Re BKR
[2015] SGCA 26

Case Number	:	Civil Appeal No 27 of 2014
Decision Date	:	19 May 2015
Tribunal/Court	:	Court of Appeal
Coram	:	Sundaresh Menon CJ; Chao Hick Tin JA; Andrew Phang Boon Leong JA
Counsel Name(s)	:	Sarjit Singh Gill SC, Terence Seah Yong Guan and Sarah Liyana binte Yazid (Shook Lin & Bok LLP) for the appellants; Lee Eng Beng SC, Kelvin Poon, Low Poh Ling and Wilson Zhu (Rajah & Tann LLP) for the first and second respondents; Alvin Yeo SC, Monica Chong, Aw Wen Ni and Chan Xiao Wei (WongPartnership LLP) for the third respondent.
Parties	:	Re BKR

19 May 2015

Judgment reserved.

Sundaresh Menon CJ (delivering the judgment of the court):

1       This appeal concerns an application made pursuant to the Mental Capacity Act (Cap 177A, 2010 Rev Ed) (“the MCA”) for a declaration that the third respondent is unable to make decisions as to her property and affairs because of an impairment of or a disturbance in the functioning of her mind or brain, and for a consequential order that deputies be appointed to make all decisions relating to her property and affairs on her behalf. The application was made by the appellants, both of whom are sisters of the third respondent. It is also supported by a number of their siblings and two of the third respondent’s three children – one son and one daughter – though none of these others are named as parties to the proceedings. Opposing the application are the third respondent herself and the first and second respondents, who are wife and husband respectively. The first respondent is the third respondent’s daughter and youngest child.

2       It is common ground that the functioning of the third respondent’s mind is impaired in some way, but there is no accord as to the nature or the degree or extent of that impairment. The appellants take the position that the third respondent has dementia. They say that this has caused (i) significant deterioration in her memory, (ii) a decline in her executive functioning – which refers broadly to the cognitive abilities essential for goal-directed behaviour – and (iii) the development of paranoid and/or false beliefs. They claim that these consequences of her dementia prevent her from being able to make decisions relating to her property and affairs. The respondents, on the other hand, deny that she has dementia; according to them what she has is a condition referred to as Mild Cognitive Impairment (“MCI”). They accept that this has had an effect on her memory but they contend that this is not so pronounced as to have deprived her of the capacity to make decisions.

3       It is useful to note at the outset that this appeal abounds in allegations of undue influence and ulterior motives. It is a disputed matter whether these allegations are even relevant to the appeal. The parties come from a prominent and very wealthy family. It is against this background that the third respondent has repeatedly expressed the belief that these proceedings brought under the MCA are in reality an attempt by her son, abetted by the appellants, to achieve control of her substantial wealth. On the other hand, the appellants and all who support their application maintain that they have no interest in the third respondent’s money, and that it is in fact the first and second respondents who seek to manipulate her for financial gain.

4       The application was heard at first instance by a Senior District Judge in the Family Court (“the SDJ”), who agreed with the appellants and found that the third respondent lacked decision-making capacity. The appellants were duly appointed as the third respondent’s deputies. The SDJ’s written decision was published as [2012] SGDC 489 (“the First Instance GD”). All three respondents appealed and the matter went before the High Court judge below (“the Judge”). In her judgment reported as Re BKR [2013] 4 SLR 1257 (“the Judgment”) the Judge reversed the decision of the SDJ. The appellants now appeal to this court.

5       As this is a long judgement, it may be of assistance to the reader if we signposted it by providing a broad outline of how its contents have been organised:

(a)     We begin with a brief summary of the facts and the background to the case ([6]–[37]);

(b)     We then review the course of the litigation in the Family Court and the High Court ([38]–[51]);

(c)     Next we set out the broad contours of the law on mental capacity ([52]–[57]). This is done essentially to set the context against which the decisions of the two earlier courts may be considered and understood;

(d)     We then consider the two earlier decisions from which this appeal arises ([58]–[75]);

(e)     Then we set out the broad issues that arise for our determination ([76]–[79]);

(f)     We then set out our decision. There are several parts to our decision:

(i)       We first deal with a preliminary objection that the third respondent’s litigation representative in the High Court proceedings was not properly appointed ([80]–[87]);

(ii)       We then deal with the central issue of law which we were presented with, which is the proper approach that should be taken by the court in dealing with applications of this sort and more specifically, to what extent the court may have regard to the actual circumstances in which the subject of the application is placed and where overlapping allegations of undue influence have been raised ([88]–[127]);

(iii)       We then consider the question of whether the third respondent lacks capacity. In this section we examine the medical evidence that was led, as well as the evidence that was given by the third respondent herself and by the other factual witnesses. We also examine the evidence in relation to two specific sets of events or transactions ([128]–[205]);

(iv)       We then set out our findings on the issue of capacity ([206]–[207]) and our conclusions and further directions ([208]–[210]);

(g)     We end with a post-script where we offer some observations as to how matters of this nature may be conducted in a manner that may be preferable to what transpired in this case ([211]–[216]).

Facts and background

6       The litigation commenced when the appellants made an application under the MCA by way of Originating Summons (Family Matters) No 71 of 2011 on 18 February 2011. The critical events leading to the application began about four months earlier, in late October 2010, as a result of behaviour on the part of the third respondent that might fairly be said would cause an objective and reasonable observer to ask whether she might lack decision-making capacity and/or be acting under the influence of the first and second respondents. In particular, this behaviour consisted of (i) setting up a trust, which we shall call “the Trust”, on terms and in circumstances that might be said to be questionable, (ii) giving a succession of contradictory instructions to her bankers in relation to the transfer of her assets that had been held in two banks to a third, and (iii) going to live with the first and second respondents in Hong Kong and then cutting off contact with her other children and her siblings despite their efforts to gain access to her.

7       As to the facts that are relevant to the issues, we are primarily concerned with events that took place in the last months of 2010 and the early part of 2011. But it is first necessary to go back a little so as to locate those events within their proper context. We touch on two points in particular. First, we describe the state of the relationships within the third respondent’s family; then we trace the decline of the third respondent’s memory in the period from 2005 to 2010. This will lead us to the events in the critical period from the end of 2010.

The third respondent’s family

8       As mentioned above, the third respondent has three children, of whom the first respondent, a psychiatrist, is the youngest. Her son, the middle child, is a barrister practising in Hong Kong. We refer to him as “CK” as was done in the Judgment. Her eldest child is a doctor in general practice and we refer to her as “NG”.

9       When the third respondent’s father passed away in 2004, he left her and her various siblings a vast legacy. By July 2010, her share approached $200 million.^{[note: 1]} The third respondent’s late husband was himself well-to-do with a considerable fortune and when he passed away in 2007, the distributable part of his estate was to be divided in the proportions of 30% each to NG and the first respondent, and 20% each to CK and CK’s son.

10     Unfortunately, the relationship between the third respondent’s children could not be described as harmonious. There had long been friction but matters deteriorated after their father passed away. The lines were drawn with CK and NG on one side and the first respondent on the other, but the antagonism was strongest between CK and the first respondent. There were a number of specific disagreements between them as summarised by the SDJ at [101] of the First Instance GD. It is not necessary to go into them in any detail.

11     The animosity between her children vexed the third respondent, and on many occasions following her husband’s death, she wrote letters in which she lamented their inability to get along and urged them to work past their differences. Most of these letters were addressed to the first respondent alone and copied to her siblings, and these contained some form of criticism of the first respondent. It appears that the third respondent would ask CK and the first appellant for assistance in drafting these letters, a circumstance which has led the first respondent to question the degree to which these letters truly reflect her mother’s sentiments.

12     As we have said these family tensions came to a head towards the end of 2010 and culminated in the appellants’ filing the MCA application on 18 February 2011. We will turn to that shortly, but first, we touch on the matter of the third respondent’s memory.

The third respondent’s declining memory from 2005 to 2010

13     It is not controversial that the third respondent’s memory is far from satisfactory. This had been evident from as far back as a decade or so ago: in a report dated 12 October 2005, a clinical psychologist from Hong Kong found that the third respondent had “significant impairment” in memory when tested with immediate recalls and 20- to 30-minute delayed recalls; that her “ability to acquire new materials after repeated presentations was limited”; and that her “ability to recognise what had been presented was also impaired”. Her ability to recall verbal materials on a test known as the California Verbal Learning Test was two to three standard deviations below what was expected of her age group, based on a research sample from the United States. It also appears that in 2006 the third respondent started taking “Prophylactic medication” for “Senile Dementia”.

14     Moreover, there is documentary evidence that various persons who interacted with the third respondent found her memory deficits sufficiently significant to warrant comment from at least 2009. For instance, in a handwritten note dated 10 March 2009, one of the third respondent’s sisters informed NG that the sisters had noticed that the third respondent’s memory was “not so good” and that she “[kept] on repeating a lot of questions”, and that “it might be a good idea to bring her to consult a geriatrician to advise whether some medication may be helpful for her symptoms”.

15     More noteworthy is documentary evidence that reflects concerns about the third respondent’s mental fitness expressed in the middle of 2009 by her bankers at JP Morgan, with whom she had opened accounts in late 2006 or early 2007. In notes attached to an e-mail exchange on 22 June 2009, she was described as having “shown some signs of being forgetful and incoherent lately”, and another internal note, sent with an e-mail on 23 June 2009, suggested that her forgetfulness might mean that she was “unsuitable for [redacted] with us going forward especially on more complex products”. A month later on 24 July 2009 an e-mail recorded the fact that some action had been taken “due to client’s deteriorated capacity”. In another e-mail a few days later, a bank officer wrote that having conversed with her, he had begun to “realise that her memory has deteriorated especially in the last couple of months”. In an e-mail of 28 July 2009 it was written: “Banker/Investor brought the case to our attention in June when they found that client could not remember [redacted] she placed with us”.

16     Furthermore, there was the testimony in court of a former JP Morgan banker who had been her investment advisor at the bank in 2009 and 2010. He testified that, in early 2009, he had received from the third respondent faxed instructions to send all her bank statements to the first respondent in Hong Kong. At the time of these instructions, the third respondent was living in Hong Kong with the first respondent, and when she subsequently returned to Singapore and met in person with this banker, she had no recollection that she had given such an instruction, and was in fact “very upset” that the statements had been sent to her daughter.^{[note: 2]} The banker also recounted how the third respondent would discuss with him the carrying out of a transaction, in particular transactions pertaining to renminbi, and after the transaction had been completed, she would suggest carrying out the same transaction as if it were a fresh idea. The banker added that JP Morgan’s concerns about the third respondent’s capacity eventually reached a point where it decided that it would no longer accept her oral instructions to trade on her account unless another party also reviewed and confirmed those instructions – she designated her son CK for this purpose.^{[note: 3]} The evidence of this banker as to JP Morgan’s decision to cease accepting her oral instructions was corroborated by another JP Morgan banker who also testified in court.^{[note: 4]}

The creation of a trust dated 26 October 2010

17     We turn to the critical events near the end of 2010, and we begin with the Trust. The date that appears on the Trust documents is 26 October 2010 but, as we shall later describe, it appears that the third respondent signed the relevant documents on 6 and/or 26 November. In any event, nothing turns on this. The bank with which the Trust was created was DBS. By way of background, the third respondent had previously in 2007 created two other trusts with two different banks. One was with JP Morgan and broadly speaking it was to provide for herself; the other was with UBS and it was to provide for the education of her eight grandchildren. According to the first appellant, the idea of setting up both these trusts had originated with her: she had previously set up similar trusts for herself and her own grandchildren, and upon her guidance and advice, the third respondent followed suit.^{[note: 5]}

18     The third respondent has said that the Trust has two objectives.^{[note: 6]} The first is to provide for herself during her lifetime, and the second is to make donations to charitable organisations after her death. The Trust names one specific beneficiary, a company which we refer to as “[B] Ltd”, and one class of beneficiaries, being “Charities to be determined”. The trustee of the Trust is a BVI trust company, and it has “absolute discretion” in applying the money in the Trust for the benefit of either of the beneficiaries.^{[note: 7]} In a separate letter signed by the third respondent, the first respondent was appointed as Protector of the Trust. It would appear that she remains the Protector to this day.

19     In a letter of wishes also dated 26 October 2010, the third respondent expressed her wishes that (i) an amount of $5 million be settled into [B] Ltd’s bank account and be replenished periodically, (ii) upon her death, [B] Ltd should have its bank account balance topped up to $10 million and [B] Ltd should then become the sole property of the Protector, and (iii) after her death, the trustee should apply the trust property to charitable causes as recommended by a philanthropic panel to be chaired by the first respondent. Until 27 July 2012, it does not appear that there was any express provision as to what the money in [B] Ltd’s bank account would be used for during the third respondent’s lifetime; it was only on that day that a deed of understanding was entered into between the third respondent, the trustee and the first respondent which provided that [B] Ltd’s monies would be used for the “exclusive purpose of maintaining [the third respondent]”.

