in other words their expertise comes from experience not from scholarly learning and so on and so forth some people may be treated as experts in the criminal court now I'm not saying that CCTD evidence and identification drawn there from its pseudoscience but I think some scientists in the audience might raise an eyebrow at the thought that it might be regarded as expert evidence or can't but in the law it can what we have said about these kinds of cases is that expert evidence should only be admissible if the trial judge before the trial begins in fact or at the preliminary stage should only allow the trial judge to be a gatekeeper and should only allow the evidence in if persuaded or satisfied that it's sufficiently reliable to be laid and to that end the judge should be assisted by guidelines and I won't go into this because it's a technical guideline that is needed by a court expert where it really is outside the scope even of the judge's ability to comprehend but the owners in our proposals will be on experts or those who claim to be experts producing their own background methodology literature insofar as it can be boiled down to a reasonable degree which supports their claim to be an expert or indeed and it's a bit like going back to trial by ordeal having experts actually tested alright the judge will say Mr Soto you claim you're a lit reader right well we'll have a little trial shall we and we'll all get as good as soon as we can say the same thing and can you tell all their say now I'm not suggesting that the judge will do it in that way because of course the statistical significance of just doing a handful of people might need to learn but it would certainly be possible in some instances and appropriate to see whether lit readers can actually do the business and the police officers who claim to identify people actually do it with a random sample of people put in front of them in that way that judges should get more involved in doing that sort of thing before this evidence is allowed to go forth now I don't want to bore you with the statements of the present law of the psalm but the present law relies very heavily perhaps not surprisingly as the bold element in there indicates on expertise being part of a body of knowledge or experience sufficiently organised or recognised to be accepted as a body, a reliable body of experience so the law at least in theory relies on consensus opinion if you like which is a sort of safe haven if you like for judges when deciding something is a matter of expertise now one of the problems about that is that not only do you need to ignore what I'm writing in some cases the fact that the subject matter of opinion must be part of a body of knowledge or experience but also in this case at the bottom the court of opinion could be rejected the view that in order to be admissible there's a preliminary requirement that the expert be tested as to their methodology and the foundations of what they're saying they said no no no there's no preliminary requirement to be tested in cross-examination and so to come back to the question of gait analysis or the analysis of how you walk it was a case it's part of the consultation that an old baby judge wrote to me and said we might be interested in this case that I had to try last week in which an expert or a ledge expert came along and said well I know it was the defendant because if you look at the CCTV of him right away you can see he runs in this particular sort of a way I don't know if he's dragging his left leg or whatever the evidence is and here we've got some CCTV of him running as he was requested to do from HV by the prosecution or whoever they've got the evidence I'm not sure we know it's him in the second one and if you look at the way that he runs it's quite obviously him quite obviously he's very dubious about this but the problem was there I was as a judge faced with luxury I couldn't say to him how do you what's the scientific basis of this how do you know yes to the common garden person it looks the same but have you got anything more than just an ordinary person I wasn't able to do that so I let it go for he said you know what happened he said the the jury acquitted I sent me a note and the note said what did you think you were doing with that ridiculous evidence we just had to discount it straight away we can't understand why it was admitted so there we are but he said what you're doing could help to stop that kind of absurdity occurring that would be great but again I don't rule out the possibility that kind of evidence could tip the balance of supporting evidence for something more major possibly it's not easy but it was wrong actually to try to use it as a main bit for the prosecution's case now Paris cases and one of the problems I mentioned earlier on when I talked about the text and the law's reliance on a body of experience is that actually of course the trouble with relying on a body of experience is that what the courts have got in mind here is evidence for the prosecution you want to be sure really sure that evidence is reliable if it's going to form part of the case for the prosecution but actually I've suggested to you I'm not sure that that is really or should be the test when you're talking about expert evidence used by the defence all they have to do is throw a bit of sand in the prosecution's face they don't have to prove anything so they seem to me should be given a little bit of latitude to use minority opinion and if the minority opinion is pretty absurd and indeed there's one case in which the defence called an 80 year old alleged expert from Australia to show or suggest that a baby with cranial injuries in fact died of scurvy which seemed a little bit unlikely but now that kind of case clearly is well beyond the