AS, R (on the application of) v Kent County Council

The application of the benefit of the doubt is nothing more than an acknowledgement that age assessment cannot be concluded with complete accuracy, absent definitive documentary evidence, and is in the case of unaccompanied asylum-seeking children who may also have been traumatised, unlikely to be supported by other evidence. On that basis, its proper application is that where, having considered the evidence, the decision maker concludes there is doubt as to whether an individual is over 18 or not, the decision-maker should conclude that the applicant is under 18.

The benefit of the doubt is not of use where a specific date or age has to be determined except insofar as it requires a sympathetic assessment of the evidence as indicated in R (CJ) v Cardiff City Council [2011] EWCA Civ 1590 .

Human teeth develop as an individual progresses through childhood and into adulthood; that much is clear. How, and to what extent, the stages of that development are indicative of age (and the extent to which it can been assessed by a dental examination) is a matter of significant debate as was noted in R (on the application of ZM and SK) v The London Borough of Croydon (Dental age assessment ) [2016] UKUT 00559 (IAC).

In addition to the issues considered by ZM & SK the Mandibular Maturity Markers (MMMs), Root Pulp Visibility (RPV) and Periodontal Ligament Visibility (PLV) are unreliable.

While age assessment determinations are findings of fact, this application raises specific issues of wider importance:

the reliability of dental assessments of the type considered in R (on the application of ZM and SK) v The London Borough of Croydon (Dental age assessment) [2016] UKUT 5 59 (IAC);

the use of photographs in age assessment determinations;

the proper scope of the benefit of the doubt in age assessment.

The applicant’s case is that he is from Sherkheil village, Tagab district, Kapisa province, Afghanistan. He has two younger brothers and his father was a commander in the police. Approximately a month before he left Afghanistan his father was kidnapped by the Taliban and the family were later told that he had been killed. The Taliban came looking for him and a decision was taken to leave the country for his own safety. His cousin, the son of his maternal uncle, arranged for him to be taken out of Afghanistan by an agent. He was brought to Europe via Turkey, Hungary (where he was fingerprinted) and France. He arrived in the United Kingdom on 7 September 2015 and was taken into the care of the respondent county council. He claimed asylum but since the commencement of proceedings, that application was refused and it is understood that an appeal against the decision has been lodged.

There has been a substantial number of orders and interlocutory applications made in this case. These relate primarily to the request for disclosure of material and in respect of the instruction of expert witnesses. Of particular note in addition to the above is the application made by the respondent firstly to have disclosure of the applicant’s dental records which was granted and for the terms of Professor Roberts’ instructions to be changed such that he be solely the expert instructed by the respondent (hereafter “Kent”). That application was refused by Upper Tribunal Judge Rintoul on 3 May 2017.

Although Kent sought initially to challenge that decision, at a hearing on 10 May 2017, the matter was resolved by way of consent such that Professor Roberts was to be instructed by the respondent alone and provision was made for further questions to be put to Professor Cole. It should be noted that the hearing in this matter had initially been scheduled to commence on 10 May 2017 but owing to difficulties with the interpreter who had been booked, the hearing had been abandoned.

The parties’ respective positions have varied significantly during the course of the proceedings. Initially, as can be seen from the statement of facts and grounds the challenge was essentially that the initial age assessment was flawed and unreliable in that:-

the assessors did not give the applicant an opportunity to deal with the points adverse to his case at a time that the decision was only provisional and/or at all;

the written age assessment failed to include all and/or the real reasons why the applicant’s age was not accepted;

that the assessors erred in their consideration of the applicant in concluding that he was not credible;

that the manner of questioning was unfair; and

there are a number of obvious inaccuracies in the age assessment.

These were then detailed in grounds of quite extraordinary length.

The respondent’s initial defence was that the claim was without merit, there being no reasonable basis for asserting that the multiple professional social/support workers were all wrong in their assessment at various points in time that the applicant is the age assessed. The response to that defence took issue with those assertions.

Kent’s position, however, changed significantly as is set out in its statement of case dated 3 April 2017 which postdates the assessment by Ms Corby and Ms Carter and also by Professor Roberts. This assessment, that the applicant is 24½ years of age, is based on the assessment of Rose Corby putting him in the age range of 22 to 26 years with a most likely age of 24 and the evidence of Patrick Shier putting him in an age range consistent with Professor Roberts’ assessment. Kent avers that on this calculation the applicant was at least 18 when he first came into contact with Kent on 7 September 2015.

Kent’s contentions as set out in the statement of case are [5]:-

the judicial review grounds were irrelevant to the fact-finding exercise;

the benefit of the doubt in favour of the asylum seeker is irrelevant;

Professor Roberts’ evidence is reliable ; and

the applicant was born before 7 September 1997.

In R (A) v London Borough of Croydon [2009] UKSC 8 , the Supreme Court held that the question of whether an individual was a child or not was, upon challenge by judicial review, one of fact for the decision of the Court itself. Baroness Hale said at paragraph [27]:

Lord Hope, in his concurring judgment, said at paragraph [51]:

In R (CJ) v Cardiff City Council [2011] EWCA Civ 1590 at paragraph [21], Pitchford LJ stated:

In R (AM) v Solihull Metropolitan Borough Council (AAJR) [2012] UKUT 00118 (IAC), the Vice President of the Upper Tribunal stated at paragraph [15]:

In R (B) v London Borough of Merton [2003] EWHC 1689 (Admin) , Stanley Burnton J held at paragraph [36]:

In R (AS) v Croydon LBC [2011] EWHC 2091 (Admin) , HHJ Thornton QC (sitting as a Deputy High Court Judge) stated at paragraph [17]:

The applicant contends that he is entitled to the benefit of the doubt. The relevant case law is set out in the reply to the grounds at [10] – [12] supported by quotations which are selective. What was said in A v LB Croydon; WK v Kent CC [2009] EWHC 939 (Admin) per Collins J at paragraph [9] is:

The applicant also seeks to rely on R (CJ) v Cardiff CC [2011] EWCA Civ 1590 at paragraphs [21] and [23] where Pitchford LJ said:

This does not, contrary to the applicant’s submission, mean that if a decision-maker concludes a child is between 15 and 17, application of the benefit of the doubt would lead to a decision that he or she is 15.

We consider, on reviewing the case law, that application of the benefit of the doubt is nothing more than an acknowledgement that age assessment cannot be concluded with 100% accuracy, absent definitive documentary evidence, and is in the case of unaccompanied asylum-seeking children who may also have been traumatised, unlikely to be supported by other evidence. On that basis, its proper application is that where, having considered the evidence, the decision maker concludes there is doubt as to whether an individual is over 18 or not, then in those circumstances, the decision-maker should conclude that the applicant is under 18.

Thus, the benefit of the doubt is not of use where a specific date or age has to be determined except insofar as it requires a sympathetic assessment of the evidence as indicated in CJ v Cardiff .

Human teeth develop as an individual progresses through childhood and into adulthood; that much is clear. How, and to what extent, the stages of that development are indicative of age (and the extent to which it can been assessed by a dental examination) is a matter of significant debate as was noted in ZM & SK . We take that decision as our starting point in evaluating the merits of dental examinations in age-assessment.

As did the Upper Tribunal in ZM & SK , we consider it necessary to set out some of the background to how teeth develop in humans. What follows is a summary of what is said in ZM & SK at paragraphs [14] – [27]:

Human teeth emerge individually and develop gradually. This can be observed by x-ray, the standard means being a dental panoramic tomograph.

Human teeth pass through several stages of development. The system used by Professor Roberts and many others is derived from a paper by Demirjian and others, which identified 8 stages lettered A-H;

Individual teeth attain different stages at different speeds; The third molars are the last to emerge and normally the last to achieve full maturity at stage H. The permanent lower left third molar is LL8 (lower left, 8th tooth from the front) according to the British Dental Journal nomenclature.

A tooth which has not reached stage H is somewhere on its road to doing so, but a tooth which has reached stage H is fully mature; and if all the teeth in an individual have reached stage H (“mature dentition”), no further development in the 8-stage process can take place.

Until mature dentition is reached, it is possible to provide a description indicating which (if any) of the teeth have reached stage H and, in addition, listing those teeth which have reached other and if so what stages. Because a stage earlier than H is both achieved and lost, it is possible, by comparison with the database, to generate information about the ages between which the stage in question typically occurs in the tooth in question. The more immature teeth there are in the mouth, the more information can be generated in this way and, it is said, the more accurate is the prediction of the age of an individual of unknown age.

As the number of immature teeth decreases and more attain stage H, the results of this test lose accuracy.

This method is not applicable when all the teeth have reached stage H: the dentition is mature and will remain mature whether the individual is 17, 37 or 87. Further, because of the way in which the data is assembled, that is to say, typically, by the examination of individuals at unplanned stages of their dental development, there is no way of knowing how long an individual showing mature dentition has been in that state. The tooth that most recently achieved stage H may have done so six months ago or two years ago or (in the case of an older individual) twenty years ago.

What can be done is to record the ages of those whose tomograph shows completely mature dentition. This provides a record of the lowest possible age at which dental maturity was reached. In a large database, however, these data can usefully be compared with the ages of those who were at the last stage before completed dental maturity, that is to say, with one tooth only at stage G. In the result, data for the attainment of stage H in all teeth, or in the lower left third molar, which is regarded as a diagnostic sample, have been collected for a number of populations.

There is difficulty in determining whether those data can properly provide probabilistic estimates of the chronological age of a person of unknown age who is dentally mature, even more so in assessing the probability of the individual in question being more or less than 18 years old not merely because of the inherent dangers of calculating the facts from probabilities, but because the data collected do not have a Normal Distribution.

One method of deriving an estimate of probability from data of this sort is by a process of censoring, which attempts to confine the data used in the analysis to data which might properly contribute to it. The symmetry or assessable skewness of a Normal Distribution demands that appropriate censorship criteria are used at both the upper and the lower ends of the age range. There is considerable dissent between experts, apparent in the present applications, as to how censoring should properly be done and whether Professor Roberts' method is effective or accurate.

While it is assumed that no development takes place beyond stage H consideration has been given to determining whether teeth which have reached stage H show any further signs of maturation. The result has been the identification of three separate " mandibular maturity markers ". The examinations have been confined to cases where the lower left third molar has attained stage H. It is not clear whether the markers described are dependent upon the achievement of stage H or whether any of them may be observed in relation to earlier Demirjian stages.

The three proposed mandibular maturity markers are as follows: -

Root Canal Width (“RCW”). This marker proceeds upon the assumption that following the attainment of stage H, the width of the root canals, as perceived on the tomograph, narrows in each tooth. Three stages 1- 3 of A to C are identified.

Root Pulp Visibility (“RPV”). This is based on an analysis of the appearance of the root pulp in the lower left third molar and is divided into 4 stages, referred to by Professor Roberts as A to D. It is on his evidence not clear whether this phenomenon is a result of dental change, or of maturation and perhaps greater opacity of the mandible.