20     The Trust stipulates a class of persons called Excluded Persons who are entirely excluded from enjoying any benefits under the Trust. The Trust as it was originally drawn up named two Excluded Persons, namely NG and CK; in addition, any trustee or Protector was an Excluded Person for the duration of their occupation of those offices. The appellants’ complaint is that this aspect of the Trust favours the first respondent over her siblings because she would no longer be an Excluded Person should she relinquish her office of Protector. We were informed by counsel for the third respondent that, although he considered this complaint to be groundless, the wording of the Trust has since been “tighten[ed] up” so as to “make it clear that none of the children can benefit from the assets in the Trust”. But counsel also indicated that the provision of a gift of $10 million to the first respondent on the passing of her mother has been retained.

21     Much of the evidence on the circumstances in which the Trust was set up proceeded from the first respondent. She testified that the third respondent came up with the idea of setting up the Trust after speaking to two individuals. One was a close friend of the third respondent’s late husband whom we call “Mr Z”. Mr Z is a wealthy businessman and he is also chairman of the company that wholly owns the Trust’s corporate trustee. The other was a Singapore lawyer whom we call “Mr L”. Mr Z had previously referred the third respondent and her family to Mr L for legal assistance in relation to an estate matter.^{[note: 8]}

22     According to the first respondent, Mr Z told the third respondent sometime in the second half of 2010 that she should take care of her money lest CK get his hands on it. Acting on that advice, the third respondent personally called Mr L and arranged to meet him so that he could assist her in taking steps to protect her money. Mr L duly came over to Hong Kong and met her. The first respondent testified that, although she had not been present when Mr L took her mother’s instructions, she subsequently met her mother and Mr L, and on that occasion Mr L told her what her mother’s instructions had been, which were that she wanted to set up something so that no one could touch her money, and that she desired to give most of her money away to charity upon her passing. Mr L thus advised the third respondent about setting up the Trust.^{[note: 9]} Eventually, the third respondent signed the Trust documents (as well as related banking documents) in Singapore on two separate occasions in November 2010, in circumstances we shall describe below.

23     The third respondent also gave evidence on the genesis of the Trust and her account was in certain respects at variance with that of the first respondent. In the third respondent’s version of events, Mr Z and Mr L had a fairly limited role in setting up the Trust. She said that no lawyers had been involved in advising her,^{[note: 10]} although she did also say elsewhere in her evidence that she had asked Mr L to “set up a trust for [her]”.^{[note: 11]} As for Mr Z, his participation seems to have been confined to choosing the trustee.^{[note: 12]} The primary actor was instead a senior DBS banker whom we shall call “Ms D”: the third respondent testified that Ms D had come over to Hong Kong and explained the Trust documents to her, and that she had thereafter signed a single document, also in Hong Kong, with no one present except Ms D.^{[note: 13]}

24     There is an important point on which the first and third respondents are in agreement, and that is the first respondent’s lack of involvement in the setting up of the Trust. The first respondent’s position is that she “did not provide any input in the setting up or structuring of” the Trust,^{[note: 14]} and the third respondent denied in cross-examination that the first respondent had made any suggestions in relation to the Trust.^{[note: 15]}

Events from November 2010 to 18 February 2011

25     Following the creation of the Trust, the third respondent gave a series of conflicting instructions to her bankers over a period of more than a month and, in the midst of this, went to live with the first and second respondents in Hong Kong. This move to Hong Kong was not in itself unusual given that prior to this she had periodically spent significant time there. However, what was unusual on this occasion was that she appeared not to want to have any meaningful contact with her siblings and with NG and CK.

26     On 1 November 2010, the third respondent – who was in Hong Kong at this time – signed a limited power of attorney in favour of the first respondent concerning the operation of her UBS accounts. A few days later, on 5 November 2010, the third respondent arrived in Singapore with the first respondent. The first respondent testified that, the following day, she attended a meeting between her mother, Mr L, Ms D, two relationship managers from DBS and representatives of the Trust’s corporate trustee. At this meeting, the third respondent signed some documents relating to the Trust and/or banking arrangements ancillary to the Trust.^{[note: 16]} On 8 November 2010, UBS received a letter in which the third respondent stated that she wished to transfer all her assets that were kept or maintained with UBS to DBS within a week. According to the third respondent, she requested this transfer in order to inject funds into the Trust; it is possible that the letter was one of the documents that she signed at the meeting on 6 November. On 10 November 2010, a female banker from UBS whom we shall call “Ms B” contacted the third respondent. Ms B’s testimony in court was that, during this conversation, the third respondent had no recollection of the 8 November letter and instructed her not to act on the instructions contained there.

27     On 19 November 2010 the third respondent again wrote to UBS. She expressed her unhappiness at the bank’s failure to carry out her instructions in the 8 November letter, and demanded that this be rectified within four days. In response, UBS arranged a face-to-face meeting with her and this took place on 23 November 2010. The first appellant was present at this meeting, as was Ms B. By the time the meeting concluded, the third respondent had signed a number of documents. The first revoked her instructions to transfer all her assets held in UBS to DBS. The second revoked the limited power of attorney granted to the first respondent on 1 November. The third authorised UBS to disclose the third respondent’s account information to four persons, namely, the two appellants, CK and the third respondent’s personal assistant.

28     On 25 November 2010, the first respondent arrived in Singapore. As noted above, she had flown into Singapore with the third respondent on 5 November and following that, she returned to Hong Kong on 7 November.^{[note: 17]} On the next day, 26 November, the first and third respondents had a meeting with at least Mr Z and Ms D, and here the third respondent signed other documents relating to the Trust and/or the associated banking arrangements.^{[note: 18]} On 28 November 2010, the third respondent left Singapore for Hong Kong in the company of the first respondent. The following day, 29 November 2010, a letter signed by the third respondent was faxed to UBS containing fresh instructions to transfer all her assets held in UBS to DBS within a week. In response, UBS wrote back on 2 December 2010 requesting a face-to-face meeting. The third respondent replied on the same day without addressing the bank’s request for a meeting but instead reiterating her instructions in the 29 November letter. Further, this letter of 2 December identified a specific account in DBS into which her UBS assets were to be transferred.

29     Meanwhile, on 29 November 2010, CK wrote to the appellants informing them that the third respondent was in Hong Kong. He asked the second appellant to conduct a land search on a bungalow in Singapore that the first respondent had apparently purchased in her name (“the Bungalow”). This purchase had taken place fairly recently – the option to purchase had been granted on 30 October and it was exercised on 19 November. The second appellant conducted the land search the day after being requested to do so and it turned out that the Bungalow was being purchased in the name of the second respondent rather than the first respondent. The appellants conveyed this to CK.

30     Arising from this discovery concerning the Bungalow, CK and NG decided to visit the home of the first and second respondents in Hong Kong on the morning of the following day, 1 December 2010. They arrived intending to meet the third respondent, who was residing there. However, they were prevented from doing so. The second respondent evidently stood at the front doorway and physically blocked them from entering; at the same time the third respondent remained in an upstairs bathroom with the first respondent. The encounter ended with CK and NG being escorted from the premises by security personnel.

31     Those involved put forward conflicting accounts of what had transpired. The second respondent said that he had obstructed CK and NG because they were being aggressive and unruly; the first respondent said that her mother went to hide in the upstairs bathroom because she was surprised and frightened by the commotion that CK and NG were causing; and on the other hand, CK and NG maintained that they had conducted themselves appropriately and politely and that the second respondent’s refusal to permit them entry was unreasonable. On the next day, solicitors acting for the first and second respondents sent CK a letter warning him not to attempt to enter their residence again.

32     A few days later, on 6 December 2010, CK showed up at an acupuncture clinic in Hong Kong where the third respondent was having treatment while accompanied by the first respondent. There was something of a confrontation there between them; once again the protagonists are not in agreement as to what exactly happened. CK said that his mother was pleased to see him and that he was unable to interact with her only because his sister came between them without good reason; on the other hand, the first respondent said that she was merely trying to stop her brother from harassing their mother into signing legal documents of an undetermined nature. Whatever the case, this did not prevent the trio from going for tea thereafter together with NG and the second respondent. Subsequently, the third respondent wrote by hand a letter dated 8 December 2010 addressed to CK and NG in which she castigated them, and CK in particular, for their attempt to enter the home of the first and second respondents on 1 December.

33     Returning to the third respondent’s dealings with her banks, on 13 December 2010, bankers from JP Morgan met her in Hong Kong at the residence of the first and second respondents. Also present were the Singapore lawyer Mr L and two bankers from DBS. According to a contemporaneous attendance note prepared by the JP Morgan bankers, this meeting had been arranged in order that the JP Morgan bankers might obtain instructions from the third respondent face-to-face. It was also recorded in the attendance note that the third respondent did not speak much at this meeting. Instead it was Mr L who mainly spoke on her behalf. Parts of the note were redacted due to considerations of banking secrecy and for this reason it cannot be unequivocally ascertained exactly what the meeting was about, but it may be inferred that it had to do with the transfer of the third respondent’s assets in JP Morgan to DBS. The first appellant’s evidence was that, on 24 November 2010 – about three weeks before the meeting – a DBS officer had turned up at the third respondent’s apartment with letters for her to sign. These letters were addressed to JP Morgan and contained instructions to transfer her funds in the bank to DBS. The first appellant also testified that the third respondent refused to sign the letters at the time.^{[note: 19]} Given the context, it is probable that this meeting on 13 December 2010 concerned those letters and the instructions therein.

34     Two days later, on 15 December 2010, bankers from UBS, including Ms B, met the third respondent. This was a follow-up to the bank’s request about two weeks earlier for a face-to-face meeting with her in order to clarify her instructions on the transfer of funds to DBS. Once again, Mr L was present with two bankers from DBS. According to Ms B, the third respondent was unable to recall the previous conflicting instructions she had given, but she did personally confirm that she wanted to close her UBS account. Ms B also testified however that, at the conclusion of the meeting, the third respondent asked her more than once to purchase more renminbi products, apparently having forgotten that the entire purpose of the meeting was to close the UBS account.

35     On 17 December 2010, CK met the third respondent while she shopped at a crafts shop. Little was said about this meeting in evidence except that the third respondent testified that she had been “quite happy to see him”.^{[note: 20]} A few days later, solicitors acting for the third respondent wrote identical letters dated 21 December 2010 to CK and NG demanding that they furnish undertakings not to harass their mother. However, the third respondent then had lunch with all her children and grandchildren on Christmas day, 25 December 2010. On this occasion, they did not speak about financial matters or the intra-familial conflicts; in NG’s words, “superficially it was like a happy Christmas lunch”.^{[note: 21]}

36     On 27 December 2010, solicitors for UBS wrote to the third respondent setting out the chronology of events since 1 November. They traced the contradictory and conflicting instructions they had been given as to transferring her UBS-held assets to DBS, and informed her that UBS would need her to receive independent legal advice and to provide medical proof of mental capacity. To that end, on 13 and 20 January 2011, medical reports were submitted by two doctors certifying her capacity to execute statutory declarations.

37     Subsequent to this, between 21 January and 18 February 2011, correspondence flowed between Shook Lin & Bok LLP, counsel for the appellants, and WongPartnership LLP, counsel for the third respondent, the gist of which was that the third respondent declined to return to Singapore and meet her sisters face-to-face as requested. On 18 February 2011, the appellants commenced proceedings under the MCA.

The litigation

Pre-trial interlocutory applications and clinical examinations

38     When the appellants filed their MCA application on 18 February 2011 it was accompanied by a summons in which they sought, among other things, an order that the third respondent attend alone before “a neurologist and a psychiatrist and/or any other reputable independent medical specialists” for the purpose of evaluating her capacity to deal with her property and financial affairs. These medical specialists were to be engaged by the appellants and approved by the court. The appellants were evidently anxious to ensure that any clinical examination of the third respondent be carried out in a neutral way by impartial medical experts.

39     The appellants’ summons was not heard until 22 July 2011; by then the third respondent had sought out a number of medical experts with whom she had had no previous relationship and requested that they examine her. These experts comprised three psychiatrists and two neurologists: the psychiatrists were Dr Calvin Fones (“Dr Fones”), Dr Tan Chue Tin (“Dr Tan”) and Dr Ung Eng Khean (“Dr Ung”), and the neurologists were Dr Tang Kok Foo (“Dr Tang”) and Dr Wong Meng Cheong. Over several days in the first half of March 2011, the five experts took turns to examine the third respondent; on one occasion, three of the doctors conducted a joint examination of the third respondent but the other meetings were private and involved only a single doctor with the third respondent.

40     The examinations of the third respondent consisted of informal interviews in which she was asked questions on such topics as her life history, family, finances, beliefs, and likes and dislikes. In addition, a number of formal standardised mental tests that are widely used by clinicians for the purpose of assessing cognitive function were administered. The tests are meant to be administered according to clearly-defined instructions and use a fairly strict scoring system. The unanimous opinion of all five experts following this round of clinical examinations was that the third respondent remained in possession of her decision-making capacity.