sorts of expertise that one that is in mind must be drawing on a body of experience about cranial injuries I mean I'll come back to that case in a moment but there are I think different ways in which this canon should be approached depending on whether you are appearing for prosecution or defence here are some areas where this issue has cropped up and it has to be said it's increased I don't say every single criminal case involves at least some expert evidence that would be wrong but it's probably not far away to say to the majority in some way shape or form probably I mean there's no way to know there's no statistics I don't believe me I'm trying to find out there's no way to know but it's becoming more and more significant judges are telling us this voice identification based on auditory comparisons so sounds like is that a field of expertise do you know how reliable is it again you ought to be able to test that before it gets into court handwriting comparisons diagnoses of bad women's syndrome repression theory of sex abuse memory games comparison I've talked about that that was a case of ear prints a man was accused of very serious crime and the evidence against him was that he left his ear print on the window of the house where he was accused of crime took place and two I think ear print experts came on to say yes it's definitely here definitely and of course he was convicted and when you can see the conclusion cantering over the horizon can't you which was that when they finally got some DNA evidence it proved it wasn't him and the reason why the error occurred was that was not that these people were charlatans or anything like that no but they are once a comparison about the ear were figure but let's say they were 8 point comparisons whereas if in the case they called a finger print expert to say well what would you regard as reliable the finger print expert would have said what 8 points of comparison no way I mean that's not nearly good enough the error rate would be too high you need something like a 12 or a 16 point or something of that nature and that would have caused much more doubt I think on the confident assertions of a finger print expert in that case I mentioned DNA evidence now DNA evidence is often absolutely central to the prosecution's case and so there are particular issues about reliability of this evidence and I've drawn down here from the CPS guidelines for low copy number DNA evidence what they now say about this and as you can see there are actually quite a number of checks in place that the CPS now have to make sure that this evidence is not which is so powerful and persuasive when it gets in the core is not in the end unreliable and you can see here qualifications, procedures in the lab does the evidence involve composite results that's the messy data problem coming back the interpretation of the results that's another issue which Professor Han raised with us in the consultation paper where I didn't mention it that sometimes when confronted with data which are actually quite good the way that they're interpreted by the expert is wrong so that's another thing that has to be checked so I'm glad, I hope you're reassured to see that the Crown Prosecution Service is actually alive to the possibility of errors being made in the way that DNA is kept and then presented and it's quite clear that there is a need I think to try to get a grip on this problem pointed out by the House of Science and Technology committee who wanted a way of testing where the theories are sufficiently robust and so on and this actually relates to something in Rule 702 of the Federal Rules of Evidence a much discussed rule where in embryonic form not only very sophisticated form but in embryonic form there is something along the lines that we would like that there should be some way of testing the theory or technique has it appeared in peer review and publication that kind of thing been tested out in that way what is the known or potential error rate that's number three there but number four they sort of go back into their bolt hole really run like dangerous courses or wide spread acceptance is very important and minimal support should be viewed well as I said maybe that's true for prosecution but not all now what our provisioning proposed solution involve basically following that case I mentioned doubt that the American case I just looked at and we are proposing a game keeping role for the trial judge involving a discretion to exclude what might otherwise be quite relevant evidence on grounds that it's unreliable and there are going to be guidelines on reliability that means and as I said earlier on I'm not looking into this here in difficult cases the judge could have the power to call on independent assessor although we think those cases would be extremely rare in some very arcane fields of statistics or whatever it may be but that's not something I want to go into today now the way that this breaks down is that the in general the as I said the evidence must be admissible only of course deans is sufficiently reliable and as a question of law the evidence must be predicated on sound principles and assumptions techniques and so on in other words no astrology essentially no science of that nature but beyond that it's too stringent to test other than ruling out things like astrology I mean I think one could rely in general terms on the prosecution not to present on as an expert unless they meet at least this minimum question but it nonetheless is an important one to have out in the open then is the question how have the principles been properly applied to the case now you might say well why should the judge be concerned with that well there's a case called gilfoil in which an evidence of a psychologist was called to suggest