Periodontal Ligament Visibility (“PLV”). The x-ray visibility of the periodontal ligament or membrane reduces progressively as the tooth (again, in particular, the lower left third molar) settles into position. Radiographically, a visible ligament appears as a space between the tooth and the mandible. Professor Roberts identifies the stages as A to D but in the pioneering paper on this topic, again by Olze and others (Footnote: ³) the stages are 0, 1, 2, 3 and the definition is in descriptive rather than numerical terms

Having evaluated the evidence before him, the Vice-President concluded:

The fact that all teeth are mature in the sense that all have reached Demirjian stage H is a sign of chronological maturity but is not a reliable indicator of whether an individual is more or less than 18 years old. The use of the Demirjian stages below stage H does appear to be more reliable in the prediction of age, particularly in the lower teens.

None of the three mandibular maturity markers so far identified appears yet to have attained such acceptance in the scientific community that it can be accepted as a reliable pointer to chronological age in the late teens in males.

Two main issues arise from the above analysis:

Interpreting the data in respect of achieving Stage H: is Professor Roberts’ methodology effective or accurate? And, is “censoring” of data appropriate?

Whether the three proposed Mandibular Maturity Markers (“MMMs”) are a valid basis for assessing age once Stage H has been achieved?

An issue common to the above is the extent to which ethnic origin may result in different results: for example, there is evidence that teeth mature at different rates in different populations; in Caucasian populations maturity appears to be earlier than in Han Chinese populations.

It was primarily for these reasons that the applicant sought permission to adduce the expert evidence of a medical statistician, Professor Cole whose evidence is also relevant to the issue of the statistical validity of Professor Roberts’ work.

In addition to producing reports, the experts have, through questions put to them, refined their opinions. Although the issue of censoring of data was raised in ZM & SK that was no longer pursued. Rather, Professor Cole advised the use of a technique known as Probit analysis.

Probit analysis is a technique whereby the curved or s-shaped curve on the graphs showing the data of dental maturation is transformed into a straight line. This straight line can then be extended by regression to predict values for, in this case ages, which the data does not exist owing to the lower and higher age range cut-offs. Thus, where a study was undertaken with a range of individuals from, say, 16 to 19, a graph can be produced with a line indicating predicted values for those outside that age range.

Professor Cole’s view is that the appropriate statistic to use when considering stage H is the median age of attainment, that is the age when half of the population has achieved that stage, which is established by Probit analysis.

Professor Cole also advised that it would be useful when comparing the data, to undertake a “kappa analysis”. This is a technique by which a mathematical formula is applied to the data from different researchers which produces a value between 0 and 1. The purpose of the analysis is to identify the reliability of different data collectors when observing various phenomena such as the stages reached in the MMMs. It seeks to take coincidence out of the calculations.

Accordingly, and in the light of the observations made in ZM & SK , the focus of Professor Roberts’ evidence in this appeal has been to address the criticisms made in that decision; the evidence of Professor Cole is more directed to the issue of whether Professor Roberts’ conclusions are statistically sound and/or reliable.

Professor Roberts’ evidence is set out in his witness statement of 7 February 2017, 16 May 2017 and 4 June 2017. He also adopted the assessment made on 23 February 2017 and the skeleton produced by Mr Holbrook dated 31 May 2017. He also continued to rely on his own paper, Lucas et al , "Dental Age Estimation: A Test of Reliability of Correctly Identifying a Subject over 18 Years of Age Using the Gold Standard of Chronological Age as a Comparator" Journal of Forensic Sciences 2016 (Footnote: ⁴) (”the Gold Standard Report”) as well as an extract from the Encyclopaedia of Forensic and Legal Medicine entitled “Dental Age Estimation: Theory and Practice” (Footnote: ⁵) (“the Encyclopaedia Chapter”) ^.

Professor Roberts accepted that the respondent is no longer relying on part 2 of his second statement entitled “censoring” but did stand by what he said at page D89 in respect of age of attainment and current age although he accepted Kent were not relying on this. He denied resiling from what he had said and that he no longer used attainment of stage H except as a trigger for consideration of MMM. He said that, when asked about the assessment of the applicant, the observations on the stage of development were redundant and did not apply, adding it was a small part of the evidence relied upon.

Professor Roberts accepted that the Gold Standard Report had in its abstract stated the assignment to above or below the 18-year threshold, in the age range 17 years to 19 years, could be wrong on up to 50% of occasion but considered it inappropriate to refer to that in the Encyclopaedia chapter. He accepted that the Encyclopaedia had not been updated but did accept that techniques and processes had evolved since it had been written. He accepted also that when he had said in the report on the applicant that “all the information required for a full understanding of the anatomical statistical and theoretical aspects of dental age estimation are covered in the Encyclopaedia Chapter,” he had overstated the case.

Professor Roberts’ report records that the applicant’s lower left third molar is at stage H, meaning that it had completely formed. On that basis it is recorded that the probability that he is under 18 years is 20.7% the corollary being that the probability that he is over 18 years is 79.3 % using the UK-Caucasian dataset.

It is noted that following ZM and SK the use of LL8H (Footnote: ⁶) alone is not enough to justify an assessment and noted there is strong evidence of ethnic differences such as the Chinese dataset indicating a 99.09% chance of a subject with LL8Hm being over 18 but it is acknowledged that there is no reference data for Afghan nationals. He accepted that caution would be needed before extrapolating the above data directly to Afghan nationals.

Professor Roberts accepted that he should in his report on AS have said “the bulk” of the evidence not “all” but said that this was simply a semantic difference. He said that it would have been inappropriate in the Encyclopaedia chapter to refer to doubts and that if writing it now he would adjust it adding that what he had said as at February 2017 was correct and that he and Dr Lucas were willing to change their minds; this is an area which was moving very fast.

Professor Roberts said that he did not recall when he became aware of the range within which there is uncertainty referred to in the Gold Standard Report adding that it takes time for ideas post-publication to become accepted.

Professor Roberts considered that the Vice-President’s comments in ZM & SK at [64] were overzealous and that even when a report has been written, the result is nonetheless tentative and needs to be accepted by colleagues.

In ZM & SK at paragraph [65], the Vice-President wrote:

Professor Roberts said that he no longer reports stage H as a stage and, if this is present, goes on to consider the MMM. He said that he had not told the Vice-President he did not use stage H in the assessment but had done so only as a threshold to consider the MMM. Professor Roberts also accepted that there were differences between African populations and those in the UK and between the UK population and the Han Chinese. He said that the variation was minimal, the data might have been collected differently. He agreed that there would be potential for uncertainty in this area.

Much of the cross-examination was in fact directed towards showing that Professor Roberts was not a reliable witness.

Professor Roberts accepted that they had no idea of whether ethnicity affected RPV or PLV. He said that their methodology was objective but that the interpretation was difficult. He said they do not provide an actual assessment of age but when asked if they believed that their results were better than social workers, said that was a matter for the Tribunal to decide.

It was put to Professor Roberts that he had been four years out in his assessment of the age of SK (whose decision JR/3413/2016 was put to him – a decision of UTJ Blum). He added that SK had not developed 3 ^rd molars which made assessment difficult. Professor Roberts accepted that his witness statement [B15] had been intended to address the concerns raised in ZM & SK at paragraph [70].

In response to what had been said at paragraphs 60, 72 and 73, he said that they had validated their methodology.

Asked about the report which appears at B15/B20, he said that it should not be there. It had been adduced in error.

Asked about the issue of root canal width, Professor Roberts said that they would have used it but they were not able to in this case. Mr Holbrook said that Kent did not rely on this. Professor Roberts, however, thought that it was reliable (see witness statement at 53).

Asked to respond to Professor Cole’s comments in his report of 31 March 2017 that there had been no work to date on MMM in any other ethnic group apart from Caucasians and that thus applying them to non-Caucasians involves making a series of untestable assumptions, he said that it was reasonable to apply the assumptions to other ethnic groups in the absence of any other evidence.

Asked about a pattern of progression between stages A to D, he said he did not know if there is progression and that they had to assume that this was so. He accepted there is ethnic variation in development. He said it was not a wild guess to assume extrapolations could be made from Caucasians and this had been the subject of validation. He said that the validation study was multi-ethnic.

Commenting on the data of Root Pulp Visibility, he said that the sample data for Olze is different from that of Mongiovi. He noted that dental development across Europe appears to be different and that this may account for the difference between the Olze and Mongiovi reports as the age threshold appeared to be different.

Commenting on the kappa values, he said that these appeared to be very high here, because there have been only four categories. The purpose of kappa values is to attempt to take account of the chance of results being identical.

On re-examination Professor Roberts says that there was a progression from stage H to stages A through to D of RPV and also PLV. He said other people were working on this but it is still a work in progress.

Professor Cole said that it was dangerous to assess age from developmental markers, given the wide range of variation in those. He said it was fundamentally unscientific to seek to determine age from observation of physical characteristics as a result.

When asked about a Probit analysis, he explained that this was a measurement which allowed a straight line method for predicting results from a graph plotting a range of values which was unbounded. He had a number of criticisms of Professor Roberts’ work. He was in particular concerned about the apparent validation exercise [D81/D97] pointing out that as the age range of the group was recorded as 16 to 26 there could be nobody aged 29 (as appeared to be the case) and there were six other inconsistencies identifiable.

He considered that the kappa values identified in Professor Roberts’ work were extremely high and he questioned them, considering they appeared to be less in the other papers.

When cross-examined by Mr Holbrook, Professor Cole said that dental age assessment was fundamentally unscientific and that it was fundamentally flawed although he later rode back from this assertion with respect to stage H. He accepted that the Demirjian H stages are clearly stages of development and that the MMM might be developmental markers but although one would expect to see evidence of progress between the stages, this is not so. He said that there was no longitudinal study. He said again there was no evidence of progress between stages.

Professor Cole was also asked about his papers written on the issue, including published letters in which he had responded to Professor Roberts. He said he had not chosen the title of the article entitled “hot potato”. He also said that he had not in his 2008 paper referred to the absence of probit analysis as that had not been referred to, but he confirmed that it ought to be used, but it is unhelpful where a person’s age is unknown as you can quote the probability that they were under 18 when stage H was attained, but they will certainly be older and to an unknown extent.

While Professor Cole accepted that any error in the figures given for the applicant under Professor Roberts’s analysis was in his favour, that was not relevant, given that on either basis only a probability of when stage H had been achieved could be identified.

Asked about his research paper, ’The evidential value of developmental imaging for assessing age of majority’ (Footnote: ⁷) , he stated that it would not be correct to extrapolate from the sample, to an individual. While accepting that the table showed that in the case of Japanese, Chinese and Koreans, achievement of Demirjan Stage H was 100% indicative of having reached age 18, it could not be inferred that the same applied to Afghan nationals.

Re-examined, he said he did not have a political axe to grind, and was not a campaigner.

The respondent submitted that we should have regard to Kennedy v Cordia [2016] UKSC 6, and that this raised concerns about Professor Cole’s ability to give opinion evidence.

At paragraph [44] of Kennedy v Cordia the Supreme Court set out the considerations to be adopted in addressing opinion evidence.

We are satisfied, having considered their testimony, that both Professors Roberts and Cole’s evidence have assisted us. We are also satisfied that both have the necessary knowledge and experience in their relevant areas of expertise, forensic dentistry, and medical statistics. For the reasons set out below, we do not consider either lacks the necessary degree of impartiality.