41     Since the appellants’ prayer for an order that the third respondent be examined by independent medical specialists had not yet been adjudicated upon, the appellants had no means of compelling the third respondent to undergo an examination by any medical expert nominated by them. Nonetheless, on 7 July 2011, three experts instructed by the appellants – namely, Professor Kua Ee Hock (“Prof Kua”), a psychiatrist, Dr Chan Kin Ming (“Dr Chan”), a geriatrician, and Dr Benjamin Ong, a neurologist – produced opinions on the third respondent’s mental health. They based their opinions on the factual material asserted in the affidavits that had been filed by the parties to the litigation, as well as on the medical reports put forward by the third respondent’s five experts in March 2011 and also on various other medical reports produced by her Hong Kong doctors prior to the commencement of the MCA proceedings. The appellants’ experts all expressed concerns about the third respondent’s ability to manage her property and affairs but none of them went so far as to say that she lacked capacity to make decisions. Instead the consensus among them was that it was imperative that the third respondent be clinically evaluated by a panel of independent specialists.

42     In early July 2011, the third respondent was again examined over a few days by three of her own medical experts, namely Dr Fones, Dr Tang and Dr Ung, individually. In their affidavits, the experts stated that these examinations were requested by the third respondent in order to ascertain if there had been “any changes in her clinical and mental status” since the earlier examinations. All three experts repeated their earlier conclusion that she was mentally capable of making decisions.

43     A few weeks later, on 22 July 2011, the SDJ heard the appellants’ application for an independent clinical examination of the third respondent. This was heard together with two applications which had been made two months earlier, one by the first and second respondents and the other by the third respondent. Both the latter applications sought an order that all the deponents of affidavits in the proceedings be cross-examined; among these deponents was the third respondent herself. Thus, in effect, she had sought an order that would necessarily expose her to cross-examination.

44     The hearing of the applications spilled over to a second day before the SDJ in August 2011. On 13 September 2011, she gave her decision. She allowed the respondents’ applications for cross-examination but made no order on the appellants’ application for independent medical examination. The appellants were dissatisfied with this and lodged three appeals relating to the SDJ’s decisions in the three applications. The appeals were heard by the High Court on 1 February 2012, less than two weeks before the trial in the Family Court was scheduled to begin. At the conclusion of the hearing, all of the appeals were dismissed. But the High Court judge gave some directions as to how the cross-examination should be conducted, including a direction that the third respondent be cross-examined first.

45     The determinations on these interlocutory matters had two key consequences; first, the third respondent would have to undergo cross-examination in court; and, second, the medical experts instructed by the appellants would not have the opportunity of examining the third respondent in a clinical setting. Thus, their opinions on her mental capacity would have to be based on what they knew of her clinical interviews with the experts instructed by her, as well as their own observations from her examination in court.

Examination of the third respondent in and out of court

46     The trial in the Family Court began with the examination of the third respondent on 13 February 2012. This continued for eight days and ended on 22 February 2012. It consisted almost entirely of cross-examination conducted by counsel for the appellants. Whereas the usual objective of cross-examination is to establish facts by eliciting evidence relevant to those facts, the cross-examination of the third respondent was meant in large part to test her mental functioning. Thus she was asked, for instance, to explain what an affidavit was and what her understanding was of the terms of the Trust. Even though it has not been suggested that counsel for the appellants was anything other than polite and courteous throughout this process, the third respondent generally adopted an attitude of suspicion and hostility towards him. Following cross-examination by counsel for the appellants, counsel for the first and second respondents indicated that he had no questions to ask her, and her own counsel then indicated that he also had no questions to ask her by way of re-examination.

47     On 23 February 2012, the day after the completion of the third respondent’s examination in court, she met four of her experts – Dr Fones, Dr Tan, Dr Ung and Dr Tang – for a clinical interview. No formal cognitive tests were administered; the entirety of the interview was a conversation between her and the four experts, and much of this traversed the ground that had been covered in cross-examination. The appellants complain that, since counsel for the third respondent had chosen not to re-examine her in court, this clinical interview in effect was a “leading re-examination” of the third respondent calculated to mitigate the damage done by her poor performance under cross-examination. But the opposing perspective is that the adversarial and stress-inducing nature of cross-examination does not provide an appropriate environment in which to assess a person’s true mental capacity and that a more accurate measure of this would be the person’s responses given in the kinder and gentler setting of a sympathetic clinician’s office. We will deal with this divergence of views in due course.

Joint conference of the medical experts in court

48     After the examination of the third respondent, the medical experts were examined in joint conference. The third respondent brought six experts, among whom were the three psychiatrists and one neurologist who had conducted the clinical interview of 23 February 2012 with her. On their part, the appellants brought five experts, including Prof Kua, Dr Chan and Dr Michael Yap (“Dr Yap”), a neurologist. During this phase of the trial, the experts did not give evidence individually in the usual way but instead took the stand together and gave evidence concurrently. In this manner, they were able to respond immediately to one another’s views, and questions from counsel and the SDJ could be put to all the experts at once. Each side’s complement of experts included a radiologist and a nuclear medicine physician. They gave evidence on brain scans that the third respondent had undergone prior to the commencement of the MCA proceedings.

49     The medical experts’ conference lasted five days, during which time they discussed (i) the brain scans, (ii) the results of the formal cognitive tests administered to the third respondent by her experts, (iii) excerpts from the transcripts of her clinical interviews with her experts, and (iv) the transcripts of her cross-examination in court by counsel for the appellants. From this mass of material, two quite different conclusions were advanced. As we have mentioned, the third respondent’s experts took the position that although she had MCI, she did not lack the ability to make decisions. The appellants’ experts on the other hand, were of the view that she was suffering from a mixture of Alzheimer’s disease and vascular dementia – which is dementia that is linked to inadequate blood supply to the brain – and that the effect of these conditions was that she did not have decision-making capacity.

Examination of the other witnesses

50     After the joint conference of the medical experts, the appellants and six witnesses aligned with them took turns to give evidence. These six witnesses were the third respondent’s children, CK and NG, and four bankers. Finally the trial concluded with the testimony of the first and second respondents.

51     The SDJ reserved judgment after hearing submissions and gave her decision some months later. Before we turn to her decision (and then to the decision of the Judge against which this appeal has been brought) it would be useful for us first to describe the legal framework governing matters of mental capacity in Singapore as this will provide useful context for understanding the two earlier decisions. At this stage, we limit ourselves to a broad overview of this framework.

The law on mental capacity under the MCA

52     Under s 19(1) of the MCA, the court “may make declarations as to whether a person has or lacks capacity to make a decision specified in the declaration” or “decisions on such matters as are described in the declaration”. That is what we are being asked to do in this appeal. Following the MCA, we refer to the person whose capacity to make decisions is under consideration as “P”; to illustrate, P in the present case is the third respondent.

53     In relation to the capacity to make decisions, there are definitions laid down in the MCA. Under those definitions, P lacks capacity if he is “unable to make a decision for himself” because of an impairment of, or a disturbance in the functioning of, the mind or brain: s 4(1). The MCA then goes on in s 5(1) to explain when P will be considered to be “unable to make a decision for himself”. There are four possible situations: (i) he is unable to understand the information relevant to the decision, (ii) he is unable to retain that information, (iii) he is unable to use or weigh that information as part of the process of making the decision, or (iv) he is unable to communicate his decision by whatever means. The remainder of s 5 elaborates on these four situations, eg, s 5(4) expands on the meaning of “information relevant to [the] decision”.

54     For ease of reference we reproduce the material parts of ss 4 and 5 of the MCA:

Persons who lack capacity

4.—(1) For the purposes of this Act, a person lacks capacity in relation to a matter if at the material time he is unable to make a decision for himself in relation to the matter because of an impairment of, or a disturbance in the functioning of, the mind or brain.

(2)    It does not matter whether the impairment or disturbance is permanent or temporary.

(3)    A lack of capacity cannot be established merely by reference to —

(a)    a person’s age or appearance; or

(b)    a condition of his, or an aspect of his behaviour, which might lead others to make unjustified assumptions about his capacity.

(4)    In proceedings under this Act (other than proceedings for offences under this Act), any question whether a person lacks capacity within the meaning of this Act must be decided on the balance of probabilities.

…

Inability to make decisions

5.—(1) For the purposes of section 4, a person is unable to make a decision for himself if he is unable —

(a)    to understand the information relevant to the decision;

(b)    to retain that information;

(c)    to use or weigh that information as part of the process of making the decision; or

(d)    to communicate his decision (whether by talking, using sign language or any other means).

(2)    A person is not to be regarded as unable to understand the information relevant to a decision if he is able to understand an explanation of it given to him in a way that is appropriate to his circumstances (using simple language, visual aids or any other means).

(3)    The fact that a person is able to retain the information relevant to a decision for a short period only does not prevent him from being regarded as able to make the decision.

(4)    The information relevant to a decision includes information about the reasonably foreseeable consequences of —

(a)    deciding one way or another; or

(b)    failing to make the decision.

55     In the Judgment, the Judge observed (at [70]) that the test for capacity in s 4(1) of the MCA contains a functional as well as a clinical component. The functional component is that P is “unable to make a decision for himself”, and the clinical component is that this inability stems from an impairment of the mind. We would emphasise that there must be a causal connection between the inability to make decisions and the mental impairment – going back to the statutory language, P must be unable to make a decision “because of” his mental impairment.

56     Moving to other material parts of the MCA, s 3 sets out a number of principles that are of overarching importance, the first three of which – contained in subsections (2) to (4) – apply to the question of whether P has or lacks capacity. It will be seen that the starting premise is that a person has capacity. The question for the court is whether the lack of capacity has been established (see s 3(2)). For purposes of the instant appeal, a material principle is that contained in s 3(3), which is that P is not to be treated as unable to make a decision unless all practicable steps to help him to do so have been taken without success. What precisely this entails is a matter that we turn to later in this judgment. Again for ease of reference we reproduce s 3:

The principles

3.—(1) The following principles apply for the purposes of this Act.

(2)    A person must be assumed to have capacity unless it is established that he lacks capacity.

(3)    A person is not to be treated as unable to make a decision unless all practicable steps to help him to do so have been taken without success.

(4)    A person is not to be treated as unable to make a decision merely because he makes an unwise decision.

(5)    An act done, or a decision made, under this Act for or on behalf of a person who lacks capacity must be one, or made, in his best interests.

(6)    Before the act is done, or the decision is made, regard must be had to whether the purpose for which it is needed can be as effectively achieved in a way that is less restrictive of the person’s rights and freedom of action.

57     When P is found to lack capacity, the court may appoint deputies to make decisions on his behalf: s 20(2)(b) of the MCA. The powers of deputies are potentially wide-ranging but the MCA contains provisions that circumscribe them. In particular, when deciding whether it is in P’s best interests to appoint deputies, the court must have regard to the principle that the deputies’ powers should be as limited in scope and duration as is reasonably practicable: s 20(4)(b). Furthermore, s 25(5) mandates that deputies have regard to the best interests of P and s 25(1) provides that deputies do not have power to make a decision on behalf of P if they know or have reasonable grounds for believing that he has capacity in relation to the matter. As to what constitutes P’s best interests, this is explained in s 6.

Decisions below

Decision of the SDJ in the First Instance GD

58     Perhaps the central finding that the SDJ made was that the third respondent was “generally unable to understand or appreciate the relevance or importance” of information in context due to her “poor memory”, and that she had “very poor” short-, mid- and long-term memory: at [30] and [128] of the First Instance GD. The key concern in the SDJ’s view was that the third respondent was not able to retain relevant information long enough for her to be able to use and weigh that information when she had to make a decision.

59     Broadly speaking, the SDJ identified two “main overlapping threads of evidence”. The first was directly connected to the third respondent’s mental capacity and it consisted of the evidence of the medical experts taken together with the third respondent’s performance under cross-examination, among other things. The second bore a more indirect relation to the question of capacity and it concerned the intra-family relationships and events from late 2008 to the first quarter of 2011: at [37]–[38] of the First Instance GD.

60     On the first “main thread” of evidence, the SDJ began by considering the results of the formal cognitive tests administered to the third respondent. She found that the third respondent had poor orientation to time, which is to say she had difficulty supplying the correct day of the week, date, month or year when asked: at [50] of the First Instance GD. The SDJ added that the third respondent had “poor concentration” (at [52]) and “very significant memory problems” (at [53]).

61     Next, the SDJ considered the clinical interviews of the third respondent by her medical experts. The SDJ did not agree with the experts’ view that these interviews showed the third respondent to be a “well-functioning individual”. This was because the interviews had not been “rigorous” and “probing” and did not feature “focused and detailed questions” such as would probe the “rationality or consistency of her answers”; all told, the interviews sometimes resembled a “somewhat deferential social conversation” and did not go beyond merely “accepting [her] statements and her efforts at ‘masking’ her deficiencies by giving generalised or evasive answers”. The SDJ opined that the interviews had not so much been a test of mental capacity as they had been a “platform” for the third respondent to state her views on a variety of subjects “with a view to making submissions on [her] behalf”: at [57]–[58] of the First Instance GD. Moreover, the SDJ found that the evidence of the third respondent’s experts was “less than objective” in that they were “in sympathy with and were trying to help” her “obviate a possible finding by the court that [she] lacked capacity”: at [61].