that a person a victim of murder as they were at following a conviction of the defendant for the murder might actually have committed suicide instead but this psychologist is not examine the defendant or ask any questions about the case was doing it purely from a view of the documents and really the court appeal said they have absolutely no way of being able to apply the principles that they wanted to put in evidence about predisposition to commit suicide they had no way of applying that to the actual facts of the case in a proper manner and so that evidence was simply excluded as just not sufficiently up to the mark and the evidence is the evidence that the expert gives must be supported by those principles as applied to the facts of the case in other words it's no good being an expert on A if actually the principles in play or relevance are to do with subject matter big that will not do now in terms of responses we have over 80 individuals and organisations mainly supportive of the general idea although a lot of the devil remains in the detail and I thought I'd just take it through a selection of some of the observations that they mentioned in the course of consultation I mean it's obviously very selective but it's quite a substantial document the responses to consultation one solicitor said well shouldn't there be a warning to the jury that expert evidence can be wrong rather than there's a warning about identification evidence of that kind and in fact we think that's rather a good suggestion and that probably there ought if not in all cases at least where the judge thinks it's appropriate for such a warning perhaps particularly where the evidence is central to the case it might be sensible to have such a warning not what Lloyd's called a full corroboration warning but nonetheless which will allow for a bit of nostalgia on the part of those who studied this before rules have changed but nonetheless a warning that the expert evidence can be wrong in certain cases another campaigner said in child death and injury cases the expert should come from a separate health authority we'll have to go outside and admit but nonetheless an interesting suggestion now here are some more detailed things some of which we have actually added to our recommendations attention to possible variations in the meaning of the opinion or its wording where you've got a documentary evidence that's true you may need to cross-examine the expert on that the confidence with which the experts is expressing themselves now that is important actually we don't look that in the first when we consulted and we are going to include something on that one of the problems in the Dallaher case the ear print evidence was that the experts really were very confident about this they did not express the ifs and buts and the hesitations that one might have looked for perhaps and no doubt that's because they were they had the CPS prosecuted with a hand behind their back it was urging them to remain confident but that is a very important factor to bear in mind does the opinion address other explanations also very important I think that in other words as an expert and this is a well known phenomenon that is easily forgotten you should not consider yourself purely an expert for the prosecution or the defence and concentrate only on the evidence that supports it there there should be attention paid to other possible explanations that you know now this is something we couldn't deal with the proposal should be extended to civil proceedings for example proceedings in relation to taking children into care for example that may be based on expert evidence that a child has been abused physically or something of that nature should there be some testing of the experts and give evidence in that context well quite plausibly but a bit outside out of mid then some more then some other suggestions I won't take you through all of those not all of which we've actually adopted or will propose or recommend now another interesting one was that our proposal for lawyers anyway that our proposal should apply to expert assessments that an offender is dangerous at the sensitive stage crucial obviously for the offender because if they are so labelled they're going to end up with an indeterminate sentence or for example pretty substantial but again should probably be able to test the credentials of the expert in dangerousness before the evidence comes in at that stage pretty important a little bit of what we were talking about we were literally just talking about reliability but now in one there we should also I think include as part of the things that the judge should be chesting for competence, impartiality strength of opinion are very important one in my view and the necessity of the evidence for the jury to understand and decide really some other things that I think is less relevant perhaps for this audience but not we drew a distinction originally between scientific and non-scientific expertise but that proved not terribly workable now there will have to be testing I mentioned that but sometimes of course it's not always possible to test expertise before you get into court and in that case the judge will have to take a view just purely on the person's qualifications or experience or whatever it may be a rigid test that everyone reporting to me next must produce a great sheet of papers of proof of their methodology it's not designed to work in that bureaucratic way now here if I could focus on bullet point 3 I've stressed already the importance of the defence not having to go through the same hoops as the prosecution who of course have to prove the case to a much higher Stanley now what we're running with here is the idea that if all the defence has to do with the punch holes in the prosecution's case