We are satisfied that there is, in the light of Professor Cole’s CV, and evidence, supported by the publications cited, a reliable body of knowledge in medical statistics. It is of note that he has over a period of many years been involved in the compilation of data relating to development and growth in humans.

Further, and in any event, we observe first that the rules of evidence applicable in the Upper Tribunal do not prohibit the giving of opinion evidence (other than by an expert), albeit that little weight could be given to opinion evidence not coming from an expert.

At its core, the difference between the two experts is as to the use of data in assessing probability. Professor Cole’s primary contention is that given the wide variability of developmental factors in human populations, it is unsafe to apply them

This is best illustrated by the following examples.

Suppose we take a box containing 100 balls, 75 of which are red, 25 of which are blue. One is taken out. The rules of probability demonstrate that there is a 75% chance of it being red and 25% of it being blue but without other evidence such as seeing it, we do not know which it is. The evidence of the number of balls and the application of the rules of probability only permits a finding as to the chance of the ball being red or blue; it does not permit a finding that it is red or blue. That can only be determined by other evidence such as looking at it.

If, however, the box contained 100 red balls, then the probability of a ball withdrawn from it being red is 100%, but a probability assessment is irrelevant in such a case as, by a matter of logic, the ball can only be red; that is established by the evidence of the box containing only red balls. Other possibilities are excluded.

Suppose then a set of 10,000 balls, all of which are black apart from a sole red ball. It is easy to say that the chance of that ball being drawn is very unlikely. More importantly, even if a sample of 1000 at a time were taken, the chances of the red ball being in that sample is only 10%.

These are, however, closed sets. That is, there is a fixed number of balls, and the numbers of each colour are known. With any such set, even if more complexity such as the number of colours or the number of each is changed, the chance of drawing a ball of a particular colour can be calculated with certainty.

Suppose, however, that we do not have a closed set. We have a pool of 1000 balls, and we know that there are only two colours of balls in that pool. Samples can be taken and statistical methods applied to provide more or less reliable probabilities of the colour of a ball drawn, or as to the numbers of each colour.

Professor Cole does not suggest that one cannot extrapolate from a closed set to a larger population, but questions the statistical and scientific validity of doing so when there are both wide variations in human development, such as the age at which Stage H develops and, because although it is known that those values vary among ethnic populations, the extent of that variation is not known and thus cannot be allowed for in any calculations.

Professor Cole’s principal concern is, as we understand it, that if the age of a young person is known, one can from data compiled on, say, height, give a range of height within which that person is likely be, but what is not permissible, given the size of that range, and the interplay of other factors, such as ethnicity, is to seek to determine accurately the age of a person from his or her height or any developmental factor.

The datasets considered by the authors of the various papers shown to us were closed sets. The number in each was known, as were their ages, as was whether, for example, Stage H had been reached. On that basis, as Professor Cole said in his evidence, when asked about the possibility of an individual Hispanic being a particular age, he could give a probability, but only for a person from that dataset.

The applicant submits that Professor Roberts’ methodology is flawed for the reasons set out in ZM & SK and that these errors identified cannot be rectified on the basis of the evidence produced in this case. Mr Wise QC submitted that Professor Cole’s evidence is compelling and undermines Professor Roberts’ methodology and, it is submitted, that Professor Roberts’ methodology is unreliable as is his credibility.

Professor Roberts’ report relies on three factors.

that the lower left third molar had Demirjian stage H, indicating the probability that he was over 18 years is 79.03 percent;

that his RPV was observed as being at stage B, giving the likely age of 22.77 years; and

that his PLV was at stage D giving him a likely age of 23. 61 years.

The applicant makes the point that Professor Roberts says that he no longer relies on stage H as an indicator of likely age.

The respondent does, however, seek to rely on this as evidence on the basis that this is not challenged by Professor Cole and that even on Professor Cole’s basis, disregarding the issue with ethnicity, the effect of Probit analysis applied to this case showed that the median age of attainment of stage H falls from 20.2 years to about 19.5 years and the probability of being an 18 year old fell from about 20% to 10%. Thus, the likelihood of him being stage H whilst he is at the claimed age of 16.5 was slim. It is also submitted that in any event Professor Roberts’ treatment of stage H is insignificant in comparison to the conclusions he draws from the MMMs which are strongly indicative of him being over 18.

Mr Holbrook submitted that on any view of the Stage H data and in particular the Probit analysis, it was not likely that the applicant is the age claimed, there being little chance that he is under 18. He said that Kent did still rely on that part of Professor Roberts’ report from which he had distanced himself, it being for Kent to seek to rely on evidence as it was the party, not the expert.

Mr Wise submitted that the ethnicity of the subject was likely to introduce an unknown error. It was submitted there was no evidence that the evidence for Caucasians could properly be extrapolated to those of Afghan origin.

We consider that there is nothing in the materials to show that we should depart from the conclusion reached in ZM and SK that the stage H data is unreliable. While we noted that once the Probit analysis was undertaken that gave a likely age for the applicant as slightly older than on the basis of the analysis initially undertaken, that, however, does not advance the respondent’s case; the reality remains that the Stage H analysis can only say that he has achieved dental maturity, not when. It can provide a probability that he is older than claimed, but that is not sufficient evidence. Further, the evidence shows that there is a variance in the achievement of dental maturity across different ethnic populations.

We are satisfied that, in addition to the issues considered by ZM & SK , RPV and PLV are unreliable for the reasons set out below.

Kent’s case is that MMMs can be determinative or least highly suggestive of a subject being an adult. It is of note that in the assessment by DARLinG (Footnote: ⁸) on 23 February, 2017 that the applicant had achieved an RPV stage B and a PLV stage D. In response to questions put by the applicant on 18 March 2017, Professor Roberts set out how those stages would have been interpreted according to the different published papers by Olze (cited above)(Germans) and Sequeira, Perez-Mongiovi , Teixeira A, Calda IM and others (Footnote: ⁹) (“Caldas”) (Portuguese) as well as by Lucas (Footnote: ¹⁰) on whom DARLing/Professor Roberts rely.

Olze 26.20 (as indicated by PLV)

Caldas 19.10 (as indicated by PLV)

Lucas 18.67 (as indicated by PLV)

Olze 28.10 (RPV), 33.70 (PLV)

Caldas 21.90(RPV), 26.90 (PLV)

DARLinG/Lucas 22.61 (RPV), 23.37 (PLV)

It is noticeable that there is a very wide variance by using the different published data, but equally we note that Professor Roberts was of the view that owing to the different study designs, age ranges and small numbers, it was unwise to over analyse the details.

The maximum ages for males identified at Stages 0-3/A-D in males by Lucas, Mongiovi, and Olze, were, as is submitted by the applicant, similar and were located close to the cut-off point chosen by each team of researchers:

Lucas maximum ages Stages A-D (cut-off point 25.99 years)

RPV - 25.49, 25.93 25.96, and 25.19 (Footnote: ¹¹) .

PLV - 22.8, 25.43, 25.43, and 25.93 (Footnote: ¹²)

Sequeira maximum ages Stages 0-3 (cut-off point 31 years for RPV and 30.99 years for PLV)

RPV - 30.0, 28.7, 30.6, and 30.7 (Footnote: ¹³)

PLV - 20.4, 30.6, 30.6, and 30.7 (Footnote: ¹⁴)

Olze maximum ages Stages 0-3 (cut-off point 40 years)

RPV - 39.1, 40.0, 40.2, and 40.6 (Footnote: ¹⁵) .

PLV - 24.0, 26.4, 40.6, and 40.5 (Footnote: ¹⁶) [MB/9/3].

Aside from any difficulties in the analysis of the data, there is a fundamental problem. As Mr Wise submitted, whilst it is well known that an individual progresses gradually through the Demirjian development stages there is insufficient evidence that the same occurs with the RPV or PLV stages. In particular, as Professor Cole observed there are no longitudinal studies of RPV or PLV. That is, there is no evidence base taken from examining the same individual at different ages to ascertain whether the stages progress. It is, we note, remarkable that the achievement of the various stages of RPV appear very much higher in terms of the median in the Olze study which included individuals up to the age of 40 when compared to the figures generated by Lucas et al who used a different age range. While there is in that paper at section 4 a consideration of issues of methodology, there is also an indication of differences in the approach to categorising the different stages. It is, consequently, not possible to attach statistical significance to the data from DARLinG, Olze or Caldes.

We do not consider that this is an issue in which Professor Cole is stepping outside the area of his expertise. Rather, in noting the absence of the longitudinal study, he is identifying why the data presented is, from a medical/statistical point of view, unreliable. As he explains in his response to Kent’s questions about his criticism of Professor Roberts’ assertion that his criticism of RPV studies were matters of clinical judgment, only radiological appearance was such a matter; ethnicity, size of sample age range and age are relevant to data interpretation. We consider also that it was open to him to note that DARLinG assessors had, in Professor Robert’s statement, had difficulty in matching Olze’s descriptions with RPV and PLV stages indicating a possibility that different stages of development were being assessed. The same observations apply to PLV.

We consider that Kappa values are relevant in the consideration of the evidence put forward by Caldes, Olze and DARLinG. It is a means to identifying whether there could be errors in the data collection. It is also of note that DARLinG and implicitly Kent accepted that the absence of Kappa values from the Olze RPV paper is a weakness as it is with the PLV paper.

There is also the difficulty in that, although Professor Roberts uses stage H as a trigger for considering MMMs, it appears that an individual may display an MMM without being in stage H as appears from the validation study attached to his second statement. Professor Roberts referred to this as an error, stating

While it would, clearly, be dangerous to rely on a single incidence, equally it is unclear why Professor Roberts is sure this is a recording or typographical error.

Whilst it is clear that Professor Roberts did seek to rely on the data attached to the skeleton argument, he did accept that there may be a question as to whether there is progression through MMMs but that the same criticism could not be made of whether the minimum age for the occurrence of the initial stage arose.

Mr Wise relied also on Professor Roberts’ first witness statement as to why MMMs are unreliable. It was, as we observed, unclear if there was evidence other than that obtained from a Caucasian population. There appears to be no indication from the data of Olze or Caldas of consideration of anything other than a Caucasian population.

We consider that none of Mr Holbrook’s submissions rebut Professor Cole’s criticisms of Professor Roberts’ methodology. Indeed, Kent accepted Professor Cole’s conclusions as to the DARLinG data regarding Stage H.

We find no merit in the submission that Professor Cole’s method strengthens Kent’s case because it shows (disregarding the issue of ethnicity) that the probability is of the applicant being older than claimed.

Regardless of whether or not the MMMs are Stage H dependent, there is a serious gap in Professor Roberts’ evidence because there is no data on how MMMs present in Stage H individuals aged below 17.16. The youngest male in any of the Olze, Caldas, or Lucas tables is 17.16. However, it is known that a subject can be in Stage H as young as 15.47. This means that the Tribunal simply does not know what a person of AS’s claimed age (approximately 16 years 5 months at the time of the dental examination) would look like in terms of MMMs.

We accept Professor Cole’s evidence that there are different levels of validation and this validation is at the lowest level that is validation by the researchers themselves. Further, and as Professor Cole noted, there are a number of difficulties with the results.