62     Turning to the third respondent’s performance in court under cross-examination, the SDJ held that it demonstrated a “memory impairment so severe that she cannot retain information for a short while and definitely not long enough to be able to use and weigh the information in context together with other longer term information”: at [68] of the First Instance GD. The third respondent was found to be unable to keep track of lines of questioning; unable to recall documents she had read or things she had been told moments earlier; and to have difficulty reading and understanding documents: at [66]–[67]. This was so despite the fact that, in the SDJ’s view, the third respondent had not been “tense” during her cross-examination, as evidenced by the fact that she had engaged in banter with counsel for the appellants who in his cross-examination of her had been “patient and non-threatening” and had extended due respect, courtesy and consideration to her: at [66].

63     The SDJ then dealt with the evidence of the witnesses of fact. She found the evidence of the four bankers “credible” and considered that some of that evidence tended to show diminished capacity on the third respondent’s part: at [78]–[82] of the First Instance GD. The bankers had testified that she would give instructions verbally and forget about them, sometimes repeating the same instruction moments later as if it were a fresh one. They had also given evidence that was suggestive of her susceptibility to manipulation and in particular manipulation by the first and second respondents. Likewise, the SDJ found the appellants to be “credible witnesses” whose concern over the third respondent’s failing memory and consequent vulnerability was “genuine”: at [86]. She thus found that, prior to the events of November 2010, the appellants had been “trusted sisters and advisers” to the third respondent, but all this changed after she left for Hong Kong on 28 November 2010: at [83]–[84]. The SDJ found CK to have been a “credible witness” as well (at [90]), and although she did not expressly apply that epithet to NG it can be inferred that the SDJ found NG credible too as she was said to be “the most spontaneous in giving evidence”: at [93].

64     By contrast, the SDJ found that the first respondent had been “evasive and less than truthful in her answers” and “selective in her memory”: at [96] of the First Instance GD. As for the second respondent, it would appear that the SDJ was unimpressed by his performance on the witness stand. She highlighted the inherent incredibility of the second respondent’s claim that the third respondent’s memory in 2012 was little different from what it had been when she was working at the legal department of a major bank decades ago: at [98].

65     In dealing with the second “main thread” of evidence, the SDJ examined the terms of the Trust and although she said that it was not her role to rule on the Trust, she put forward some observations suggesting that the way the Trust was structured made it open to abuse. She noted the fact that, shortly after the creation of the Trust, the second respondent had purchased the Bungalow using money from the first respondent. The SDJ found that, given the first respondent’s “active interest and participation” in helping her mother move the assets that she held in JP Morgan and UBS to DBS, it was “not a coincidence” that the attempts to move funds into DBS for the purposes of the Trust had taken place around the time of the purchase of the Bungalow, and that the purchase price of the property was similar to the approximate amount of funds in the third respondent’s DBS accounts. In short, the SDJ appeared to think that the first respondent might have engineered the creation of the Trust as well as the migration of her mother’s funds into DBS from other bank accounts in order that she might get her hands on those monies: at [111] of the First Instance GD.

66     The SDJ proceeded to the conflicting instructions that the third respondent had given to her bankers in November and December 2010, and opined that this showed that the third respondent had been “in a very confused state” and had had “difficulty following and understanding the meetings and discussions with her bankers”: at [119] of the First Instance GD. Finally, the SDJ considered the events that had occurred after the third respondent returned to Hong Kong on 28 November 2010, and her conclusion was that the first respondent had been “intent upon precluding access to [the third respondent] by members of her family and her bankers” so that there would be no interference with the first respondent’s self-interested plans relating to her mother’s property and affairs: at [126].

67     Summing up, the SDJ found that events beginning in the middle of 2009 and continuing to 2011 and 2012 showed that the third respondent had “extremely poor memory”, faced “significant difficulties with orientation as to day and date”, had difficulty with “calculation, focusing and concentrating”, and had an “unrealistic perception of the value of her property and assets”: at [127] of the First Instance GD. The SDJ further found that the third respondent had at the very least “mild dementia, Alzheimer’s Disease combined with vascular dementia with very poor short, mid and long-term memory”, and that due to these “extreme memory deficits” and “consequent lack of ability to understand and appreciate the implications of her actions”, she did not “have the executive functioning to enable her to decide on her property and affairs”. The third respondent, it was found, was “not able to recognise if and when she needs help and if so, what sort of help and to whom she may turn for help”: at [128].

68     The SDJ also found that the third respondent was “vulnerable and susceptible to undue influence and manipulation” (at [129] of the First Instance GD), and that in fact the first respondent had been “influencing and causing [the third respondent] to act in a manner that [was] contrary to [her own] best interests”: at [130]. Finally, the SDJ also found that the third respondent did not have capacity to litigate: at [132]. For all these reasons the SDJ granted the appellants’ application and appointed them as deputies to make “all decisions” on their sister’s behalf “in relation to her property and affairs”.

69     The respondents were dissatisfied with the SDJ’s decision and lodged an appeal accordingly. This appeal came before the Judge and we now turn to summarise the essence of the judgment.

Decision of the Judge in the Judgment

70     The Judge considered that she could dispose of the case simply on a point of jurisdiction. She started from the premise that the court has jurisdiction under the MCA only where the substantial dispute revolves around the mental capacity of an individual or matters connected therewith: at [17] and [19] of the Judgment. It followed that the court would not have jurisdiction if the dispute concerned the vulnerability rather than the mental capacity of the person: at [26]. She said that it was “a matter of discretion for the judge to decide what the substantial dispute is” in a given case: at [28].

71     In the Judge’s view the substantial dispute in the present case was “not so much” about mental capacity as it was about access to the third respondent and “the undue influence which she was allegedly subjected to”: at [30] of the Judgment. Central to this conclusion was the fact that the appellants, CK and NG had all opined at various points in their evidence that if the third respondent were afforded access to the right people to give her proper advice, she would be capable of making decisions: at [31]–[43]. Noting also the emphasis placed by the appellants and those associated with their case on what the SDJ had termed the “second thread” of evidence, which concerned events and relationships that were only indirectly related to the question of capacity, the Judge thought that they were “really concerned about something else rather than the mental capacity of [the third respondent]” (at [60]). On the totality of the evidence, the Judge held that the “substantial dispute” in the case before her “really does not centre on the mental capacity of [the third respondent]”, meaning that the SDJ had “erred in finding that she had the jurisdiction to entertain the appellants’ application”: at [61]. The Judge also set aside the findings of undue influence made by the SDJ on the basis that they were “irrelevant” to the question of mental capacity: at [63].

72     Even though the Judge would have allowed the appeal against the SDJ’s decision on the jurisdiction point alone, she went on to consider whether, on the evidence that was adduced, the third respondent could be said to lack capacity. Noting the distinction between the clinical and functional components of the test for incapacity, the Judge found that there was no doubt that the third respondent did suffer from an impairment of mind: at [96] of the Judgment. As to the extent of that impairment, the Judge did not think that she had to make a definite finding on this but nonetheless ventured some views. She thought that it appeared “less likely” that the third respondent suffered from mild dementia progressing to moderate dementia, as the appellants’ experts said she did (at [106]), and the basis for this view was that, in the third respondent’s clinical interviews with her experts, she had been able to “carry on a normal conversation” and there had been “no coarsening of emotions” in that she had “displayed a range of emotions and views over various issues” – even if she “may hold certain paranoid views”: at [103]–[105].

73     As to the functional component of the test for incapacity, the Judge found that this was not met, ie, that it had not been shown that the third respondent was unable to make decisions: at [108] of the Judgment. In the Judge’s opinion, the SDJ’s fundamental error lay in failing to consider the principle enshrined in s 3(3) of the MCA that a person is not to be treated as unable to make a decision “unless all practicable steps to help him to do so have been taken without success”: at [122]. The Judge thought that it followed from this that the third respondent’s performance under cross-examination was not the best indicator of whether she had or lacked capacity: the MCA provides, in effect, that any assessment of whether a person has or lacks capacity must be made when that person is at her best, but she cannot be at her best in cross-examination since that is characterised by its adversarial nature, “lack of assistance” and “interrogative form of dialogue”: at [131]–[134].

74     Contrariwise, the Judge took the view that there was greater probative value in considering how the third respondent fared in the gentler, more facilitative environment of the clinical interviews with her medical experts. In response to an argument that the clinical interviews swung to the other extreme of being entirely lacking in rigour, the Judge found that the experts had done their part in “probing and examining the internal consistency and logic of [the third respondent’s] answers”: at [125] of the Judgment. Although the experts had at times guided or prompted her, this was not so excessive that “it would render the diagnostic value of the interviews nugatory”: at [127]. The Judge held that the clinical interviews strongly suggested that the third respondent did not lack mental capacity: at [128]. The Judge also found that the third respondent did not lack the ability to reason logically, in that she was clear in wanting to donate her wealth to charity, and that she “retained a sufficient ability of logic to link her motivations with her eventual decision”: at [144]. Finally, the Judge noted that the third respondent had been receiving and could continue to receive assistance to aid her in making her decisions, and that she herself recognised that she required assistance in certain areas: at [145]. The Judge concluded that the third respondent would be able to make decisions if practicable assistance were given: at [148].

75     Given the Judge’s decision thus far, it was not necessary for her to go into the questions of whether the SDJ had made too wide a declaration of incapacity or whether she had erred in appointing the appellants as deputies. Nonetheless she expressed some views on these matters, holding that a declaration of incapacity should be as narrow and issue-specific as possible, and that the SDJ “could have been more specific in her declaration of incapacity with regard to different transactions of different complexities”: at [184] of the Judgment. Finally, the Judge opined that the SDJ might perhaps have given more weight to the third respondent’s wishes in deciding whether to appoint her sisters as deputies, seeing that she would have been strongly opposed to that course of action: at [193]–[195].

Issues arising

76     The single core issue in the present appeal is of course whether the third respondent has or lacks mental capacity in relation to her property and affairs. That encompasses a number of sub-issues. In order to resolve that core issue we must consider the voluminous evidence adduced at the trial in the Family Court.

77     But we should first deal with a more general and fundamental issue: what is the approach that the court should take in MCA proceedings where there is, as here, an interaction between (i) mental impairment on the part of P and (ii) the possibility that P is under the undue influence of other persons? The First Instance GD and the Judgment evince quite different responses. The SDJ considered that the allegations of undue influence did not constitute a bar to her determining the question of whether the third respondent was unable to make decisions because of mental impairment; the Judge, on the other hand, thought that the undue influence element was so dominant that the MCA did not even confer upon her the jurisdiction to hear the matter. The SDJ addressed the issue of the third respondent’s mental capacity by taking in the round all the evidence without drawing a sharp division between evidence relevant to mental impairment and evidence relevant to undue influence; but the Judge was of the view that any findings pertaining to undue influence were of no relevance and ought to be set aside.

78     There is thus a need to select between divergent approaches. In our judgment, this choice boils down to the following dichotomy: in assessing P’s mental capacity, does the court have regard to the actual circumstances that P is in, or does it adopt a more theoretical analysis that overlooks those circumstances? The answer to this enquiry will supply, in turn, the answers to the two questions directly raised by the Judgment – first, whether the Judge was correct to hold that she had no jurisdiction to hear the matter; and second, whether she was correct to set aside the SDJ’s findings on undue influence.

79     Before we turn to these substantive issues, there is a preliminary issue that we must consider on account of it having been raised by the appellants. This concerns the propriety of the Judge’s appointment of a litigation representative for the third respondent in the appeal before her.

Our decision

Preliminary issue: litigation representative

80     The appellants put forward a threshold objection that the appeal before the Judge was not properly constituted because her appointment of a litigation representative for the third respondent was improper. They begin from the premise that a proper litigation representative must be one who has no interests adverse to those of the third respondent, and who independently considers whether it is in her best interests to pursue the litigation. The representative who was appointed was Mr Z, the close friend of the third respondent’s late husband we have mentioned at [21] above, and the appellants argue that he did not fulfil these criteria.

81     The obligation to appoint a litigation representative in proceedings under the MCA is found in O 99 r 8 of the Rules of Court (Cap 322, R 5, 2006 Rev Ed). Under r 8(1), so long as P is made a party to proceedings under the MCA, he shall have a litigation representative for those proceedings. Where a deputy has been appointed for P, the deputy will ordinarily act as P’s litigation representative (r 8(3)), but that is not permitted if the deputy is himself a party to the proceedings in his own capacity (r 8(4)). If this preclusion applies to the deputy, it then falls on the court to appoint as litigation representative a person whom the court is satisfied is competent and willing to conduct proceedings on P’s behalf, and who has no interests adverse to those of P: r 8(5)(a).

82     But this requirement of appointing a litigation representative for P admits of an exception stipulated in r 8(2). This exception applies where the court is of the opinion that P does not lack capacity to conduct the proceedings himself. For convenience, we reproduce here the relevant provisions from the Rules of Court:

Litigation representative for P (O. 99, r. 8)

8.—(1) Subject to paragraph (2), if P is or is made a party to any proceedings under the Act, P shall have a litigation representative for those proceedings.

(2)    The Court may, on its own motion or on the application of any person (including P), permit P to conduct any proceedings under the Act without a litigation representative, if the Court is of the opinion that P does not lack capacity to conduct those proceedings himself (regardless of whether P lacks capacity in relation to the matter or matters to which the proceedings relate).