and it cross examinations say well I put it to you that in fact your data are not as reliable as you claim it to be and in fact it only shows us 70% probability or probability depending on your building is that not so your perfect entitled to do that without having to show the grounds of your own based saying so otherwise the thing would become to unwield however what I think we will be insisting on is that there are the judge applies these tests of reliability if the defence goes beyond just trying to rebut the prosecution's case and introduces a new scientific theory of its own to explain the course of events as in the scurvy example I mentioned earlier on there I think one would have to have a test of the expertise of the person called to give that evidence and I've mentioned also the judicial duty to consider a war that's also something we're incorporating so specific considerations for experts if I can just finish there in future experts are going to have to be prepared to come on and defend their methodology and not just their evidence and there will be a role I think here for professional bodies in producing guidelines as to what you should come to call which clearly it's over coming with a great pile of textbooks and articles and songs that would be unreasonable and not helpful anyway so there's going to be a role for expert bodies in working out exactly what it's going to be useful and sufficient for calls to text reliability there will of course be disclosure obligations as there currently are in any event and the other right for duty is to the court I know that's already a requirement but it's important I think that experts are as alive to the possibility of the case being explained on some other basis as the basis on which they've been asked or requested to to explain it I think that is important in the context of how expert evidence is used well I've slightly run over time but I will be very happy to take any questions you may have and it's been a pleasure to actually be to select the people who are I thought you would have a good do you want me to ask a question to the gentleman in the middle thank you very much for a very interesting talk I'm a pediatrician specialising in brain injury particularly type of heart practice in my accent with head injury but I'm a little disappointed to hear that we experts are going to be told and instructed how to balance the evidence because I think that's a good idea the danger there is that those of us who feel we do that already will carry on doing it and people who should be doing it haven't been doing it may listen to that and ignore the guidelines surely that's the barrister's job and where I've been disappointed in the court is when I said something that clearly should be challenged or at least they should ask for some how can I make that statement what's in the evidence of that statement there's been a chasm in the court and I almost felt that the barrister should have been instructed how to defend for what could be a factor of charge basically that's an extremely good point I think that the problem at the moment is that it's all left to be tested in cross-examination in accordance with the adversarial tradition but that as you yourself say it's not happening in a very satisfactory way but it will undoubtedly be the case that as a result if our proposals are adopted by government there will of course have to be training not just judges who will be adopting a very significant role but also for advocates because advocates will have to come along and be prepared to persuade the court along with their expert that the evidence should be admitted so they are going to have to be up on it in any case where there's going to be an issue but I'm afraid that you can't make a provision for absolutely everything and whilst it will certainly be true I think that barristers are going to have to become more adept at questioning methodology and challenging it in court because the judge will be required to do that right from the start in any event there are going to be these gaps in knowledge I mean almost inevitably I think I mean we're very happy to collaborate I think from professionals particularly in my area we're very keen to speak with the courts about this I'm not sure that the Royal College of Earprints has been formed yet there's not enough organisation ability to work closely with the courts about that field so ultimately that will have to go back to be I'm not sure if we should use anything on paper to those two guys who stood up in that case who really presented such strong feelings there whether the people who come from those fields will go back down again I think it's back down to the barristers Well it is but I think that the we're trying to take this in a sort of staged way I mean I think that there are already actually quite a number of rules bearing on criminal procedure which relate to the preparation of expert reports in Salomon how they are presented there's already quite a lot there but what there is lacking is a test that the judge applies or at least to moderate how the evidence is being presented because the judge might say well look of the earprint for people well I can see that this has some basics I mean it has a scientific basis but you're surely not going to put that forward as your only evidence are you because if so well the case isn't really going to get a lot further you're going to have to go back and come up with some other evidence or in some other way because it's just not reliable for that purpose and I think what we are anticipating there for is a more prominent role for the judge right at the very beginning to try and screen out the worst examples of this but there still will be an important role for the advocates and you're right I mean they