At paragraphs 59 to 60 of his second witness statement, Professor Roberts sets out results in respect of RPV and PLV, covering 85 males aged between 16 and 26” [D97-98]. There are, however, significant difficulties with these, as identified by the applicant as although said to be the results for males aged 16 to 26, 5 of the 8 maximum ages are over 26 (26.04, 29.65, 24.23, 29.65, 26.04, and 26.03). Professor Roberts sought to explain this by saying the data needed cleansing but it was not good professional research practice to analyse and clean the data whilst a study is progressing. This does not explain why Professor Roberts signed a statement to say that the table recorded the results for 85 males whose age ranged between 16 and 26 .

Further, one individual in the RPV table is said to be “Not stage H”. This person is aged 24.44, which is three years older than the oldest male previously recorded in the DARLInG dataset as being pre-stage H (21.64 [E47]). If so, it is surprising that such as result was not remarked upon. That said, in contrast to the RPV table, the PLV table records 0 individuals as being “Not stage H”. This is inexplicable, given that the tables purport to record the same 85 individuals.

The minimum age in the RPV table is 16.32 but the minimum age in the PLV table is 18.04, again despite the fact that the tables purport to record the same 85 individuals.

Whilst it may well be, that there is a requirement for data cleansing to be continued, nonetheless it does cast significant doubt on the reliability of the data. We consider that, at best, the issue of MMMs is a developing area and it is work in progress. We therefore endorse what was said in ZM and SK .

We consider further that, as the applicant submits, there is a lack of sufficient scientific consensus on the Olze, Caldas and other papers and that these do not provide the unequivocal support claimed by Professor Roberts who in his second witness statement at [54] stated that :

It was put to him in cross-examination that the Caldas paper on RPV expressly states that “the accuracy of predictions for the group younger than 21 years of age was low, meaning that this methodology may not be suitable for age estimation” [MB/18/1] and “using the different stages of pulp visibility as an age marker may not be suitable, as it may produce ethically unacceptable errors by claiming that a person is older than they really are” [MB/18/5]. This was particularly true for males, for whom it was said “other markers should be used” [MB/18/5].

Professor Roberts’ evidence was that this only applied to the 21-year threshold and that for the 18-year threshold it was helpful because the minimum values were all over 18 years. This explanation fails to recognise that the sample used in the study did not include any males under the age of 18. It was therefore inevitable that the minimum age for any stage was over 18: far from proving that no male under 18 was identified, the authors did not look at any males under 18.

We accept that to some extent, Professor Roberts seeks to cast doubt on and distance himself from the Olze and Caldas papers. In his first witness statement, he argued that the inclusion of the Olze data rendered the sample report provided to the Tribunal in ZM & SK inaccurate. He gave several reasons for why the Olze data was likely to be unreliable and made it clear that going forward he would not be relying on the Olze data (paragraphs 15-28 [B20-22] and 19-22 [B22-24]). In his written replies to Kent’s questions, he criticised the Caldas papers as well as the Olze research. Kent had asked him to state what age and age range he would have given AS had his data been drawn from data published by “a) Olze, b) Perez-Mongiovi, c) Sequeira, d) “any other reputable source” but he refused to do so, stating:

We note that it is still the case that there is no data for dental development in Afghan nationals. As Professor Roberts accepts, there is strong evidence about the differences in development and this is so particularly in respect of MMMs. Professor Roberts himself accepts that “there is strong evidence of ethnic differences” in respect of tooth development [E14]. In his oral evidence, he appeared to be seeking to confine this to Tooth Dental Stage, yet in his first witness statement at [18(i)] and [22(i)] he cited ethnic differences as a reason to distinguish the Olze papers on RPV and PLV.

There is in any event other evidence showing that ethnicity may affect the development of MMMs. In the Caldas paper on PLV, the authors noted that “Differences between studies in different populations are evident, suggesting that specific population standards should be used when applying this technique”. The same conclusion was drawn by Caldas in respect of RPV.

Whilst we note Mr Holbrook’s submissions that the likelihood is that any difference on account of ethnicity is likely to be small and irrelevant, that is not a matter which is relevant to assessing whether or not this is reliable evidence. Further, contrary to what is said, there is evidence certainly in this regard that some populations reach development stages earlier. In the report entitled “Comparative study on the effect of ethnicity on wisdom tooth eruption” from Olze and Others (Footnote: ¹⁸) ^, 2007 it is noted that a study pairing American white men and women and the black African population showed that dental development in the latter was faster (see Chagula (Footnote: ¹⁹) ). Another study indicated that the upper wisdom teeth of black Americans emerge 3.7 years earlier and the lower wisdom teeth 5.6 years earlier than white Americans. Similarly, studies of South African men indicated they developed earlier, the conclusion being that “according to our study results, we conclude that forensic age assessments should be based on population specific reference data”.

The applicant submits that Professor Roberts is an unreliable witness and that despite stating that he has learned from ZM & SK , he has not done so and has continued to present his evidence either relying on unreliable data, or failing to mention difficulties. Attention is drawn to several factors.

The continued use of attainment of stage H as a measure of age, contrary to what he had said in ZM & SK at [65];

Inconsistencies in the value of using Root Canal Width (“RCW”) as an indicator of age;

Overstating the value of the Encyclopaedia Chapter;

Reliance on a validation study which (as set out above) contains many inaccuracies;

failing or refusing to recognise that he had erred in continuing to produce reports relying on Stage H even when he knew that method was unreliable in a significant proportion of cases;

Resiling from a large part of his report.

We deal with these in turn.

We noted that in oral evidence Professor Roberts denied that giving probabilities in his report on AS that he was under or over 18 as a result of being in Stage H was a measure of age and that it was just a “threshold assignment” . We do not accept this. He could have simply noted that AS was at Stage H then proceeded to consider the MMMs. The fact that Professor Roberts does not suggest what specific age AS might be does not make this any less an assessment of probable age.

With respect to RCW we note that in his second witness statement, at [D96] Professor Roberts stated that it was perhaps too early to start using it, yet in, SK v Croydon (JR/2414/2016) he had produced a report (dated 2 ^nd November 2016) relying solely on RCW. Further, in oral evidence in this case he stood by that SK report. That is inconsistent, and not what we would expect from an expert witness. It is also troubling that he said that in this case he would have used RCW had the image been of sufficient quality.

It is also of note with regard to evidence presented in earlier cases that in ZM & SK, Professor Roberts relied on a sample report which he now accepts in his first witness statement was inaccurate. While we would not agree with the applicant that this amounts to seeking to mislead, it is equally worrying that this could have occurred.

There is a degree of exaggeration in the covering letter sent with the report in this case which stated that:

Professor Roberts accepted that he should not have said “all” and it is worrying that the chapter presented as reliable a method which is now known to be wrong in 50% of cases between 17 and 19. We found the explanation for that given in oral evidence to be unsatisfactory. We accept that there is a time delay in publishing articles, but we do not accept that it was right not to have drawn attention to difficulties in assessment.

Before us, Professor Roberts sought to rely on a validation study which, as noted above, contains significant errors. It may well be, as he said, work in progress but it is of concern that it was put forward as evidence to the Tribunal in a witness statement.

We consider that there is here a pattern in Professor Roberts’ conduct. There is a strong tendency to overstate the value of research he has undertaken and his methodology. This extends to overstating the weight of the validation of his work, an exercise which, as became clear in oral evidence, was far from satisfactory.

Further, and of greater concern, is the significant evidence of him giving assurances about future conduct when faced with the flaws in his methodology yet failing to put them into practice.

We regret to say that we did not on this occasion find Professor Roberts to be a reliable witness. Nor do we find that MMMs are of any use in assessing age.

The report concluded that the applicant was 17 years old, his birth date being fixed at 7 September 1998. This was based on an interview carried out on 6 November 2015. It was noted that he appeared physically older than 15, and seemed independent, his interaction with peers suggesting he is older, and that staff believed him to be 17 years of age. It is also recorded that they believed him to be within the age range of 17 to 19, and giving him the benefit of the doubt, assessed him at the lower age of the range, taking the date of birth as 7 September 1998 as he believed his date of arrival in the UK was his 15 ^th birthday.

The assessors observed a lack of obvious signs of mourning over the loss of his father, and a lack of any information given about village life or his community, his description of his life coming across as existing in a vacuum, yet his behaviour in going missing and going out into the community appeared inconsistent with this, as was his assertiveness.

As noted above, the respondent conducted an additional assessment on the applicant on 9 February 2017. That interview was conducted with the assistance of an interpreter; there was also an appropriate adult present. What occurred in that meeting is set out in her statement and also in the notes of the interview prepared by Melissa Carter.

Rose Corby noted ([9]) that she had taken account also of the age assessment completed on 6 November 2015 and additional observations by the on-site support workers, Danielle Weston and Dave Ansell. Contained within the witness statement are photographs of the applicant taken on 8 September 2015 when he arrived at Appledore; around 24 August 2016 and on 23 February 2017. Full size blow-ups of these were provided at the hearing.

Rose Corby stated that her initial impression of the applicant was that he was significantly over the age of 18 having adult features such as a prominent Adam’s apple, defined jaw line and developed muscle tone. She also observed signs of aging such as lines on his neck, forehead and jowl lines beginning to show, and that his skin was losing lustre. She also observed old acne scarring which had completely healed. She noted that the skin looked weathered and not smooth, discounting the possibility that premature skin aging was due to his living outside as he said that he spent all his time indoors.

Rose Corby concluded ([12]) that the applicant presents as older than the claimed age and would fit more appropriately to be an age between 22 to 26 based on his physical appearance and demeanour and taking into account the collated photographs of young people and young adults aged from 14 to 26 appended to the report. She stated:

Rose Corby noted ([13]) the applicant appeared to have difficulty in estimating ages particularly when asked about his parents and that it seemed strange that he was unable to know about the general significance of age within his community, that is, whether it is something that others are likely to know about or not. She noted that in contrast to this ([15]) that he had spoken about his younger brothers and was clear that one was 7 and one was 5 when he left, clarifying that there were discussions where his mother would speak about their ages which she found implausible if he had been told his age only when he was leaving. She noted that the ages might have been written in the Holy Koran but does not know whether or not that was so, concluding that he had used the discussions to provide false evidence of his claimed age ([16]).

Rose Corby noted that the applicant appeared to be coping very well with social situations and the learning and behavioural expectations of a college environment, being remarkable progress for someone who said he had not been allowed out, only socialised with his younger brothers and had never been to school, albeit that he had said in an earlier assessment that he played with other children from a local mosque. She noted also that a decision had been taken that he would not attend the school further away from the house due to the risks from the Taliban.

Rose Corby considered ([18]) that the applicant was deliberately trying to come across as young but was not familiar with what information he should provide about his level of self-care skills to be consistent with the claimed chronological age as that was not a significant part of his culture. She concluded ([18]) that he was trying to minimise his capabilities and responsibilities in the family home with a view to indicating that he was younger than his true age, raising doubt of the credibility of the picture he had provided of family life and his childhood experience. She again concluded that the account given of how and when he was allowed to stay at home alone was not plausible and was evidence of the applicant manipulating his account. She noted also ([20]) that he appeared to be seeking to distract from questioning, providing further evidence of his manipulating his answers. She did not find it credible ([21]) that the applicant would not have sought more stimulation than he had said and the activities described were unlikely to have satisfied him as an adolescent. She noted ([21]) he again appeared to be distracted and not answering questions, concluding that he did not want to give a true account of his activities. She noted also ([23]) that he did not give the impression of someone who had led such a sheltered childhood as described and ([25]) that he was deliberately trying to give a false impression of his development in his family in Afghanistan.