(3)    Subject to paragraph (4), if there is a deputy appointed or deemed to be appointed by the Court under the Act, or a donee under a lasting power of attorney registered under the Act, with power in relation to P for the purposes of the Act, who is given power to conduct legal proceedings in P’s name or on P’s behalf, the deputy or donee (as the case may be) shall be a litigation representative of P.

(4)    Paragraph (3) shall not apply in relation to any proceedings under the Act in which the deputy or donee (as the case may be) is a party in his own capacity.

(5)    If there is no deputy or donee referred to in paragraph (3), or if paragraph (4) applies, the Court shall appoint as a litigation representative for P —

(a)    any person who applies to be the litigation representative, and whom the Court is satisfied —

(i)    is competent and willing to conduct proceedings on behalf of P; and

(ii)   has no interests adverse to those of P; …

…

83     The appellants applied to strike out the third respondent’s appeal against the SDJ’s decision on the basis that she lacked the capacity to litigate. In response, the third respondent applied before the Judge for either one of two alternative orders: either she be permitted to proceed with her intended appeal against the SDJ’s decision without a litigation representative, or Mr Z be appointed as her litigation representative for purposes of the appeal. The Judge decided to make the latter order and so the appeal proceeded with Mr Z as the third respondent’s representative.

84     The appellants now say that it was wrong for the Judge to have appointed Mr Z as the third respondent’s litigation representative. Two reasons are advanced. The first is that Mr Z allegedly has financial interests that are adverse to those of the third respondent because he played an advisory role of sorts in setting up the Trust; and also because the Trust’s corporate trustee is a wholly-owned subsidiary of a company of which he is the chairman. The second reason is that there was a failure to consider whether it was in the third respondent’s best interests to pursue the appeal before the Judge, in that the Judge simply ordered that Mr Z “adopt [the appeal] in its entirety”.

85     There is no doubt that the appellants are correct to say that the appointment of a litigation representative is a matter to be taken seriously and that it is not merely a matter of fulfilling some formal or technical requirement. We also do not think it controversial that the interests of a litigation representative must not conflict with those of P. We consider that these are matters that ought generally to receive explicit attention at the commencement of any proceedings under the MCA, whether at first instance or on appeal. If the court concludes that no litigation representative is required by reason of O 99 r 8(2), it should say so expressly; and if it considers that one is required, then it should indicate that it is satisfied that the representative has no interests adverse to those of P, and that the representative has seriously considered whether litigation would be in P’s best interests.

86     But it is also true that any objection to the propriety of a litigation representative’s appointment will often be an issue that is largely of academic interest. This follows from the fact that the only party that would take such an objection is the party that holds that P lacks capacity. If P is found in fact to have capacity, he would not have needed a litigation representative in the first place and it would be illogical to say that a finding of capacity should be set aside on the basis that a litigation representative was not properly appointed. On the other hand, if P is found not to have capacity, there would be no need to deal with the objection.

87     The instant case presents an illustration of this. If we find that the third respondent lacks capacity, there is no need to rule on the appellants’ threshold objection. On the other hand, if we find that she does not lack capacity, it would follow that she never needed a litigation representative at any time, and it would be against all reason and common sense to decline to hold that she has capacity on the ground only that a litigation representative was improperly appointed at some earlier time. For this reason, we dismiss the appellants’ threshold objection. There might well be exceptional circumstances where it is suggested that the conduct of the proceedings is in fact being affected to P’s detriment because of the unsuitability of the litigation representative in question. But this is not such a case. We therefore do not need to and hence do not in fact consider the substantive merits of the objection, ie, whether Mr Z was in fact in a position of conflict of interest and whether there was indeed no serious attention given to the third respondent’s best interests.

Whether an assessment of P’s mental capacity takes into account P’s actual circumstances

88     With that, we turn to the question of whether the court ought, in MCA proceedings where there is interaction between mental impairment and undue influence, to take into account P’s actual circumstances or to adopt a more theoretical analysis that disregards those circumstances. In truth, this confluence of mental impairment and undue influence is not all that unusual; there are a number of English cases in which such a confluence features. Since our MCA was based substantially on the UK Mental Capacity Act 2005 (c 9) (“the UK MCA”), we look to these English cases for guidance.

English cases that have taken into account P’s actual circumstances

89     Two first instance decisions from the Court of Protection, the English court that deals with matters of mental capacity, suggest that the court should have regard to P’s actual circumstances in examining the issue of P’s mental capacity. The cases in question are In re A (Capacity: Refusal of Contraception) [2011] Fam 61 (“Re A”) and The London Borough of Redbridge v G and others [2014] EWHC 485 (COP) (“Redbridge”).

90     In Re A, P was Mrs A, a 29-year-old woman with a full scale IQ of 53 such that her general cognitive ability was in the “extremely low range of intellectual functioning”; it was not disputed by Mrs A’s litigation friend that she suffered from “an impairment of or a disturbance in the functioning of the mind or brain” within the meaning of the UK MCA (at [4]). She was married to Mr A, who was himself saddled with a “significant impairment of intellectual ability and limited literacy skills” (at [40]). The question before Bodey J was whether Mrs A lacked capacity to decide whether to use contraception. There was evidence to show that Mr A wanted a child and that he exerted control over Mrs A in many areas of her life including by preventing social workers from seeing her.

91     Bodey J found at [66] that Mrs A could “understand sufficient” about the medical aspects of contraception, which amounted to a finding that she could understand the information relevant to a decision as to contraception. But that was not the end of the enquiry: Bodey J continued at [66] to say that Mrs A had also to have “the ability to use or weigh that information”, and in that connection the question was “whether the influence of Mr A over Mrs A has been so overpowering as to leave her unable to weigh up the information and take a decision of her free will”. It is apparent that Bodey J thought that he had to take into account the actual circumstances in which Mrs A lived.

92     Having reviewed the evidence briefly, Bodey J stated his conclusions at [73]. He held that Mrs A’s decision not to continue taking contraception was “not the product of her own free will”. He went on to say that Mrs A “was unable to weigh up the pros and cons of contraception because of the coercive pressure under which she has been placed both intentionally and unconsciously by Mr A”. Such coercive pressure was the product of a number of factors, including the learning disabilities of both Mr and Mrs A, Mrs A’s dependence on Mr A and fear of rejection, Mrs A’s suggestibility and wish to please her husband, and Mr A’s own wish to start a family. “For these reasons”, which include reasons that related to Mrs A’s actual situation in life, Bodey J was in no doubt that Mrs A “lack[ed] capacity to take a decision for herself about contraception”.

93     It was suggested in argument that Re A was treated by at least one subsequent English authority as a case not on mental capacity under the MCA but on the court’s inherent jurisdiction to protect a wider class of “vulnerable” adults. We will return later to explain in greater detail this distinction between, on the one hand, persons who lack capacity because of a mental impairment and come within the protective ambit of the MCA, and on the other hand, persons who, being “vulnerable” for reasons other than mental impairment, fall outside the scope of the MCA and receive protection under the court’s inherent jurisdiction instead. For now the pertinent point is simply that the English Court of Appeal in In re L (Vulnerable Adults: Court’s Jurisdiction) [2013] Fam 1 (“Re L”), at [24], listed Re A as one of a number of cases that illustrated “the deployment of the inherent jurisdiction in different factual circumstances”. It was submitted that what the court had done in effect was to re-classify Re A as a case under the inherent jurisdiction rather than a case under the MCA, and that this amounted to disapproval of the approach taken by Bodey J in Re A in relation to the issue of mental capacity. But in our judgment, there was no re-classification of Re A, still less a disapproval of it.

94     The Court of Appeal in Re L noted that the first instance judge in that matter, Theis J, had considered a number of authorities on the court’s inherent jurisdiction over “vulnerable” individuals of which Re A was one. But this alone cannot mean that Re A is accordingly to be treated as a decision that may be defensible as an exercise of the court’s inherent jurisdiction but not of its jurisdiction under the relevant statute. In fact, nowhere does the Court of Appeal suggest this. And if regard is had to the first instance decision of Theis J, A Local Authority v DL and others [2011] EWHC 1022 (Fam), it will be evident that the judge at [31] regarded Re A as a case on mental capacity under the MCA: she noted without adverse comment that Bodey J had concluded in Re A that “Mrs A lacked capacity”, and in the passage from Re A that she quoted, also without adverse comment, Bodey J had referred to Mrs A as an “incapacitated adult” in contradistinction to “a capacitated but vulnerable adult”. Since the proceedings before Theis J concerned “vulnerable adults who do not fall within the MCA” (at [7]), her focus was naturally on the inherent jurisdiction and not the UK MCA. In this context, the emphasis that she placed on the brief remarks on the inherent jurisdiction in Re A can hardly be considered to amount to a more far-reaching opinion that Re A ought properly to have been determined on the footing of the inherent jurisdiction rather than the MCA. Hence there is nothing in Re L, either at first instance or in the Court of Appeal, which suggests that either court in that case thought that Bodey J in Re A ought not to have come to the conclusion that Mrs A lacked capacity or that his analysis was otherwise erroneous.

95     Turning to the second case on this point, namely, Redbridge, P in that case was one G, a 94-year-old lady. Living with her in her house were a woman, C, and C’s husband, F. Prior to being introduced by a friend at their church, C and F had been “total strangers” to G, and following their initial acquaintance, both had “[f]airly swiftly” moved into G’s house: at [35]. There was evidence that C at least was controlling G by placing restrictions on what she could do in her own house and preventing social workers from seeing her; C would shout at G and threaten to tell others that G had Alzheimer’s disease so that she would be taken from her home and placed in a care facility, a prospect that G greatly feared. There was evidence also of C attempting to secure financial gain at G’s expense: at [27]. At issue was G’s capacity to make decisions in a number of areas, including the people who would live with her and on financial matters generally. The proceedings before Russell J involved two concurrent applications, one under the UK MCA and the other under the court’s inherent jurisdiction to protect “vulnerable” persons.

96     A medical professional, Dr Barker, interviewed G and opined that she had dementia, “failing on a short memory task and on executive functioning tasks”: at [54]. Russell J also heard evidence from an independent social worker, Mr Gillman-Smith, who had been instructed to carry out an assessment of capacity. Mr Gillman-Smith apparently took the view that G lacked capacity “solely due to being unduly influenced by C”; Russell J did not share that view and said that, although she was certain “that the dependency and the influence inhibit G’s ability to reach decisions and undermines her capacity to do so”, she nonetheless preferred the evidence of Dr Barker, which was that G suffered from an impairment of the mind as demonstrated by “confusion and short term memory loss”: at [60] and [62]. In the realm of financial matters, that impairment “affected her ability to retain information relevant to the decisions she has to make” and resulted in “difficulty in understanding the necessary information and to use and weigh the information” (at [80]); Russell J added that C’s influence would make decision-making “even more difficult” for G: at [79].

97     As for the decisions G had to make in relation to the persons who would live with and have contact with her, Russell J held that while G was able to understand that “C and F [had] taken control of her finances”, she was “unable to use the information to make the decision about her own welfare and care and allow[ed] them to remain in her home”. She added that the “controlling behaviour of C and F … will have further compromised the ability of G to make decisions and understand what is happening to her”: at [81]. Russell J thus concluded at [82] that G lacked capacity under the UK MCA, and having so decided, she considered it unnecessary to rule on the question of inherent jurisdiction.

98     It is apparent that in examining P’s mental capacity, the courts in Re A and Redbridge have had regard to their actual circumstances, in particular whether other persons were exerting pressure on P such as would make it more difficult for P to make the decisions in question. It was not even considered as a possibility that the court should divorce P from her actual circumstances and apply a theoretical analysis assuming P’s emancipation from all external pressure and influence.

Arguments in favour of a more theoretical approach

99     But as against this there are three, possibly inter-related, strands of argument for the proposition that the court should apply a theoretical analysis that assumes that P is getting the best appropriate assistance, even if this is not in fact the case. The first strand relies on s 3(3) of the MCA, which provides that P “is not to be treated as unable to make a decision unless all practicable steps to help him to do so have been taken without success”. It was argued that this provision postulates an assessment of capacity based on an ideal or theoretical set of circumstances in which all such practicable steps are being taken, and it was said that such an approach is supported by the first instance decision of the Court of Protection in Wandsworth Clinical Commissioning Group v IA and another [2014] EWHC 990 (COP) (“Wandsworth”).

100    The second strand depends on the legislative history of the UK MCA, as described by the English Court of Appeal in Re L. What this discloses is that a conscious decision was taken not to extend the UK MCA to adults who are “vulnerable” to undue pressure, which means that the MCA was not meant to address cases of undue influence. It follows, it might be argued, that allegations of undue influence – and, more generally, P’s actual circumstances – are not relevant to an assessment of mental capacity.

101    The third strand has to do with the causative nexus required to be shown between P’s mental impairment and his inability to make decisions. This causal requirement is embedded in s 4(1) of the MCA, in the operative words “because of”: P will be declared to lack capacity under the MCA only where he is unable to make decisions because of a mental impairment. In this regard it might be contended that the English Court of Appeal case of York City Council v C and another [2014] 2 WLR 1 (“York CC”) establishes the need for a very strong nexus so that, if P’s inability to make a decision is the result not only of his mental impairment but also his actual circumstances in which he comes under undue influence and pressure, that situation would fall outside the MCA. We consider each of these strands of argument.