may still fall down in their duties but they will not be immune from the drift of the changes should they come through there will be a sense in which they'll be expected to take a more knowledgeable line across examination and to have discussed with the experts what the other side are likely to say I mean I should have thought they'd normally do this in any event actually but it just may be in some instances it doesn't work that way Thank you How would you respond to criticisms such as those expressed by Chief Justice Rehnquist in Dover itself that invisibility rules such as this they place too much reliance on the judge to become amateur scientists and ultimately they don't have that capacity to do so and the subsequent research that has been carried out that this rule has just applied indistinctly and a lot of evidence gets through anyway Sure now that's a perfectly fair observation and a challenge to me in a broad sense we have to try and get a little bit beyond the general maxim that one of the benefits of the University of Education is that you know when someone is talking rot because that would you might not in these kinds of cases but I think that where I would take issue with Chief Justice Rehnquist would be the idea that judges are expected to be amateur scientists, I mean that's not what we're anticipating or amateur institutions or whatever, what they are meant to be to use a picking phrase one of the scholars uses is intelligent consumers of science if I put it in that way in other words they're perfectly entitled to say look Mr Sanzo this is all going over my head you've got to try and show me in language like reasonably understand how does this methodology work how have you tested it how do you satisfy me basically now so in other words the others would be on the judge proactive to to get the expert to show them how they can get an intelligible critique or assessment of the evidence now that may not be easy to do but that's what we're aiming for if you like and that's what in fact the expert bodies that we talked to said well actually that is possible won't be easy, I want to be able to insist on being paid for it but it is actually possible to do that there are only some areas we think that they said where actually the thing is so arcane that you probably couldn't it just wouldn't be reasonable to expect someone to put in ordinary put something in language that an ordinary judge could understand that just can't be done and that's why we had the expert assessor provision but that's quite controversial in a number of ways and it's quite a difficult proposition but I didn't want to focus on it today because it's not our main pot but there will be provision as I say for an expert assessor for those sorts for the rare sorts of cases where you really can't explain something in a way that the judge can adopt the critical angle and take decisions about it I'm sure Chief Constable is waiting for that even more but none of that lets you, I'm not here to account to him, fortunately I meet Puddwyd, I I walk into the business court and lawyer and I read the clerk the first judgment in the first opinion which the conviction was affirmed I read it after the event and the decision is probably I found it very alarming and the reason being actually it's not because it's a bed decision or it's a bed judgment actually because it's such a thorough and detailed analysis of books and evidence and the law involved and nevertheless they got it wrong eventually in terms of result and I'm still not sure what to make of this event but I think I would expect much more concern and more intent to slightly more radical reforms of the issue of expert evidence based on this wrongful conviction and two I think general issues which come across the first one is I want to focus on the use of statistics the first one is about the perception that it was the main evidence or the only evidence etc and I think once you finish reading the persecution case you realise it was very essential as the first court I think rightly pointed out and the problem is not only about this particular it's not local to that region and you realise and I think that it might be the right kind of how it plays out or what you could see there how basic notions such as reasons without and the presumption of innocence walk because once you read the case of the court and you put aside the statistics you nevertheless I think wondering how come with all these allegedly strong defences of the accused and so many doubts running in the court and perceived such a thorough examination how come the accused was still convicted of the end so that's one thing I think which should be examined and I think the second thing is the issue of expert evidence in particular and that's I would expect that type of cases to lead to much more skepticism amongst the legal profession towards the experts and to kind of consider I think more thoroughly the legal questions arising for that sake the statistics for example assuming that it was in an alternative case that if the statistics were gathered in the utmost professional manner and everything would be arrived according to the statisticians I think the questions for us remain as forceful as they were before so if the numbers were reduced essentially is that seen as good evidence to be used for the defendants purpose or if they were seen just slightly not 1% of the 3 million but 1% of the 23 million or whatever the number would have been is that seen as good evidence to use in general not only the supporting case not only as a key evidence but also as a supporting case as to where I think it's a solid record so overall I think there are many more expressions but I think overall I mean going back to the downward principles is good start maybe but I think