Turning to the applicant’s journey to the United Kingdom ([26]) she noted that his account was not as detailed as young people are often able to provide ([27]) resulting in it not being possible to build a timeline of his journey and that he said he had known that it was three months long because others on the journey had counted the days but that there is no way of knowing if it was the same travelling companions from start to finish and there are doubts about the accuracy of the estimate. She noted also ([29]) that he had been at points sarcastic when answering questions on this matter.

Turning to the applicant’s education in the United Kingdom, it was noted that he attends Bexley College three days a week and from the support workers Danielle Weston and Dave Ansell, that he had come far with education since arriving in 2015, his English having improved massively and that he could hold a conversation easily although his understanding of time is more questionable. It is noted also that his attendance at college is good; his attitude to staff and behaviour during lessons is very good, that he completes his work and that in December 2016 he had exams where he was assessed at entry level one, English Speaking Board (International) describing this course as being “aimed at candidates who have very good literacy knowledge or experience in English”. It transpired during cross-examination she had identified the wrong test.

Rose Corby noted that he appeared to be able to travel around with relative ease ([37]) and that he was seeking to minimise his capacity to undertake tasks for himself, especially domestic tasks ([38]) and that there are concerns that he is dishonest about money and how he is spending it ([39]), appearing willing to deceive professionals who are offering support around money which he is requesting; that he is showing confidence, independence and making decisions for himself and is not seeking the support of adults when he is determined in what he wants such as purchasing larger items.

Rose Corby noted ([41]) that social work assistant Patrick Shier had attended on 6 and 9 March 2017 to offer support with independent living skills including budgeting and he appeared to be resistant to this. It was noted also that he appears to have made a financial arrangement with a friend to purchase an iPhone at £400, arranging to pay a lump sum and the rest in instalments showing advanced money skills. He appears also to have laughed when Patrick Shier and Danielle Weston challenged him as to why he had lied about Danielle lending him money ([42]). She noted also ([42]) that the applicant did appear to be able to undertake domestic skills contrary to what he had told them and that he appeared to want to be a family because he thought he would have homemade meals made for him and help with his education and also appears to expect a carer to be there all the time ([43]).

It was noted also that he expresses concerns about ill health stating that he is still in pain despite having had a kidney removed some six or seven months earlier.

Rose Corby considers ([49]) that it was very unusual that the topic of the applicant’s age would come up casually just before he left home and that his mother would be so specific about it – stating that he was 14 years and 9 months old and that his journey was three months long, saying that when challenged he said that his mother had actually said he would soon be 15 ([50]). She concluded at [52]:

Rose Corby noted ([55]) the applicant’s overall attitude was making lots of complaints, that he was unable to explain what he had done for himself. She stated that in her view AS’s experience of life in the UK is not matching up to his unrealistic hopes, the easiest option being to blame others for putting him in the wrong situation. She noted also that there were moments ([57]) when he became sarcastic in his answers, behaving in a sophisticated mature way which was inconsistent with the uneducated background and lifestyle he claimed to have had [57]. The conclusions are as follows:

The information about how he knows his date of birth is unreliable, implausible, inconsistent and vague

AS continues to make complaints about his health concerns and that no-one is helping him, however recent information from his support workers shows that he has been supported to seek numerous medical opinions and only two weeks previously was examined by the doctor who could find no problems. During the assessment he therefore was giving a false impression on his needs, particularly in relation to their impact on the review process

AS is making good progress in managing the behavioural, cognitive and social expectations of his education placement. This would seem unlikely given the isolated childhood experience he has described during the assessment and raises doubt over the authenticity of his account.

AS's account is that as a child he was able to stay at home with his much younger brothers, however he was dependent on his mother in relation to personal hygiene and he claims his mother washed him. In my view this is improbable and it is more likely that AS has tried to minimise his level of independence and skill when he was living in Afghanistan to try and support a younger claimed age.

In the UK, AS said in the meeting that he wants to live in foster care and also in contradiction said to his allocated social worker Ben Holbrook that he wants to live in a house by himself. This indicates that he wants to have the domestic support of carers to cook for him but the freedom of living alone. AS is aware that his placement is informed 'by his age and therefore this gives clear motivation for the information he has provided in his assessment to be shaped by a desired age rather than from memory.

I do not accept that it is plausible that AS's mother would regularly discuss his younger brothers ages in everyday conversation but had never mentioned AS's age, unless she did not actually know AS's age. In this scenario, it is unrealistic that she would have told him that he was 14 years and 9 months old, and AS has given conflicting accounts on whether she actually said this, or whether she actually gave a more general answer. It seems more likely to me, that AS had decided that he wanted his claimed date of birth to be 15 years old on arrival so that he could access a foster placement for at least a year, and that he then calculated the age he would have been on arrival and then said that his mother told him this.

It comes across as too coincidental that AS's mother estimated his age so precisely that would mean he was exactly 15 years old on arrival, given that young people who do not know his or her specific birthday but just an age will usually be given a birthday of 1 January. At his arrival this could have made him 15 and 9 months (if the immigration officers had to allocate a date of birth in accordance with a claimed age) rather than what he claimed which was 15 years old exactly.

AS's demeanour when he was being challenged was defensive, rude, sarcastic and distracted.

It also comes across as too coincidental that AS’s date of birth came up in conversation for the first time just before he left when it had never been discussed before, despite his brother's ages being regularly discussed.

It comes across as though AS is now in a power struggle because he is adamant that he will not cook for himself, although he is capable. He sees the best way out of this is to be in foster care and have his meals cooked for him

AS physically presents as someone who is fully developed and is now showing signs of aging such as lines on his face

AS continues to express that he needs a lot of support around him all of the time, however he seems more than happy to take himself out unaccompanied to visit friends around Kent and go to college and to purchase £400 mobile phones on a payment plan.”

In her second witness statement dated 3 May 2017 Rose Corby refers to the photographs collated as “typical photos of age” [3] because on the whole they show pictures typical of 14, 16, 18 etc. year olds, and states:-

The witness statement then sets out certain features and how those appear in those aged 14 to 18. These are then compared with those aged over 20 and finally how they related to the applicant. The features of acne, Adam’s apple, eyebrows, facial hair, facial shape, jaw line skin lines and jowls, skin under the eyes, muscle tone, are then addressed.

Turning to demeanour/behaviour, she noted that the age assessors in November 2015 observed the act that he held strong eye contact which could be perceived as passive aggressive; he gave a long, fixed glare and when asking questions did not show signs of emotion. She did not consider it would be the behaviour of a typical 15 year old. She noted also in her interview in March 2017 AS did not demonstrate the behaviour of a 16 year old but his behaviour was consistent with the assessment of him being 22 or older.

Rose Corby then goes on to give her overall view on age.

In cross-examination Rose Corby confirmed that the interview with the applicant had taken place over approximately three hours including breaks. She said she had continued to observe him during breaks.

Rose Corby explained that the delay of two months in preparing the statement was due to a long period of analysis which had been going on during which she had read through the notes. She said that there were no longer any notes of hers in existence as she had used those notes, expanding upon them, to create her statement. She confirmed that the report had been the subject of discussions between herself and Melissa Carter and also with their line manager, Rita Boboye. She said it was her understanding that it would not have been possible given that court proceedings had commenced, to have a full “ Merton ” assessment.

She also referred to undertaking a “pro forma” by which she meant a specific form which would have been sent to the Home Office and it would have been necessary to advise the applicant that his legal age might change if this had been done.

Rose Corby was asked about the letter from Kent Legal Services dated 24 March 2017. She confirmed that it was a review not a reassessment and it had been done in accordance with the ADCS guidelines. She said that they had applied what they thought was relevant in order to be fair but that they had not said to the applicant that she could change his age as it was not appropriate to say that the outcome could change the legal date of birth as she had understood it. She said that contrary to what was said in the letter as Merton and ADCS compliance were different. She accepted that there was a theme of fairness underpinning “ Merton ”. She considered that it had been fair as the applicant had had the opportunity with legal support to answer any point that had been written in the statement. It was put to her that this was not correct. She had been giving an opinion rather than a “minded to refuse” analysis. It was put to her that the exercise was not Merton compliant.

Asked about the observations made by Danielle Weston and Dave Ansell, she said she had not spoken to them but that there had been electronic communication.

It was also put to Rose Corby that her view was at odds with that of Dave Ansell.

She said that in her experience the role of an age assessor was different; it involved collecting as much detail as possible and that the view of an age assessor might be different from a key worker. She said that they had asked Dave Ansell and Danielle Weston for a view and their views had been included in the statement. She said that she had not had the email referred to [D11] but if it had come to her it would be in her report. It was put to her that this was part of AS’s statement which she said she had read though she had not seen the attachments. She accepted she had missed it but denied that this showed this was an unfair process. She did not answer whether, had she seen the letter from Dave Ansell, it would have prompted her to ask more questions as support workers like him have a different role from age assessors. They spend a lot of time with young persons and can find it difficult to keep the same objectivity as an age assessor.

Cross-examination then turned to the photographs of the applicant which Rose Corby said had been provided by the children’s services and were available on the electronic file.

Rose Corby said they had tried to make the process as fair as possible. It was put to her that she had not said in her statement that Melissa Carter was part of the process, her witness statement stating only that she agreed with Rose Corby. She did however say that the witness statement she had produced was a joint effort and that she had used “I” as it was a witness statement from her.

Rose Corby was asked to explain the delay between visiting the applicant and the witness statement which was only signed on 4 April 2017. She explained that she would have made notes of interview which were then converted into the witness statement and so there were no original “notes” left as they had now all been incorporated into the witness statement. She accepted this scenario was out of the ordinary and it was not correct to say that the whole statement had been written on 4 April 2017. She said they had discussed their observations in the office.

Rose Corby said that Melissa Carter had also observed the lines on the applicant’s neck and it is her experience that post-puberty, it is necessary to look for subtle differences indicative of age change. She accepted that she was beginning to talk outside of her remit when discussing whether Adam’s apples became more prominent, but considered that it was an indicator of aging and development.

Discussion then turned to the issue of jaw line. Rose Corby said that her comparisons regarding the jaw line were not confined to experience of white males, that she was relying on her broad experience of working with other young people from Afghanistan. Asked about the age of puberty she said that it ranged between 12 to 13 for males but it could be younger or older.

She said that she did believe that the applicant had shown acne scarring (witness statement page 10). She had seen it on the earlier day of the hearing but not today. She said that they were more prominent in February, that people can change their experience and can mask scars. She said in her experience young people went through stages of acne infection.

Asked about referring to the applicant’s skin as weathered and not smooth, she said that it was an indicator that someone was not a child though she accepted that if somebody had spent their life out of doors it might be different and that in conducting a holistic age assessment if someone had spent their life out of doors this would have been taken into account as indeed would be a long journey but that was not said in this case.