(i)   Section 3(3) of the MCA: “practicable steps”

102    The first strand of argument in favour of a theoretical analysis, which relies on s 3(3) of the MCA, is in our opinion undermined by a passage in Redbridge. In that case, Russell J touched briefly on this provision and said only that it was “more difficult to assess as any steps taken by the local authority or others have been hampered by the actions of C and, to a lesser extent, F” (at [76]). She noted that the local authority had on many occasions “attempted to intervene on G’s behalf and tried to assist her without success”. From the evidence, it was apparent that the control exerted over G’s life by C was an impediment to G’s getting assistance in making decisions, and the way Russell J expressed herself was that this rendered it “more difficult” to assess the requirement in s 3(3) that all practicable steps must have been taken to assist G without success. The precise meaning of this is perhaps a little obscure but what is clear is that Russell J thought that she had to consider G’s actual circumstances. She certainly did not think that s 3(3) called for a theoretical analysis that assumed that G was free from C’s control and also able to obtain decision-making assistance.

103    The decision of Wandsworth was cited to us as standing in support of a theoretical analysis but we do not find any such support in the judgment. In that case, P was a 59-year-old man called IA who had been the victim of a violent criminal assault that inflicted a brain injury upon him and this resulted in cognitive impairment. IA suffered from diabetes and consequently required regular medical care. The issue before Cobb J was whether IA was able to make decisions about his ongoing medical treatment, future residence and care, and management of his property and affairs. Medical professionals sought to conduct assessments of his capacity to make these decisions but IA generally refused to cooperate with them although he did make exceptions and he cooperated fully with one Dr Bashir.

104    Cobb J noted at [68] that “practical steps” had been taken to assist IA to reach the decisions in question, including “careful explanation” by Dr Bashir, and on that basis he opined that “when ‘practicable steps’ are taken to assist IA to make a decision, he is able to do so”: at [86]. For this and other reasons which need not concern us here, Cobb J was of the view that IA was not unable to make the relevant decisions. If anything, Wandsworth demonstrates the court’s concern for P’s actual circumstances. Cobb J looked at the assistance that IA had in fact been getting, and at [68] he expressed his “trust that such assistance will be available to him in the future”. Thus to Cobb J, the actual availability of assistance in the circumstances of IA’s life was an entirely relevant matter.

105    Moreover, there is the plain language of s 3(3) of the MCA: it speaks of “practicable” steps to help P. It directs us to look not at fanciful possibilities but at sensible ones. Hence, if P needs extremes of assistance which he could not realistically expect to receive in order to be able to make decisions, it would not be right to say that he possesses the ability to make decisions. By the same token, if in P’s actual circumstances there exists some positive impediment to his receiving assistance, it cannot be said that P has capacity just because he might theoretically be able to make decisions in some other imaginary set of circumstances in which that assistance might be forthcoming.

106    We are led to conclude that s 3(3) of the MCA does not suggest that a theoretical analysis is to be applied which assumes that P is able to obtain and is in fact obtaining assistance in making decisions, however removed such an assumption may be from P’s actual circumstances. We accept that in some situations, s 3(3) may oblige the court to look beyond P’s actual circumstances. But on no basis can it said that, when P’s actual circumstances are such that there is little realistic prospect of him getting the assistance he needs, the court is at liberty or is obliged to disregard those actual circumstances by reason of s 3(3).

(ii)   Legislative history of the MCA

107    The second strand of argument in favour of a theoretical analysis depends on the legislative history of the UK MCA. That history was traced by the Court of Appeal in Re L at [37]–[38], and the essential point is that the joint committee of the UK Houses of Parliament seems to have made a considered decision to exclude from the UK MCA “the lack of capacity to make a free choice as a result of undue influence (or unacceptable pressure)”. This meant that persons “who have high levels of vulnerability, who may be totally dependent on an individual for care, and of course are open to undue pressure being placed on them” would fall outside the protective scope of the statute. All this led the court in Re L to remark, at [58], that nothing in the UK MCA “makes express provision with respect to individuals who may lack capacity for a reason other than an impairment of, or disturbance in the functioning of, the mind or brain”.

108    The respondents rely on the legislative history of the MCA and the Court of Appeal’s comments in Re L to mount an argument as to the court’s jurisdiction: they contend that the court has no power to enter upon a dispute where what is being alleged is that P is unable to make decisions by reason of undue influence or unacceptable pressure, as opposed to an impairment of the mind. A related argument (though not one put forward in these terms by the respondents) is that, since the MCA was not designed to deal with cases concerning undue influence or pressure, when the court assesses mental capacity, it ought to disregard such allegations and, in effect, apply a theoretical analysis assuming that P is free from such influence or pressure.

109    In our judgment, such an argument would be wholly misconceived. At best, what can be extracted from the legislative history of the UK MCA and the court’s comments in Re L is the tautologous and uncontroversial proposition that the MCA was not designed to deal with cases in which P is “vulnerable” but not at all as a result of mental impairment. An example of such a case would be a wife of ordinary mind who is vulnerable to her husband’s influence by reason of her emotional and economic dependence on him. Such a case would fall outside the MCA, simply because it would not come within the terms of s 4(1) which requires a functional inability that arises because of a mental impairment (see [55] above). But this says nothing about what the court’s approach should be when P’s decision-making ability is affected by a combination of factors which include mental impairment. There is no indication in the legislative history of the UK MCA that such a case should be excluded from the operation of the statute.

110    Furthermore, there may be situations in which the whole reason that P is susceptible to undue influence is that he is labouring under an impairment of the mind – for instance, when P’s poor memory permits another person to plant falsehoods in his mind, or when the effect of the impairment is that P is unable even to conceive of the possibility that another person may be manipulating him. In these situations, it is inconceivable in our judgment that the court is to disregard P’s actual circumstances.

(iii)   Requirement of causative nexus between mental impairment and inability to make decisions

111    The third and final strand of argument concerns the strength of the causative nexus that s 4(1) of the MCA requires to be shown between the mental impairment on one hand and the inability to make decisions on the other. The key authority in this regard is York CC, where P was a 48-year-old woman called PC with mild learning difficulties and an IQ between 66 and 69. She was married to NC, a man who had been convicted of serious sexual offences; in fact, she married him while he was serving his term of imprisonment in respect of those convictions. The issue was whether PC had capacity to decide to cohabit with NC upon his release. A significant point in this case was that it was undisputed that PC had capacity to marry.

112    At first instance, Hedley J held that PC lacked capacity. As the Court of Appeal noted at [57] of its decision, Hedley J, having found that PC had a mental impairment and that she was unable to understand the potential risk which NC presented to her, said he was satisfied that the inability to understand “significantly relates to” her mental impairment. On that basis, Hedley J held that PC lacked capacity to cohabit with NC under the MCA. By necessary implication, Hedley J considered that the requirement of a causal link between mental impairment and inability to decide was met so long as the one “significantly relates to” the other. The Court of Appeal took issue with this. It said at [58] that the danger of Hedley J’s approach was that it “watered down” the “strength of the causative nexus” that was needed; in other words, it cautioned against replacing the statutory words “because of” with such formulations as “referable to” or “significantly relates to”.

113    Thus on the authority of York CC, it was argued that a strong causative nexus is required between the mental impairment and the inability to make decisions. In the course of the oral argument before us, counsel for the third respondent submitted that York CC meant that the mental impairment must be the effective cause of the inability to make decisions. It followed, he said, that the causal requirement was not met if the inability to make decisions was also caused by factors other than mental impairment, such as undue influence. The consequence of the argument is that the court must find that P’s inability to make decisions is inherent to the mental impairment and wholly divorced from P’s actual circumstances. If that is accepted, it would seem to follow that York CC suggests the application of a theoretical analysis that divorces P from those actual circumstances.

114    We do not think that is how York CC should be read. The specific context of that case is important: it troubled the English Court of Appeal that Hedley J had found PC to lack capacity in relation to co-habitation when it was agreed that she had capacity in relation to all other matters, including marriage. The question that naturally arose and which the court posed (at [60]) was: if PC’s mental impairment was thought not to be sufficient to “rob her of capacity in all other fields”, what was the evidence that it was enough to cause incapacity in a specific realm? Given two different decisions – the decision to marry and the decision to cohabit – that involved understanding and weighing up similar information, and that were made in the same circumstances and in the shadow of the same mental impairment, the court thought that Hedley J had not adequately explained how PC might be able to make one decision but not the other, let alone how PC’s mental impairment might permit her to make one decision but not the other. Thus, the court’s stress on the “strength of the causative nexus required” was not meant to prescribe any particular approach in a situation where the evidence indicates that P’s inability to make a decision is a result of multiple causes of which P’s mental impairment is one.

115    The court in York CC emphasised the words “because of” in the MCA. In our view, those words most naturally suggest nothing more stringent than a “but for” connection. More crucially, those words do not suggest that there can be no other cause of P’s inability to make decisions besides mental impairment; we do not think that those words indicate that the MCA was intended to exclude situations in which the inability to decide was caused by both mental impairment and P’s actual circumstances. When counsel for the third respondent submits that P’s mental impairment must be the effective cause of his inability to make decisions, we have no quarrel with the use of the word “effective”, because all that says is that there must be a real and not merely illusory cause-and-effect relationship; what we do have some difficulty with is the use of the word “the”, which suggests, erroneously in our judgment, that the impairment must be the only effective cause.

116    In our judgment, York CC is consistent with the notion that P’s inability to make decisions may be the product of a number of effective causes and that the MCA will apply so long as one of those causes is P’s mental impairment. That indeed was the approach of the Court of Protection in Re A and Redbridge: in those cases P’s ability to make decisions was significantly diminished by her actual circumstances, but the MCA was nonetheless held to apply because the inability to decide was also a result of the mental impairment in question.

The need to take into account P’s actual circumstances

117    In our judgment, York CC in fact underscores the importance of an analytical approach that does have regard to P’s actual circumstances. To recapitulate, the issue there was whether PC had the ability to decide whether to cohabit with her husband NC. One question that arose was as to the kind of understanding that she was required to have in order to be considered to have capacity. Did she merely have to understand the nature of the act of cohabitation and its attendant consequences and risks, or did she also have to understand the nature and consequences and risks of cohabiting with the specific individual that was NC, an individual who had been convicted of serious sexual offences? This question was cast in terms of whether the test of capacity to make decisions as to cohabitation was “act-specific” or “person-specific”. The court held that it was not categorically one or the other. The statutory test was “decision-specific”, and it might be “act-specific” in some cases but “person-specific” in others. A pertinent observation made by the court in this context was that “removing the specific factual context from some decisions leaves nothing for the evaluation of capacity to bite upon”: at [35].

118    The importance of that observation in York CC is that when P makes decisions in relation to other people, such as a decision to give away property to person X, it surely cannot be argued that P has capacity so long as she can understand the nature and consequences of giving away property to some theoretical or hypothetical person. On the contrary, part of the package of information relevant to the decision, which P must be able to retain, understand and use, is information about X and in particular whether X is the person to whom P wishes to make the gift. Should P be unable to retain, understand or use information relevant to that decision because of a mental impairment, P will be found to lack capacity under the MCA.

119    Finally, there is a single passage in the judgment of Baroness Hale of Richmond in a decision of the UK House of Lords which did not directly involve the MCA. In Regina v Cooper (Gary Anthony) [2009] 1 WLR 1786, at [13] of the judgment, Baroness Hale commented that the Law Commission report which informed the UK MCA envisaged that the statute would cover those who could understand the nature and effects of a decision to be made but who were prevented by mental disability from using that information in the decision-making process. One of the examples given by the Law Commission was a person whose mental disability “meant that he or she was ‘unable to exert their will against some stronger person who wishes to influence their decisions…’”. Thus it was recognised that mental impairment may in some instances affect decision-making ability only in conjunction with P’s actual circumstances. In such a situation, it would not be at all sensible to disregard those circumstances and in that sense, this too militates against a purely theoretical analysis.

120    For these reasons, we are satisfied that the court must take into account P’s circumstances in assessing his mental capacity. That is what the English cases do, and in this regard, we consider that theirs is a path that we also must take.

The correct approach in MCA proceedings where mental impairment interacts with undue influence

121    What we have said so far on the dichotomy between taking into account P’s actual circumstances and adopting a theoretical analysis does not precisely or directly address the reasoning of the Judge in her Judgment. We turn now to relate what we have said to (i) the Judge’s holding that she did not have jurisdiction to hear the matter, and (ii) her setting aside of the findings of undue influence made by the SDJ.

The court’s jurisdiction

122    The Judge took the view that she had no jurisdiction to hear the matter because the “substantial dispute” was not as to the mental capacity of the third respondent. Instead, she considered that the “substantial dispute” concerned undue influence and access to the third respondent.

123    There is no doubt that in bringing their application under the MCA, the appellants were motivated at least in part by concern over their inability to access their sister and the possibility of undue influence having been exercised over her. Indeed, it would not be inaccurate to say that a substantial part of the trial before the SDJ related to matters of access and undue influence. But it is equally clear that the third respondent has a mental impairment of some sort, and that the case mounted was that given this mental impairment and having regard to the circumstances she has been placed in, she was deprived of the ability to make her own decisions in relation to her property and affairs. In other words, there is a genuine and bona fide dispute as to the real effects of the third respondent’s mental capacity, and this is not any less so just because the appellants and those aligned with them may have expressed the belief at various points under cross-examination that the third respondent would, under certain conditions, be capable of understanding information and forming her own views.