the issue of experts and I think comparing to other legal systems the amount of trust and respect they received amongst Florida in this country might I think be questionable maybe that's how much reform is not appropriate yes well the juice to what you're getting at is that I think you have to be very careful in papers like these from quite a bit of bodies like the law division not to say things like well if our proposals have been in place of course this case would have been decided differently like Clarke I'm pretty confident we don't say that anyway and we give very few issues we did because there is no guarantee actually that even if our proposals have been in place that the particular errors that were fallen into there would not have happened it was a strange concatenation of services and things that went wrong so in a way cases for us are suggestive of problems I mean sometimes you can put forward a recommendation you can say what we need to do is get rid of the effect of that case but that's the women it sets down a rule of some kind but Clarke wasn't like that Clarke was just a bad decision and it went wrong for a number of interconnected reasons and it's only really one or two of those that we're trying to address I mean that's why in a way I started off by looking at Clarke not from the point of view of the statistical stuff but actually from the point of view of the experts themselves and what they should have been saying actually and what their GGs were to the court with respect to the inconvenient data if you like or the broader picture but which they may feel it's not their part it's not their role to pronounce Clarke's in fact as it is I mean in a way an expert should to some extent be a fly in the ointment for the barrister because the barrister should really get wanting them to say A but the expert should be saying well yes alright but I'm not going to say A unless you allow me to mention B because even though B might muddy the waters because unless I mention B you can't really set A in proper context and it's that kind of possibility I think that one has to be very careful about and to some extent that's already being addressed actually in the way that expert reports up or forward and dealt with by advocates in cases so there's already been quite a lot of response I think to these problems at high level but you can't guarantee that any of the cases actually we're looking at would be dealt with in a different way we just hope that over a period of time would be of a task My name is Robert I've been an expert on a few cases and sir I've also just recently been stuck in Hong Kong for 11 days and what surprised me was we saw all these experts on the TV talking about volcanoes and this and that and it's as though they had just been wheeled out they got their 15 minutes of fame and they were so enthusiastic but actually what they weren't good at was giving an opinion and interpreting they were good at just characterising the data or the information and from your talk I didn't see how you were going to address the complete mismatch between your field and our field as experts and that we're enthusiastic about what we do and we're just simply not trained in the way that you are in what we're trained in is the sort of stopping speculation and how do you deal with that complete mismatch and problem because whatever you want I've been caught a number of times and I just noticed it a whole time this is like two worlds clashing and how do you get sense out of that and what seems to come across to me is that it's the person with the most forceful opinion or way of presenting himself irrespective of whether what they're saying is true or not that looks the best and there's no way that you can get around that it just looks better so how do you deal with all of that? Well I'd like to reinforce that point I think because I think one of the problems with jury trials is that juries equate the expression of doubt or reservation with someone being a not particularly reliable witness and they equate confidence although not over confidence clearly but good presentation confident presentation as actually evidence that what you're saying is more reliable whereas of course whilst that might be true of course it equally might not be true I think it's quite we've tried to steer it away a little bit in our report from what you might call trial techniques and the problems about idiosyncrasies in particular the witnesses because one obvious answer to the point you've made is what you need to have is experts who are experienced at giving evidence in criminal cases and understand but there are a bit of that too because one of the problems is that they get captured by defence and they always come up and give the same basic stuff more and more confidently and professionally and expertly and that can have its own problems too actually can have a distorting effect on the way that when defendants decide whether or not to plead guilty as advised by their advocate it's not an unmixed blessing I think so I'm very alive to the problem it mentioned certainly and I think we're very clear that our paper is purely about the reliability in a provable sense of the basis of an expert opinion that dealing with problems in cross examination of people's confidence and so on to a certain extent you can address that if it emerges out of the reliability criteria I've mentioned that already the confidence must be matched to questions of reliability but the judge can't control what happens once you get into the witness body and start giving your evidence or at least only to a limited extent can you say well I'm not sure you should be expressing things as confidently as that or something because that's now trespassing on the job of counter to the other side