Asked about the photographs which formed part of the evidence, she had been given a booklet of these by her team manager. She said she cannot be sure but she thought they may have come from the legal department. She said it was in a yellow folder. She said that they represented a general cross section typical for people of that age group. Asked about the links to the photographs, it was put to her that she knew full well that all the guidance and case law counselled against looking at physical appearance, she replied that physical appearance could not be excluded. She said that she felt in this case the photographs are a tool to try the points she had made. She said that using her experience the photographs were typical. She confirmed that photo four on page F214 was typical of a 16-year-old. She said that she had been given the booklet early on in the process. She was not aware of them being used in any other age assessment. She denied that these were ex post facto justifications and accepted she had not seen a booklet put together like this for training. She said that neither she nor her colleague would have done anything unethical. She said that she described the photographs as typical, drawing on her experience. She said that she considered that the photograph of the person on E227 was like the applicant although he has a full beard (the applicant does not). She also considered that photograph two was like him concerning the eyes. She accepted that the applicant does not have a moustache. (Note to self: the photograph has a receding hairline.) She denied that she was “reverse engineering” her conclusions and it was difficult to say whether or not the applicant had a receding hairline.

It was put to her that the photograph album was a faulty tool but she did not accept that. She said that the photograph booklet was a tool used to illustrate her observations but did not know who it had been provided for.

Rose Corby confirmed that the applicant appeared to have a number of soft skills such as when going to school that one must stay in the lesson, exercise discipline and so on, which a lot of people who have not been to school before struggled to adapt to. She said she had not asked the college about him but had looked at the records. She said that the description he had given of his upbringing in Afghanistan, being very isolated with few social relationships which were very limited and not going to school was quite extreme but he had adjusted.

Asked about the applicant’s personal education plan and the reference to the course at which he was attending being aimed at candidates who have very good literacy, knowledge or experience of English, she said this was gleaned from evidence provided to her. She was happy to accept that in fact the course was aimed at a lower level, it being a slightly different course. She denied that this was “all of a piece” but she did not recall who had looked up the information. She did not accept that this had coloured her assessment. This was due to the many discrepancies in his account. She said that the course he was attending was provided to people who were 7 to 8 but that age related to the level of those whose main language was English.

Rose Corby said that she would have expected the applicant to have shown more creativity in his behaviour and explain what he did to create stimulation. She said that this account of how he had lived was inconsistent with him becoming sarcastic and with his attitude.

She did not accept that what she was saying was pejorative ([29]). She was unable to confirm whether Dave Ansell had ([35]) said that the applicant’s English had improved massively and had come far with education.

Asked about her second witness statement at paragraph [17] she had said that she could not comment on the applicant’s ability to speak or understand English. She explained that she had been able to refer to this in her first witness statement as this was based on what Dave Ansell had said and that she had used an interpreter to communicate with the applicant. She said that she had needed an interpreter to be “Merton compliant”.

Asked about paragraph 38 of her witness statement in which she had said that there was conflicting information about the applicant’s attitude and ability to do things for himself, she said that at times he said he was unable to do simple things yet other times was able to do things such as travel with relative ease.

Rose Corby denied trying to paint the applicant in a bad light, adding that she had drawn on her experience and information supplied. It was put to her it was not ethical to come to such a conclusion without putting these points to the applicant and she was satisfied the point raised was fair. Asked about the evidence at [42] that the applicant had entered into a financial arrangement with a friend to buy an iPhone for £400, arranging to pay a lump sum and the rest in instalments, she said she did not know whether he had actually gone ahead with this.

She said that when referring to the applicant being “quite derogatory” she said that she had said this because the applicant had said that he did not celebrate birthdays. She accepted this did not appear in the notes of the interview [F201], adding that what she had said at paragraph [59(a)], that the information about how the applicant knows his date of birth is unreliable, implausible, inconsistent and vague comes from her analysis, confirming that Melissa Carter had also used these same adjectives. She did not accept that Kent had wanted a reassessment to be done to show that he was older. It had simply been done to get an update. She said it was not fair to say that she had been fed information from outside to get the result. It was she who came up with the new date in consultation with her manager.

In re-examination, Rose Corby was asked about the photographs of people who appeared in the booklet of typical ages. She was asked at various times to explain why they were typical of the given ages, adding that the applicant had various features which confirmed this.

We then heard evidence from Melissa Carter who adopted her witness statements.

In cross-examination, she confirmed that the statement of Rose Corby had been a joint effort and she did not recall any disagreement although there had been discussions. She said that Rose Corby was senior to her. Asked why they did not just do an age assessment in the usual way she said that this was what they had been told to do, apparently this coming from the legal department. She said that they had followed the guidance and it was fair. She said that she had not been asked to use photographs before or since, that age assessment was a continually developing area. The photographs had been provided by managers but she did not know from where. She was asked to explain why if her statement was dated 28 March 2017 she said she had seen Rose Corby’s witness statement yet it was dated 4 April 2017. She said that she had seen a draft.

We then heard evidence from Patrick Shier who adopted his witness statement. He said that he had described the applicant’s English as “perfect”. He says normally asylum seeking children needed an interpreter which he had never needed with the applicant except when he was obliged to do so in interviewing the applicant after he had absconded. He said he had first got to know him at Appledore where he had worked until Christmas and again came in contact with AS in October 2016. In cross-examination, when asked about the applicant absconding he said that he had told the support worker who had been present on that occasion to ask for the CCTV cameras at the shopping centre to be looked at. He said he knew the shopping centre in question and that it had not been possible for the applicant, as far as he knew, to have left and that the applicant’s story about being forgotten in the toilets could not be true. It was put to him that nobody else had said that the applicant’s English was perfect and that his results from Bexley College indicated to the contrary. He said that he did not know about how well the applicant was able to read but that his speaking was good. He said that he had been told by other social workers. He said that he had never needed an interpreter when dealing with the applicant. He referred to the CHIN document [Volume 3, F1]. He said that he had assessed the applicant at 17 on arrival and said that he thought that he could have been as old as 21 but he could not give an exact date nor was he expected to do so.

He confirmed that, as with the other young men, he had been given a razor but could not recall whether he used it or not as they did not enter the young people’s rooms. He said that he had stuck by his observation that the applicant had shown a five o’clock shadow, that is what he believed to be the case. In respect of the second witness statement he explained that he had spoken to Danielle Weston about the applicant’s phones. He said that she had been told by Edward Taylor of the applicant’s solicitors not to provide the information. He had obtained the information about the telephone numbers from her.

In re-examination he said that he had only estimated the applicant’s age as he was not qualified to give an opinion on an exact age. He said that his conclusion was based on the applicant’s behaviour, adding that the applicant had been an accomplished pool player from the beginning and also played football well.

He said he had gone by the youngest age he had observed in Appledore. In response to our questions Mr Shier said that he had been supervising the applicant who was one of about 44 boys and young men all day, every second day during the time he had been in Appledore which was a period of several weeks.

We then heard evidence from Cliff Wheeler who confirmed that he is a social worker’s assistant. He adopted his witness statement. He said that he had met the applicant on a couple of occasions when he was charged with taking him to and from various appointments. Asked why he thought the applicant’s demeanour to be manipulative, he said that during the conversation the applicant kept turning around what he was saying, trying to discuss his kidney problems and the pain he suffered as a result.

The applicant’s evidence is set out in his witness statements and in his oral evidence. In his first witness statement the applicant said that he grew up in Afghanistan, either helping at home or visiting the mosque. He said he could not recall what age he started to attend the mosque but when travelling to and from that he would play games with other boys in the village. He also helped his mother at home, including helping look after his younger brothers. He said that relatives would only rarely visit the house but when they did they would ask his age and so his mother would discuss the ages of him and his two brothers but he did not always pay attention to this when he was younger but did recall her telling him on the day he left that he was 14 years and 9 months old.

The applicant said that his father was a commander fighting against the Taliban and was often away from home.

About a month before he left Afghanistan, one of his father’s bodyguards told the family that the Taliban had taken his father from the post where he had been working. The Taliban came to take him twice over the next few days and he thought that is why his mother had arranged with his maternal uncle and cousin for him to leave Afghanistan. With the assistance of an agent he was taken out of Afghanistan. They ill-treated him during the journey which he thought took about three months including about a month spent in “the jungle” in Calais and that he believes that he was 15 years old when he arrived on 7 September 2015.

Since his arrival the applicant has had to have a kidney removed and is still in pain. He has not been able to contact family in Afghanistan. He also said during the age assessment he was in some pain and became nervous when they told him he was 17 years old because he was 15. He said that he thought he had told the truth.

The applicant also said that he has mental health problems.

In his second witness statement the applicant said that he wished to add that he recalls going to a mosque aged 13 and 14 to learn the Koran and that the mullah had told him that when he turned 15 it would become obligatory for him to start praying and fasting. He had also been pressured to learn the Koran. Exhibited to that witness statement is a letter from David Ansell and a letter from his teacher.

In his third witness statement the applicant sets out the details of the phones he has owned including the purchase of a second hand iPhone 4 which he had used to make phone calls using WhatsApp. This is why there were few phone calls disclosed in his bills.

Commenting on the witness statement of Cliff Wheeler he said he did not recall missing appointments and that, contrary to Mr Wheeler thinking that he looked over 18 when he saw him in September 2016 he was just 16 and does not think he looks older than his age. He denied being manipulative or ordering people around.

He said that Ben Holbrook’s statement contained a number of errors in that the applicant is doing an entry level 1 course not a level 2 course in English for Speakers of Other Languages and that he is not that comfortable using public transport. He said that contrary to what Ben Holbrook says he is struggling a lot with reading and writing and could only speak a few words of English when he arrived having learned these from people in Calais. He said he is not likely to go to Bluewater Shopping Centre on his own as he has difficulty understanding things.

In response to Patrick Shier’s statement he said that he did not have an aggressive and domineering attitude whilst at Appledore though accepted he did sometimes misbehave. He said that contrary to Patrick Shier saying that his English is perfect, he said he thinks his English is bad and that he faces a lot of difficulties on a daily basis and recently failed a test. He said that he did not think he looked older than his age and that when he uses public transport he is given child tickets so people must think he looks like a child. He denied having a facial five o’clock shadow, only going to the barbers once a month for a haircut and a shave, only growing patches of facial hair on his chin and sides of his head.

Responding to Rose Corby’s statement he said that he did not think he had a prominent Adam’s apple, defined jaw line or developed muscle tone. He said he did have a line on his forehead but did not think this was because he was older than 16, that his body is still changing. He said he has never really had any spots and does not think he has any scarring. He denied being dishonest and he had not been inconsistent.

In cross-examination, when asked how many people lived in his village, the applicant said he had not counted them; asked to guess, he said he did not know, nor was he able to estimate the number of houses. He could not say whether there were more or less than 100 or if there could have been as many as 1000. He said his house did not have electricity, and water was fetched from a well. He said that there were two or three shops in the village, and that there was a mosque.

Asked if there was a school in his village, the applicant said he had not seen one. Asked if he had friends, he said that there were boys with whom he went to the mosque. He had not asked them if they went to school, nor how they spent their day.

It was put to him that children are naturally curious, and he said that some may be. He said that the boys with whom he had gone to the mosque did not tell him what they were going to do when not at the mosque. He said that they may have gone to school but that he did not know.