124    In these circumstances, we cannot endorse the Judge’s holding that she lacked jurisdiction to hear the matter. There is and was, in our judgment, a dispute as to the third respondent’s mental capacity, and where this is so the court does have jurisdiction under the MCA to determine the dispute. We do not think that the “substantial dispute” formulation that the Judge employed is the correct test to determine jurisdiction. Parties are entitled to demonstrate the incapacity by reference to the actual circumstances in which P is placed. It is irrelevant that the surrounding material might turn out to be quite extensive. It is only where there is no material question of any mental impairment causing the alleged mental incapacity that a court ought properly to find that it has no jurisdiction under the MCA.

Relevance of allegations of undue influence

125    For the same reason, we also think, with respect, that the Judge was wrong to set aside the findings of undue influence that the SDJ had made. She considered that these were “irrelevant” (at [63] of the Judgment) to the issue of the third respondent’s mental capacity, but we depart from her on this. The proven or potential presence of undue influence is relevant to the issue of mental capacity in at least three ways. The first is that it then becomes material whether P is able to retain, understand or use the information that relates to whether there might be undue influence being applied, for instance whether P can understand that a third person may have interests opposed to his; and if not, whether that inability is caused by mental impairment. The second is that it must be considered whether P’s susceptibility to undue influence is caused by mental impairment; if so, and if the result of such undue influence is that P’s will is so overborne that he is unable to use and weigh information relevant to the decision in question, P would be unable to make decisions “because of” mental impairment.

126    The third way in which undue influence is relevant is that it might mean that P cannot realistically hope to obtain assistance in making decisions. In such a situation, P may be found to lack capacity because of a mental impairment operating together with that lack of assistance. In this connection, there were times when the appellants and their associates expressed the view under cross-examination that the third respondent would be able to make decisions for herself so long as she was taken out of the influence of the first and second respondents. In submissions, this was seized upon by counsel for the third respondent who argued that it is illogical to say that P lacks capacity when P is in the company of X but does not lack capacity when P is with Y.

127    Attractive as this contention might sound at first blush, we do not regard it as well-founded. This is because it fails to give due regard to the idea that capacity under the MCA is a highly context-dependent enquiry. It is “decision-specific” (York CC at [35]) and, as we have said, it must take into account P’s actual circumstances. If P is unable to retain, understand or use information relevant to a decision because of a combination of mental impairment and the circumstances he finds himself in, the statutory test for incapacity will be met, and it is no answer then to say that P’s mental impairment would not necessarily rob him of decision-making ability in a different set of circumstances.

Whether the third respondent lacks capacity in relation to her property and affairs

128    Against the backdrop of those observations as to the appropriate approach that is to be taken, we consider the core issue in this appeal, which is whether the third respondent is unable to make decisions in relation to her property and affairs because of a mental impairment.

129    We start by examining the parameters of the declaration of incapacity that the appellants have urged us to make. In the English cases we have discussed, the questions of capacity before the courts generally concerned very narrow and specific decisions that P wished to make – for instance, whether or not to use contraception (Re A) or to cohabit with her husband (York CC). Thus, the courts in those cases had only to determine if P lacked the ability to make those particular decisions. The present appeal is not like that. Here, the third respondent is said to lack capacity “in relation to matters concerning her property and affairs”. That is a broad formulation that could comprehend a host of decisions of varying complexity.

130    We do not understand the appellants to be saying that the third respondent is unable to make any and every decision relating to her property and financial affairs. In written submissions tendered before the SDJ, the appellants pointed out that their sister is a person of great wealth, and that the financial decisions she would have to make are therefore “more complex, serious and risky than [those faced by] the average individual”.^{[note: 22]} The obvious inference is that the appellants’ worries are confined to those decisions pertaining to larger sums of money such as would generally require a higher degree of cognitive ability than those involving smaller amounts. But it is not practicable to stipulate an absolute monetary figure past which a decision would be considered to be beyond the third respondent’s capacity; for that reason, we can see why the appellants have couched their application in such wide terms.

131    On the facts before us, there are two particular decisions which provide important insights into the more general question of whether the third respondent lacks capacity in relation to decisions affecting her property and affairs. The first is her decision to set up the Trust, and the second is her decision to transfer all her assets that had been held in UBS to DBS. Should we conclude, based on the evidence, that the third respondent lacked capacity in relation to these decisions, we will then have to consider what the terms of the order should be.

132    For now, we examine the evidence that was canvassed at the trial before the SDJ. We begin with the medical evidence; we then consider the third respondent’s cross-examination and clinical interviews with her experts; and finally, we consider the remaining evidence.

Summary of the medical evidence

133    During the trial before the SDJ there were three distinct parts to the medical evidence:

(a)     discussion of the third respondent’s brain scans,

(b)     analysis of the results of the formal cognitive tests administered to her, and

(c)     consideration of her cross-examination and clinical interviews.

We will take each part of the medical evidence in its turn. Before we do that, however, we should say something about the proper limits of the expert evidence.

134    As we have mentioned, the test for capacity in s 4(1) of the MCA may be thought of as having a functional and a clinical component – the functional aspect is that P must be unable to make a decision, and the clinical aspect is that this inability must be caused by a mental impairment. It is not difficult to see that we require the assistance of expert evidence when addressing the clinical component of the test: we need medical professionals to tell us whether P has a mental impairment based on the observable symptoms and any other diagnostic tools available, and if so, what that impairment is, and what effect it has on P’s cognitive abilities. But as to the functional component, it is in our judgment a question for us to grapple with leaving perhaps a limited scope for the involvement of the medical experts. We consider that we are able to form our own assessment from the evidence, including the third respondent’s cross-examination and the clinical interviews, as to the degree to which her mental functioning is compromised; that competence derives essentially from the knowledge and experience that we as rational human beings have of the process of thinking and reasoning. It does not require the specialised expertise of a medical professional to see whether P has, for instance, poor memory or difficulty in understanding sophisticated concepts. It is ultimately the court which must decide whether P lacks the ability to make decisions within the meaning of s 5(1) of the MCA, that is, whether P is unable to understand, retain, use or weigh information relevant to the decisions that must be made, or unable to communicate his decision. In the instant case, there were occasions on which the experts put forward their opinions on this issue; but to the extent that they did so, their views in that regard should not be given weight.

135    Our observations in the preceding paragraph are especially pertinent in the context of the third respondent’s clinical interviews and cross-examination. These afford us an insight into the extent to which she is able to retain, understand, use and weigh information, and hence furnish us with evidence upon which we determine whether she is able to make decisions; but at the same time, they also reveal symptoms, such as a decline in memory or executive function, which a medical expert might use in a clinical diagnosis as to the nature and degree of her mental impairment. We rely on experts to tell us what view of her mental impairment emerges from her symptoms, but questions such as whether she is able to retain, understand, use and weigh information relevant to the decisions she makes are matters that are within our capability to assess and do not call for expert assistance.

(i)   Brain scans

136    The brain scans were of two types. One was magnetic resonance imaging (“MRI”); two such scans had been carried out, one in September 2006 and the other in January 2010. The radiologists testified on this. The other type of brain scan was positron emission tomography (“PET”); one such scan had taken place in October 2010 and this was the domain of the nuclear medicine physicians. In brief, the MRI and PET scans were intended to shed light on what type of mental impairment the third respondent had. The MRI scans showed the structure of the third respondent’s brain, including the presence or absence of bleeding within, while the PET scans indicated the degree of metabolic activity in various regions of the brain.

137    We do not propose to say much about the evidence on the MRI and PET scans because, in our judgment, it is unprofitable to go into the details. The experts from both sides agreed that all these brain scans were, in the final analysis, of limited utility in distinguishing MCI from dementia. That ultimately depended on a “clinical diagnosis” made on the totality of the available evidence, including the clinical interviews.^{[note: 23]}

(ii)   Formal cognitive tests

138    The two cognitive tests that were most administered to the third respondent were the Mini Mental State Exam (“MMSE”) and the CLOX clock-drawing test. Each of these tests was administered on at least five occasions; we do not intend to go into the results of every one of those tests and will instead highlight the more noteworthy ones. There was also something called a “trail-marking test” which we will elaborate on later.

139    Before we examine the cognitive tests, we should explain a concept called “executive function” because that is one aspect of the third respondent’s mental ability that is said to be affected by her mental impairment, and all of the tests are designed in whole or in part to assess that facet of cognitive activity. Executive function refers to a relatively diverse constellation of skills and behaviours “all of which are involved, to some extent, in the maintenance of goal-directed behaviours”.^{[note: 24]} A non-exhaustive list of the skills and behaviours considered to be part of executive function includes: (i) volition, which is the “process of ascertaining one’s needs or wants and forming a plan to achieve those goals”, (ii) inhibition and impulse control, (iii) planning and organising, which is the ability to “think prospectively, to conceive of the various available options, and to make decisions based on those options”, and (iv) abstraction, which is “the ability to think in terms of concepts or generalizations as well as the ability to think about people, events, or situations of the past, future, or imagined”.

140    Executive function has also been defined as “a set of mental processes that helps connect past experience with present action”, which people use to perform activities such as “planning, organising, strategising, paying attention to details and remembering them and managing time and space”. It “involves the ability to think abstractly and to plan, initiate, sequence, monitor and stop complex behaviour”.^{[note: 25]} It can be seen that executive function is a rather all-encompassing concept that collects within it various cognitive skills and abilities which might be quite dissimilar to one another. In the present case, the medical experts tended to speak of executive function as a unitary concept and generally did not seek to break it down into its component parts.

141    We begin with the MMSE. This test consists of a number of questions and tasks, including the following: (i) P is asked to give the year, month, date and day of the week, with one point awarded for each correct answer; (ii) three objects are mentioned for P to remember, eg, apple, table and penny, and P is asked to recall them a few minutes later, with one point awarded for each object correctly recalled; (iii) starting from 100, P is asked to count backwards by subtracting 7 five times to get 93, 86, 79, 72 and 65, in what is called the “serial 7s” part of the test, with one point awarded for each correct subtraction; (iv) P is asked to repeat the phrase “No ifs, ands, or buts”, with one point awarded if successful; and (v) P is referred to a drawing of a pair of interlocking pentagons and asked to reproduce it, with one point awarded if done correctly. The maximum possible score is 30, although the third respondent’s experts varied the test from time to time such that the highest attainable score was 28.

142    An important point to note about the MMSE is that it is not meant to be a comprehensive test to determine whether or not P has dementia or is otherwise mentally impaired. Rather, it is a “broad screening test”,^{[note: 26]} designed to give a quick indication of the likelihood of cognitive impairment on P’s part so that the tester can decide if P ought to be sent for further mental evaluation. For this reason, even though the medical experts spent some time debating the implications of the third respondent’s numerical scores – for instance, what it meant to obtain 20 or 21 out of 30 – we do not think that this was entirely helpful. We prefer to focus instead on the specific parts of the MMSE which the third respondent consistently struggled with because that tells us something about the nature of the decline in her mental functioning.

143    The third respondent often had trouble with identifying the date, month, day of the week and even the year, thus demonstrating poor orientation as to time. She was unable most of the time to recall the three objects mentioned by the tester after a lapse of a few minutes, thus demonstrating compromised memory. Finally, on the serial 7s test she would usually get the first subtraction right but fail on two or three of the remaining subtractions; by contrast, she showed an ability to perform more complex arithmetic calculations with pen and paper outside of the MMSE. This suggests that she had poor “working memory”^{[note: 27]} as a result of which she was unable to retain information long enough to process mentally the subtractions required in the serial 7s test, which is an aspect of executive function.

144    Moving to the CLOX clock-drawing test, this is said to test a fairly wide range of cognitive functions: executive function, comprehension, planning, visual memory, visuospatial abilities, motor programming and execution, numerical knowledge, abstract thinking, inhibition, concentration, and frustration tolerance.^{[note: 28]} There are two components to the test but our focus is on the one called CLOX 1, in which P is given a blank sheet of paper and told to draw a clock with the hands pointing at a certain time stipulated by the tester. The clock is then scored according to a detailed and structured set of criteria, the maximum attainable score being 15.

145    Over the course of her many sessions with her experts, the third respondent was asked to draw a total of just over ten clocks. During the joint conference, the experts discussed at length the correctness of the scores that had been given to the respondent. It suffices to say that, on their own admission, the third respondent’s experts did not always grade her clocks in strict compliance with scoring criteria; they acknowledged that they had taken a more impressionistic approach on a few occasions, and in respect of three clocks, they conceded during the joint conference that they should have given one or two points less than they did. Other than this, we do not think that there is much to gain by scrutinising the scores closely and we content ourselves with the broad observation that two of the clocks drawn by the third respondent stand out as being quite obviously abnormal. In one, she placed the numbers 1 and 7 where 12 and 6 should be;^{[note: 29]} in another, the numbers 10, 11 and 12 were omitted entirely and the numbers 3, 6 and 9 were repeated.^{[note: 30]}

146    It is safe to say that the third respondent’s clock-drawing indicates that her executive function is far from good, and we accept that this is a symptom that should be taken into account in deciding whether she has MCI or dementia. But we would add that the evidence on the CLOX tests assists us little in our task of determining whether she is able to make decisions, because it is difficult to ascertain how her imperfect clock-drawing skill relates specifically to her ability to make such decisions as she would encounter in the circumstances of her life.