there are two separate phases in this the problem about world's climbing is a problematic one but actually we found when I was doing a different product on murder we were looking at diminished responsibility that experts give there on mental disorder actually what was helpful was that some experts who were used to giving evidence of criminal cases and in some cases had actually given qualifications were prepared to sit round the table with their colleagues explaining exactly what the nature of the problem was that we had if we believed the product to you and then say well look how can we help the law commission to make some suggestions that make sense to us that our job a little bit easier in terms of what we're expected to say when it comes to court and that is a way to at least get some element of overlap between the worlds but the problem with that is that we were talking there about a very specific kind of evidence about mental disorder whereas in this project obviously we're dealing with everything from police officers looking at video screens through to pathologists or what the whole range and it's a bit harder to do that actually I want to talk about baby shaking from a lawyer's point of view I've done a number of the cases both as a barrister and as a judge indeed I've crossed the time in Dr Tasker I hope I'm not one of those who missed the most of points we've been talking about before I am a firm believer in juries and I do not have any difficulty in thinking that juries can weave through the overconfident witness I don't think we really need to worry about that because they're perfectly sensible people and in practically every case I have complete faith in juries but not in baby shaking cases and the reason for that is because the evidence is simply so complicated I've just done one in Cambridge with nine experts giving evidence all extremely distinguished and all having entirely contradictory views and I think that some way has to be devised of us putting more things by agreement to a jury yet having informal meeting and then giving the jury a document we're saying what the issues are and defining the issues we did that in the case of Cambridge that Cambridge is well ahead of the field obviously that's so clear but the reality is it was still too complicated for a jury to understand and I think that it's a heavy responsibility on the doctors how do you get so many doctors from the same fields the Royal College of Pathologies got together in December of last years trying to reach some agreed thing the total amount of their agreement was well we must look very carefully at cases which involve very young babies and that's really the whole extent of it after a day or two meeting also if you don't mind me saying this is not meant to be a have a go at doctors but I think it was having a good lawyer's back they do talk techno about I said to the jury in Cambridge when you say anterior do you mean towards the front and they said yes and when you say posterior do you mean the back yes but would you mind using front and back no they said we won't and they refused so somehow we have to get the witnesses to simplify as well otherwise we have to find some way of giving the jury a clear medical picture you cannot expect 12 women sitting there who haven't had the reports who have no knowledge at all to reach these very difficult decisions on it and I think that if I may say so where the lawfish action needs to be going rather than just having more and more experts coming along with developments which is where we're going there what can I say I mean you're right of course but I would have myself thought that actually with questions like anterior posterior and so on it's part of the job isn't it part of the job for the advocate to translate this stuff into terribly boring every time you say a long word you put it in English and they say do you mean that and they say yes couldn't you have said that the third time that'd be true that's a bit outside out of the myth it's not necessarily true I could be absolutely very serious but it's not it's not exactly what our report is about it's about scientific status of certain kinds of discipline and how reliable they really are rather than what actually may occur of trials and so on I mean you're right but in some respect and this is true of the law is there any other discipline that you do need technical terms in certain instances I mean you can do a certain amount by getting rid of Latin and all that sort of thing and talking about faults and certain things but at some point you are going to need technical language to describe some things and that's true across the board isn't it okay but that's not the main thrust the main thrust is we have to get somewhere of getting on the difficult areas more agreement and I think that has to be done somehow may I interpolate something for the benefit of those who may not know which is the criminal procedure rules in 2005 created under the aegis of Lord Wolff who was very concerned about expert evidence actually contains a provision enabling a judge of the pre-trial phase now where there are conflicting experts to order them to get together and produce a statement of what they agree on and what they don't which is perhaps a little step in that direction what force are I'm not a great single expert said you will have one expert who will do the report and will do it on that now you can't do that in crime the most you can do is if you have two defence experts for two defendants you can sometimes now plenty of other questions a gentleman there in the blue shirt I'm afraid I will also appeal attrition and I'm not going to defend the disagreements that occur but I'm interested as well in this idea of informally getting together and agreeing the extent of