Asked to explain why he had been able to confirm in his initial age assessment that there were children in the village who did go to school, he said he had told then that there may have been such children, but could not explain why that had not been recorded. It was put to him that during his age assessment he had said there were not many houses, and that there was a school; he said that he had not calculated the number and that he may have said there was a mosque he attended. Questioned further, he maintained he had meant there was a mosque, and then that there might have been a school outside the village. He denied being quite well educated.

The applicant said he had learned a few words in English during the time he had spent in Calais but that this was very limited. He said he did recall being collected by a social worker (Mr Kelly) and driven to a residential care centre. He denied saying he grew up knowing his age. He said he had been asked his age and said 15, as his mother had told him he was 14 years and 9 months when he had left Afghanistan 3 months earlier. He said that is what he had told Mr Kelly but did not recall saying that he grew up knowing his age.

It was put to the applicant that his English was not that limited. He did not accept that. He said also that he had not told Mr Kelly that he had learned English at school. He said his English was not of a standard to say that, and did not know what he said. He said he would not have known how to say his skin was itchy, or that he had wanted to go on Facebook and that he had been able to say “yes” to questions. He asked how he could have given all the answers if he had little or no English.

The applicant recalled that he had had met Ben Holbrook while at Appledore but did not accept his comment that he spoke English well while at Appledore but did accept that it improved there.

The applicant confirmed that he had seen Patrick Shier at Appledore and had spoken to him sometime but did not agree that his English was “perfect” and that his English is not that good.

The applicant said that his father had been a police commander and his post had been attacked by the Taliban. He said one of his father’s bodyguards had told his mother that the father had been taken; his mother told him of that a few days later, and that he had been presumed dead. She had later told him he is dead. Asked to explain why during a home visit on 25 August 2016 he had he had said he thought his father was dead, he said that had said his mother told him that.

Asked his mother’s age when he left, he said 30 to 40. He explained saying she was about 60 when asked on 6 November 2015 as being due to being asked too many questions and being stressed. He denied she and his father were about 60, and confirmed his siblings were 5 and 7 years of age. He said his mother had told him their ages, and when neighbours had visited she had told them the ages.

The applicant was asked if he was curious about his age, replying that when he was 13-14, the Imam at the mosque had told him he would soon be 15 and would have to say the prayers like an adult. He said that his mother had told him to say, if anyone asked him, that he is 14 years and 9 months of age. He did not know why it had been relevant at that time. Asked if his mother had never told him his age before, he said she may have told other people. He then went on to describe being given presented such as clothes, on the basis that he was told this was because he was 13 or 14. It was not a birthday celebration as such, and had started when he was about 12. He said it was not correct to say that he grew up knowing his age.

It was put to him that in the initial age assessment he had said he had not known his age until he left Afghanistan. He replied that his parents had told him from 12 onwards. Asked whether he had been told “you are 14 years and nine months”, or as was recorded elsewhere, that he would soon be 15, he said his mother had said both. He said that although his mother had spoken of his siblings ages she had not mentioned his.

The applicant said the date of birth 1 January 2000 had been assigned to him by the Home Office.

Asked about his age now, the applicant said he would be 17 in September, and denied being 24. The applicant did not accept that his mother had told him an age which was younger than she knew him to be so that he would get better treatment.

The applicant said he had not contact with family in Afghanistan, and had an appointment with the Red Cross soon. He said that his health and being in and out of hospital was the reason for the delay. No appointment with the Red Cross had yet been fixed. It was put to him that the social workers had arranged this, and he had asked for the appointment to be cancelled. It was put to him that he was avoiding making contact.

It was also put to him that he had said his mother has a mobile phone, and that he had said he did not know her number. He denied that. He said that R S (F153) is his maternal uncle’s son, and did not accept that he had told the social worker taking him to Appledore that he could contact him. He denied contacting his cousin. He also said that what had been recorded during the review meeting on 5 February 2016, at which an interpreter had been present, that he was in contact with his mother, was not correct.

In re-examination, the applicant said that he had not previously seen the documents prepared by social workers for Kent, nor had anyone translated them for him. He said that he attends Bexley College and had recently failed an exam in reading at Entry Level 1.

Although the applicant’s witnesses gave evidence after the experts, we consider that that it is, in this case, more appropriate to assess first the evidence from the professional witnesses relied upon by Kent.

It is remarkable that in this case Kent do not seek to rely on the first age assessment conducted by Ben Holbrook and Anita Budai, a social worker of considerably greater experience than either Melissa Carter or Rose Corby. That assessment, carried out in November 2015, assessed the applicant to be 17 with a date of birth of 7 September 1998. On Kent’s case, as now put, that assessment was out by as much as six years.

The circumstances of the review conducted by Rose Corby and Melissa Carter are unusual; it was accepted in evidence by Rose Corby that there had been no intention to carry out a “ Merton ” compliant assessment yet correspondence between the parties indicates that the intention of the review was that if a significantly different conclusion might be reached, they would advise a fresh assessment. We do not, however, draw any adverse conclusions from this.

It is submitted by the applicant that the review undertaken was unfair, and that Rose Corby had, in particular, acted unfairly, taking the most negative view possible of the applicant and failing to take into account evidence from others, such as David Ansell which was more favourable.

The respondent submits that the interview and review was fair, and objectively carried out. It is accepted that it was not, strictly speaking, “ Merton ” compliant, but that it is reliable, and, apart from the dental evidence, it is the best objective evidence before the Tribunal.

Before analysing the evidence of Rose Corby, we consider it appropriate to consider the booklet of photographs which has been produced. Most of paragraphs [2] to [13] of Rose Corby’s second statement relates to the booklet.

The respondent seeks to rely on the booklet of photographs said to be of males typical of their ages, arranged by age groups. We take judicial note of the fact that the human population is in excess of 7 billion, and it follows that there must be many millions, if not tens of millions of young men who fall into the age groups identified in the selection. We do not know who compiled the selection, or the criteria by which they selected the individuals as being “typical”. That in itself is a word which requires further consideration. It is not legally defined and we see no reason to depart from the ordinary meaning of the word that is having the distinctive qualities or characteristics of a particular type of person or thing. It thus presupposes that the distinctive qualities or characteristics, in this case of males of a certain age, are identifiable. We have been provided with no reliable evidence that this is so. Further, unless such characteristics are identifiable, it is not possible to state, except in an entirely subjective manner, that an individual is a typical 16, 18 or 20-year-old male.

The respondent relies on an example of a well-known journalist often taken to be much younger than he is. This, we consider, reinforces the strong thrust of the case law and guidance that little weight can be attached to physical appearance.

We consider that these observations apply even more so to photographs which are not three-dimensional. How photographs are lit, the type of the exposure, the quality of the camera and other factors, not least including the clothing a person wears can affect significantly their appearance, to such an extent that we question why it could ever have been thought that the compilation of an album of photographs could have assisted anyone.

Further, the stated purpose of the collection is not at all clear, nor is its provenance. It is telling that neither Rose Corby nor Melissa Carter could give any example of it being used elsewhere, or in any other age assessment process. It is also telling that Rose Corby accepted it from management without question, despite the strong strictures and guidance militating against attaching weight to physical features.

There is simply no evidence, other than that which arose in cross-examination of Rose Corby, that it was seen as potentially useful to the Tribunal as a means by which Rose Corby could identify certain features such as a prominent Adam’s apple, as being typical. This response has all the hallmarks of desperation to justify use of the booklet in the full knowledge that this was unfair and unprofessional in an age assessment. We wish to make it clear, that if that was the true intention (which we doubt), it failed; we found the booklet to be of no evidential use whatsoever.

In submissions, Mr Holbrook gave further reasons for the use of the collection of photographs: that they were useful for benchmarking age assessment, and that without proper benchmarking, (and by extension, exposure to adolescents and young people of known age) it is easy for social workers involved in age assessment to make errors.

There is, however, insufficient evidence that such benchmarking by use of photographs is necessary, desirable or even professionally mandated. There is no proper evidence identified that over time, social workers involved in age assessment, are more likely to assess young people as younger or for that matter older than they are; Mr Holbrook’s metaphorical yardstick would presumably extend in both directions. It is also somewhat strange that this justification is put forward when the evidence is that it is only in the assessment of this applicant that photographs have been used in this way.

Accordingly, insofar as Rose Corby’s evidence is drawn from the booklet of photographs we attach no weight to it. If anything, her reliance on it diminishes her credibility. Her second statement in consequence adds little; it has all the hallmarks of an ex post facto justification for her earlier testimony.

We reject the applicant’s submission that Rose Corby’s evidence is wholly unfair. Still less do we accept the unnecessarily emotive submission that Kent’s witnesses have attempted a character assassination of the applicant.

There are, however, as noted above, matters upon which Rose Corby was mistaken: the level of English being studied, and there is also her failure properly to consider Dave Ansell’s evidence. While we conclude that these errors could have arisen from pressures of time and work, and were not deliberate, nonetheless the issue of competence in English was a significant factor. We note, however, that she sought to diminish the significance of the material missed. She similarly sought to diminish the significance of the error about the course being followed.

There is another aspect of Rose Corby’s evidence which troubles us. When asked about acne scarring, and it being apparent that there was none on the applicant’s face while he sat in court, having said there was some before, she began to give evidence about how people conceal scarring using cosmetic products. We formed the distinct impression that she was seeking to blame the applicant (and to cast aspersions on him) on the basis of little or no evidence. This fits into a pattern of seeking to deflect any suggestion of fault on her part onto others, as with her attitude to the omissions noted in the paragraph above. This is not what we would expect from a professional witness.

It was submitted that there are instances in Rose Corby’s witness statement at [14], [21] and [22] where she summarises what the applicant had said (recorded in the notes of the meeting prepared by Melissa Carter) in an inaccurate way, casting doubt on the applicant. These are instances where the age of consent is said to have been mentioned by the applicant, omitting the whole of answers about whether he found having to live indoors boring and omission of a direct answer about whether he was proud of his father.

While it is necessary to summarise what was said, equally that must be done fairly in order to avoid creating a misleading impression. Given that what was omitted were direct answers to questions and the criticism of the applicant is not answering direct questions, the impression given is misleading. A single instance of this might have occurred by accident, yet there is not just one instance, but several. Further, this is a witness statement we were told had been worked on over a period of weeks. We do not, in the circumstances, consider that this mischaracterisation of the evidence was inadvertent and it is misleading, casting significant doubt on Rose Corby’s professionalism and reliability as a witness.

Taking these factors together, and viewing all of the evidence in the round, we did not find Rose Corby to be a reliable or objective witness.

It is difficult to attach much weight to the evidence of Melissa Carter independent from that of her colleague Rose Corby. It is, however, of significant concern that she signed and dated a witness statement stating that she had seen and approved Rose Corby’s statement when that must, given its date, still have been in draft form.

Mr Wheeler saw the applicant for a limited period while taking him to and from appointments. It is, we consider, difficult to attach much weight to his evidence as to the applicant’s age, given that he said only that the applicant could have been 17 in September 2016, and when pushed said only that the oldest he could be is “18 plus.”

Of all the witnesses from whom we heard oral evidence, Patrick Shier had spent the most time with the applicant, having been on duty at Appledore Reception Centre for a period of several weeks, interacting with the applicant and observing him on a daily basis. We found him to be a candid witness whose evidence was direct and to the point. Contrary to what is submitted by the applicant, we do not consider that he tried to row back from his statement on 7 October 2015 that he considered the applicant to be 17. His evidence is, in any event, that he “would assess” the applicant as 17. That was, in effect a snapshot at a point when the applicant had just been inducted. As is evident from the material provided, age assessments should be done over a period of time to allow reflection and a holistic assessment of the individual.

We do not consider that Patrick Shier could be said to have crossed the line into being an advocate. The submission that he is impartial, and has attempted a character assassination, has more than a hint of hyperbole. We do not consider, having observed him giving evidence, that his use of two exclamation marks in an email, or any other part of his testimony could rationally form the basis of submission that he is unreliable. His evidence was clear and forthright unlike the evidence of Rose Corby or Melissa Carter.

Accordingly, we consider that we can attach weight to Patrick Shier’s evidence, and we find him a reliable witness of fact, albeit that his description of the applicant’s English being perfect is an exaggeration. We accept, naturally, that he is not expert at age assessment but he has significant experience in working with young people.

In assessing the applicant’s evidence we bear in mind that he may be a minor, and have therefore approached his testimony on the basis that he should be treated as vulnerable person. We note also that he was fasting for Ramadan, and was permitted breaks during his evidence. No additional adjustments were requested by his representatives.

As outlined above, there are a number of apparent discrepancies in the applicant’s evidence. That much is accepted by the applicant’s counsel, but they submit that the inconsistencies do not go to the core of the claim, and that the basic themes in respect of his childhood, his family, the loss of his father, his departure from Afghanistan, and his journey, have remained the same. It is submitted that he maintained his account over lengthy cross-examination, and that it is entirely credible.

Further, it is submitted that his evidence is the most direct evidence of his age available to the Tribunal given that the other witnesses are, estimating his age on other evidence .

Much of the cross-examination of the applicant focussed on what he is recorded as having said during interactions with various social workers outside the formal age assessment process. We bear in mind that what is recorded does not purport to be a verbatim record of questions and answers. Further, the dates on which the reports of the conversations were finalised are in several cases some time after the date on which it is said they took place.

We would not have expected the applicant’s evidence to be completely consistent; in our experience, each retelling of an event is likely to vary given the passage of time, and imperfect recall. Given also that on many occasions interpreters have been used, the danger of over reliance on small discrepancies which may arise from nuances is greater.

That said, we consider that the applicant’s inability when questioned to describe the size of his village, and not knowing even if there was a school in the village, undermines his credibility. Further, he has been inconsistent as to whether there was a school or not. We do not accept that this is a minor discrepancy, or can otherwise be explained. We found the applicant when questioned about both of these issues to be evasive when saying that he had not counted the houses or seen a school, yet this was a village in which he had lived for all of his life until he left Afghanistan. We find that this undermines his reliability.

We find that the applicant has also been inconsistent about whether and when he knew of his age. In the initial assessment, he said that he had not been told his age before leaving Afghanistan, but in evidence before us that from 12 onwards he had been given clothes on the basis of him being 12, 13 or 14. That does not, we consider, constitute birthday celebrations, as submitted by Mr Holbrook. While there is no real indication that it was on the precise anniversary of his birth, the description of what occurred does not sit well with the evidence from the applicant recorded in his interview with Rose Corby and Melissa Carter that birthdays are not celebrated, but it does accord with what is recorded by Mr Kelly, that the applicant had said he had grown up knowing his birthday. We do not accept that this inconsistency is properly explicable, and again, reliability is undermined.

We do not accept that the respondent has accepted that the applicant’s mother told him he was 14 years and 9 months of age. That was simply a question put to him in cross-examination. His evidence on this issue has, however, been inconsistent between whether he simply overheard his mother discussing his age with his uncle, and being told, expressly, that if he was asked by anyone he should say he is 14 years and 9 months old.

Further, we find it inconceivable that he knew the rough ages of his siblings which his mother had discussed with neighbours, yet that his age had not been discussed. This part of the applicant’s testimony appears to have been fabricated to account for him knowing his siblings’ ages, it being initially his evidence that he had not known his age until shortly before he left Afghanistan.

In fairness to the applicant, we consider that no adverse inferences can be drawn from the slight differences in whether he knew his father is dead, and the sequence of events in which his mother was told by a bodyguard, and then told him what had happened. The difference here is effectively between “I believe” and “I know” because of what I have been told. Such subtle distinctions can easily be misinterpreted.

We do, however, find that the applicant has been inconsistent in his evidence of whether he is in contact with his mother. While we bear in mind the dangers of relying on the documents produced by Kent, as noted above, that does not mean that we cannot attach weight to them.

The applicant was provided with an interpreter at the meeting in February 2016 when he told staff that he is in contact with his mother. While it is, we concede, possible that he said he was not in contact, we note equally his evidence on 25 August 2016, again with an interpreter, that she has a mobile phone and we do not accept his explanation for the inconsistency.

The evidence of Mr Kelly is relevant to this issue also. While we accept that no interpreter was present, and that the applicant may well have been tired, it is the nature of the conversation which was recorded shortly thereafter which concerns us as much as what was said. While we note that Mr Kelly was not called to give evidence, we find no reason why he would have recorded the conversation if it had not taken place in the manner stated; he had not reason to do so.

It is as much the fact that the applicant knew about Facebook and wanted to contact a cousin to tell his mother he is ok which is relevant as the fact that he was able to say this in English. We reject the applicant’s explanation, in cross-examination, that he had just said “yes” in response to leading questions. It is wholly implausible that such detailed leading questions about a cousin and Facebook, and a computer, could have been put. Nor do we accept that he had indicated by gestures that he was itchy, rather than saying so, as none of this explains why Mr Kelly would have recorded in some detail what the applicant had said during the journey to Appledore. While Mr Kelly’s evidence was not tested in cross-examination, nonetheless we consider that we can attach weight to it. The account is relatively contemporaneous and there is no reason for him not to have recorded what he was told.

We do not accept the submissions of the applicant that it was possible that much of the information in Mr Kelly’s report arose from suggestions made by him, or that the applicant simply assented to them. Still less is there any merit in the entirely speculative submission that the evidence may have been gleaned from information gleaned at the police station.

Further, we do not accept that because the pathway plan was withdrawn that reliance on the note of 25 August 2016 that preceded it should be disregarded. The challenges to the plan set out in the letter of 27 September 2016, extending to several pages do not in reality challenge what the applicant is recorded as saying; rather, the attack is on the interpretation of what he said and it is alleged that the plan was being used to attack and undermine his character. In any event, the reason given for its withdrawal was that it was incorrectly worded and did not represent the proper care and support being given to the applicant.

We accept that the notes were not read back to the applicant at the time, but that is not a sufficient reason to disregard them. It was open to the applicant to have commented on them in a witness statement as he did with other parts of the respondent’s evidence including that of Ben Holbrook who was not called to give evidence. While the applicant prays in aid Ms Johnson’s evidence that there are “huge problems” with local authority record-keeping and that misinterpretations happen, that is little more than anecdote and does not engage with the specifics of the documents in question.

The evidence of the applicant’s competence in English is inconsistent, but it is of note that both Mr Kelly and Mr Shier record in their initial observations that he spoke English to some extent, if not actually quite well. That is confirmed also by Ben Holbrook in his statement at [25].

In contrast, the evidence from the college he has attended indicate he is not able to pass tests in written English even at a basic level but there is not much indication of his spoken English.

Taking all of these factors together, and viewing the evidence as a whole, we find that, notwithstanding the applicant’s possible vulnerability, and that he may be a child, that he is not a reliable or credible witness. On the contrary, we are satisfied that he has not told us the truth about the circumstances of his village or his family, and that his testimony as to his age is unreliable and not credible.

As noted above, we heard the applicant’s witnesses prior to hearing evidence from Kent’s witnesses.

In assessing the evidence of MY and AK, we find that it adds little to the evidence of the applicant. While we agree it is unlikely that they would gravitate towards men in their early twenties, it is notable that MY describes the applicant as very immature (but not in comparison to what), and comments that the applicant was unhappy with the DOB of 01/01/2000 at [4] but says nothing about what he himself believes or why.

AK says that the applicant looks like a child [3] and behaves like a child, he does not qualify “child” by reference to any age group, given that it could cover up to 18.

While it is submitted that their evidence is that they clearly regard the applicant as the same age as themselves, that is not what is recorded in their witness statements; that is a speculation by the applicant’s counsel. We find that their evidence adds nothing of weight to that of the applicant.

Helen Johnson is Head of Children’s Services at the Refugee Council. She met the applicant at his house, and was of the view that there was nothing in the applicant’s appearance or behaviour that made her doubt his claimed age. She did not agree with Rose Corby’s observations about the applicant, and was of the view that his behaviour was consistent with being a child who is coping badly with his situation and not helping himself.

Helen Johnson accepted in cross-examination that she was not in a position to carry out a “ Merton compliant” assessment. She also accepted that, apart from a short period using an interpreter over the telephone to check some points, her conversation with the applicant had been conducted in English.

We found much of Mr Holbrook’s cross-examination, except as it related to those two issues, to be of little or no assistance.

We cannot attach weight to the photographs adduced by Kent. There is simply no basis for that assertion that they are “typical” of the individuals at the age stated. There is, in fact, nothing to suggest that they were representative as opposed to skewed, and worryingly, as Mr Holbrook accepted, several of the photographs are distorted by being stretched vertically or horizontally. We are, frankly, concerned that such inadequate evidence could have been thought capable of bearing any weight whatsoever.

We were invited, as was Helen Johnson, to consider photographs of the applicant taken some months apart. It is difficult to assess these. There is insufficient evidence that the lighting conditions were the same, and given natural shadowing, we do not consider if it was possible to discern either that there had been changes or no changes. We bear in mind also that the applicant had been unwell and had a kidney removed between the taking of some of these photographs.

For the reasons set out above, we do not consider that we can rely on Professor Roberts’ report, nor do we consider the evidence of the applicant to be reliable or credible. As noted above, we do not find the evidence of MY and AK to assist that of the applicant. Helen Johnson’s evidence is of little or no value in assessing the applicant’s age.

We do not find the evidence of Rose Corby to be reliable, nor that we can rely on the evidence of Melissa Carter for the reasons given above.

The respondent no longer seeks to rely on the initial age assessment, and has not sought to defend it in submissions. That is hardly surprising, given that their case is now that the applicant is many years older than initially assessed.

While the respondent has not sought to rely on the evidence of Ben Holbrook, his evidence is still evidence before us, and which we take into account. His evidence is still that the applicant is most likely aged 19, giving a date of birth in 1998. It is of note that he was part of the initial assessment team, and also has worked with him since January 2017.

The assessment that the applicant is now most likely 19 is consistent also with the evidence of Patrick Shier who did give evidence before us.

Taking all of this into account, we consider on the basis of the evidence of Patrick Shier, as supported by Ben Holbrook that the applicant is 19 years of age. Doing the best we can with the evidence we have, we conclude that the applicant’s date of birth is, on a balance of probabilities, 7 September 1998. He is, as a consequence, at the date of our decision 19 years of age.