disagreements and then formulating a consensus isn't there a danger that the jury are not going to be presented with and able to hear the cross-examination of that expert evidence you may well have a bias introduced by dominant parties producing a report on behalf of perhaps a more powerful group and if there are examples around the world of legal systems where this has worked and where maybe there are drawbacks with that that's an extremely good question there have been examples yes where if expert evidence has to be given in certain fields it is given by an independent expert in other words you don't get an expert produced by one party and then of course the other one has to produce their expert and then you get a punch of duty show that follows but it sounds like a great solution the idea that you have this someone from Mount Olympus to give him a partial view but first of all of course it undermines the idea that there is very intensive cross-examination of that person because they're not meant to be hostile with using that term very loosely they're meant to be this independent person so what's the incentive to cross-examination in a hostile way you must have agreed that they're being appointed as a birthday so one gets into a slight problem there I think the other difficulty is the more familiar one is that in some and indeed this is by no means a small scale problem in some cases there will only be a relatively narrow range or number of experts who can perform that role and if in and so far as they become associated with particular views and theories and that would hardly be surprising the whole trial may end up revolving around which one it gets picked if you like and that may skew very much the course of the trial and that I think will be extremely undesirable certainly when we were discussing this at the preliminary stage there was a very hostile reception amongst lawyers to the idea that you would not be able to cross-examine an expert that is exactly the way you can do now and part of what makes that legitimate is that the other person's witness that you're not just leading evidence in chief cross-examining so I mean I think we really have trouble making that stick now what's different I think is the idea that General Williams said is the idea that actually the you yourself called on both sides are able to get together and say well look over this bit actually we're not disagreeing so you shouldn't start we don't want you probing for error over there because that's where we agree on that point and the advocates one would hope would buy into that that's the point of the book reform that Professor Spencer mentioned and so the areas where the battle will be joined are known beforehand but that still doesn't in the end completely get you over the problem for which you alluded and which was the first thing when we went to the Old Bailey judges very kindly agree if we want them to at the law commission that we will meet to discuss any given product that we've got with them all and they're extremely helpful and the very first person when I set out our visual proposals on this said okay we'll solve this then I'm doing a problem in which you've got three MFXs on one side three MFXs on the other side are you shaking the case what am I going to say to jury that they've got to because somehow they've got to be convinced that the one side that the defence experts is just the evidence is absolutely nothing and they're never going to come to that conclusion so how do I solve that to which of course I don't really have arms as I say what we're looking at it's not quite that problem that's not in any way to undermine the significance of the problem not people just that just contrasting with the criminal side on the civil side we have not we've almost rejected the single joint expert in any serious case in particular if they be shaking cases I think that it's almost a fundamental principle of natural justice that if the area has any controversy there should be at least a second but we do case manage in a much more open and strong way every expert before being consulted has to be approved by the court after the case management the process of knocking their heads to has to be open and to use that all the political word transparent to say when they meet everything they say to each other should be logged in so the process forgive me for saying so but there is a certain medical school version of bullying that's to say you are in the corner by his die and say you are aligned with us mate and that person it's pushed into changing their view so it needs to be apparent that somebody's changed their view and why but I really don't think that it's necessarily a just process to lean towards lenience an expert for the defence an expert that the court permits to give a defence should be just as fine just as well and there is in the criminal business of getting the dodgy experts out of the cupboard and then being given within the trial setting exactly the same sort of depth as the mainstream does I agree and we were very clear although it did cause some controversy we were very clear here that what we were saying would apply equally to the defence of the time of prosecution that is to give the reliability or the expert's sense of the expert but clearly as I was saying all their evidence has to do their evidence doesn't have to do as much in order to fulfil its purpose not by the prosecution and so there is that slight difference we're having a lively discussion but people are beginning to start to have to go and there are indeed drinks out in the reception area shall we have one more question and then wind up for drinks over which I hope Jeremy is prepared to continue to answer questions when he's actually got by the tie as described a minute ago