147    The last cognitive test that we mention is the trail-marking test. This is said to test executive function, in particular: set shifting, mental flexibility, speed of processing, scanning and visual search. There are two parts to this test. In the first, the numbers 1 to 25 are each enclosed in circles and spread around the page, and P is asked to start from 1 and draw a line to 2, then 3 and so on until 25. In the second, the numbers 1 to 12 and the letters A to L are likewise enclosed in circles and spread around the page, and P is asked to start from 1 and draw a line to A, then 2, then B, and so on until 12 and finally L. The relevant metric is the time that P takes to complete the task.

148    For the first part of the test the third respondent took 63 seconds, which even the appellants’ experts said was within normal limits; for the second part, however, her time of 218 seconds was two standard deviations below the mean within her age group (75 to 79). Her expert, Dr Tang, maintained that this did not mean much by itself because the mean scores and standard deviations in the medical literature were in respect of “normal” individuals,^{[note: 31]} hence she might well still be “normal”. We are not sure that it is useful to employ such labels as “normal”; the real point is whether her executive function is impaired. In our judgment, the results of the trail-marking test tend to suggest that the answer is yes, and this too is a symptom to take into account in deciding whether her mental impairment is MCI or dementia; but again, it is not easy to relate her performance on this test to her decision-making ability in the context of her actual life circumstances.

149    The upshot of the cognitive tests administered to the third respondent is that there does appear to be deficiencies in her memory and executive functioning. Beyond this, as we have already said, her performance on these tests does not by itself tell us much as to whether she is able to retain, understand or use information relevant to actual decisions of the sort she will have to make in her life regarding her property and financial affairs. In that light, we turn to her cross-examination and clinical interviews.

(iii)   The third respondent’s cross-examination and clinical interviews

150    There is first a preliminary question as to the weight that ought properly to be given to the third respondent’s cross-examination. As we have mentioned, the submission was made that her responses under cross-examination are of little utility in assessing capacity under the MCA. Section 3(3), which sets out the principle that P must not be treated as being unable to make a decision unless all practicable steps to help him do so have been taken without success, was cited in support of that submission. It is argued that the third respondent did not receive assistance during her cross-examination, which was conducted as a memory or mathematics test, and that this is antithetical to s 3(3) of the MCA.

151    This was a submission that the Judge accepted. In the Judgment, she opined that “the adversarial nature of cross-examination is generally inconsistent with the principles enshrined in the MCA”, in that the function of cross-examination is for one party “to prove his case” by asking questions “designed to show a lack of capacity”, whereas an assessment of capacity under the MCA ought to be approached in an “open and neutral way” (at [131]–[132]). We do not disagree with these observations of the Judge in a general sense. We agree that cross-examination is a less-than-ideal environment in which to take the measure of P’s mental capacity, and that is why, as we will explain at the end of this judgment, we consider that in future cases, P ought generally to be examined in a clinical setting by an independent expert or experts.

152    But we would caution against a shift to the other extreme of assessing capacity by putting to P only questions designed to show that P has capacity. Certainly, there must be a degree of rigour involved in that the logic and consistency of P’s views should be challenged. Further, we do not think it inappropriate to ask P whether he can recall events or facts in relation to documents placed before him, or to ask P to read a document and explain his understanding of it. The principle in s 3(3) of the MCA cannot be stretched too far. Hence, when it comes to clinical interviews intended to serve a diagnostic function, that is, to give the court an idea of the state of P’s cognitive functioning, s 3(3) of the MCA does not give a tester of P’s capacity the license to coach P on the answers she should give or to prompt and guide her at every turn.

153    We are not suggesting that this is what the third respondent’s experts did in their clinical interviews with her. However, we do consider that those clinical interviews should be treated with a degree of circumspection because they were conducted entirely by medical practitioners instructed by her. While it is true that her experts did not have any prior personal or professional relationship with her, we do not think that the experts’ independence is secured by this fact alone. In the main, we agree with the SDJ’s assessment that the clinical interviews had not been “rigorous” and “probing” enough. We would also point out that during the witness conference one of her experts, Dr Ung, acknowledged that he felt “empathy” for her.^{[note: 32]} Another of her experts, Dr Fones, accepted the suggestion that he had “felt some sympathy for her because she was trying her best” in the context of a clock-drawing test, and he said he wanted to “give her some recognition for her efforts”.^{[note: 33]} And, as we have mentioned, her scores on the clock-drawing tests were on three occasions inflated by one or two points. The experts maintained of course that none of this undermined their neutrality and we do not question that. But, we also do not feel satisfied that the clinical interviews give us an entirely accurate impression of the third respondent’s mental abilities.

154    In the circumstances, we do not disregard either the third respondent’s performance under cross-examination, or her clinical interviews; but we consider them in the round having regard to the overall picture that emerges from a consideration of all the materials. As we have said, the cross-examination and clinical interviews provide (i) evidence as to whether the third respondent is able to retain, understand, use and weigh information relevant to decisions, and (ii) evidence of her symptoms such as might shed light on the nature of her mental impairment. Since these two facets of the analysis are closely related to each other, we will deal with them together in the paragraphs that follow.

155    First, we consider the quality of the third respondent’s memory. In our judgment, her cross-examination demonstrates a pronounced inability to remember important and fundamental details, and sometimes even after having been reminded of those details just a short time earlier. She could not remember how she came to sign any of her affidavits.^{[note: 34]} On 14 February 2012, she could not recall having asked her lawyers to file a court summons dated 3 November 2011 seeking permission to withdraw or transfer a sum of up to HKD200,000 a month from a certain UBS account,^{[note: 35]} and about an hour after having been shown the summons, she was no longer able to remember the amount per month she had asked to withdraw.^{[note: 36]} She did not seem aware that her sisters, the appellants, had asked that they be appointed as her deputies under the MCA.^{[note: 37]} She was surprised when told that it was her own lawyers who had put in an application seeking cross-examination of all witnesses.^{[note: 38]}

156    She could not remember having been shown an e-mail about 35 minutes after it had been shown to her.^{[note: 39]} She could not remember having been shown a letter just over 20 minutes after it had been shown to her, this being a letter in which she had reprimanded the first respondent;^{[note: 40]} similarly, the fact of having been shown another letter in which she chastised her daughter escaped her after no more than 12 minutes had elapsed.^{[note: 41]} In the course of proceedings she was on multiple occasions referred to the limited power of attorney she had executed in the first respondent’s favour on 1 November 2010, but despite this she could not recall having signed it.^{[note: 42]} She could not remember whether she had given a sum of HKD1 million to the first respondent,^{[note: 43]} and she could not recall a document relating to the transfer of assets from JP Morgan to UBS.^{[note: 44]} She did not remember having threatened to disinherit the first respondent.^{[note: 45]} She was under the erroneous impression that UBS had acted on her instructions to transfer her assets to DBS.^{[note: 46]}

157    There are in truth a number of other examples of the third respondent’s poor memory but we do not need to belabour the point. It is apparent that her lapses in memory are not merely trivial and incidental; on the contrary, they concern documents and events with which she ought to have been intimately familiar; which involve substantial sums of money; or which pertain to her relationships with the people closest to her. In our judgment, these raise serious doubts about her ability to retain information relevant to the kinds of decisions relating to her property and affairs which she would have to make.

158    Next, we consider the third respondent’s ability to understand information, documents and events. Her cross-examination reveals that she faces some difficulty in this regard. When she was shown the summons in which she sought permission to withdraw HKD200,000 a month, she said that it was a bank statement of her account in Hong Kong.^{[note: 47]} She did not know what the consequences were of deputies being appointed for her – although in fairness, in a subsequent clinical interview, she was able to explain that deputies would make decisions for her. When she was shown one of her own affidavits, she initially said that it was an affidavit by the appellants, and this even after having been asked to read the first paragraph of the affidavit which clearly identified its author; she did eventually realise that it was her own affidavit.^{[note: 48]} She also did not appreciate the risks involved in being represented by lawyers who had also represented the first and second respondents at the beginning.^{[note: 49]}

159    These and other instances suggest a deficiency in the third respondent’s abilities of comprehension, but as against that, there are parts of her cross-examination and clinical interviews which show a grasp of the reasons behind things. To take some examples: she said that she would not seek financial advice from anyone who had an interest in her property,^{[note: 50]} and when questioned by the SDJ she was able to state in simple terms what the court proceedings broadly were about.^{[note: 51]} But this needs to be qualified in that while the third respondent did in some instances display an ability to comprehend matters of some complexity, she did not seem able to work through the consequences of such comprehension. By way of illustration, in relation to the Trust, she seemed to accept that even though her desire expressed in her letter of wishes was that the money in the Trust should be given to charities after her passing, the first respondent had the discretion to do what she liked with the money; in other words, the first respondent had no legal obligation to carry out her wishes and she accepted that she had simply to trust that her daughter would follow her wishes.^{[note: 52]} But then, as we note below, she did not seem to appreciate that she might be better off by making a will that provided for this to happen, instead of depending on her daughter’s goodwill in carrying out her wishes.

160    Taking all this in the round, in our judgment, the third respondent’s ability to understand information, concepts and ideas relevant to the sort of decisions she would have to make is somewhat compromised. It is possible that some of her most obvious lapses in her comprehension of relatively simple matters might at least in part be due to fatigue or stress or the impaired memory we have referred to or a combination of these rather than a complete lack of cognitive ability, but we are satisfied that there is some compromise in this regard.

161    Then we turn to the third respondent’s ability to use or weigh information as part of the decision-making process. The third respondent was able to give broad explanations for some of her decisions that sounded sensible and reasonable. For instance, she explained that she wanted to transfer her assets to DBS because it was a bank with the backing of the Singapore government. On the other hand, she also said that she made the deliberate choice to have banking relationships with more than one bank so that they would compete with one another and do the best for her. As we note below, she did not seem able to reconcile these two positions.

162    In her clinical interviews, the third respondent was able to explain a preference for investing in low-risk fixed deposits,^{[note: 53]} and to say that she would not invest in derivatives and similar financial products because she did not understand them.^{[note: 54]} She also seemed to appreciate the prudence of seeking advice from more qualified individuals when making financial decisions. But for her to be able to make such decisions, she must have the ability to appreciate and to articulate the risks involved in embarking on a particular course of action, or to explain why she chose one course of action over another. In this respect, she was lacking. She said that she did not see any risks involved in signing a power of attorney in favour of someone else – although she did say that she would seek legal advice before signing^{[note: 55]} – or in removing all signatories to her bank accounts except herself.^{[note: 56]} When she was asked to explain why she wanted to move all her assets to DBS, she put forward her reasons but was unable to address the counter-arguments or to conceive of alternatives that would also fit with the reasons that she gave; and when she was asked why she set up the Trust rather than draw up a will to achieve certain objectives, she did not demonstrate much awareness of the various pros and cons of either course of conduct. These are matters that we elaborate on below.

163    Finally, it does appear from the cross-examination and clinical interviews that the third respondent labours under a number of paranoid and false beliefs. The first appellant’s unchallenged evidence was that the third respondent had levelled accusations at her sisters, alleging that one of them had stolen her purse or various possessions, and that she had labelled as “treacherous” bankers whom she had previously trusted. In the same vein, both NG and the first respondent recounted an incident in 2009 in which the third respondent misplaced her jade beads at NG’s house and thereafter expressed the belief that they had been stolen by someone in the household. The first respondent testified that the third respondent believed that NG herself had been the thief;^{[note: 57]} NG’s testimony did not quite accord with this but she added that the third respondent had gone so far as to say that she was going to make a police report.^{[note: 58]}

164    In one of the clinical interviews, the third respondent suggested that the appellants and her son CK were hiding something and alluded vaguely to an investigation by the Commercial Affairs Department, implying that the two things were related;^{[note: 59]} but in fact the Department had contacted her concerning their investigation against a former relationship manager of hers who has since been charged and convicted in court for having cheated her of money. She also said that when she was living with CK and his wife in Hong Kong, she drank in fear the soup placed before her because she “don’t know whether they put anything in”.^{[note: 60]} In cross-examination, she testified on 14 February 2012 that she had, during a break in the court proceedings, spoken to her counsel and asked him to be one of her deputies^{[note: 61]}, but counsel later informed the court that this had not taken place.^{[note: 62]}

Evaluation of the medical evidence

165    The central dispute in the expert medical evidence was what the third respondent’s mental impairment was, ie, whether it was MCI or dementia. According to the medical literature adduced in evidence, MCI and dementia may be conceived of as existing along a single continuum called the Global Deterioration Scale ranging from normalcy to severe Alzheimer’s disease.^{[note: 63]} Put another way, MCI “represents an intermediate state of cognitive function between the changes seen in ageing and those fulfilling the criteria for dementia and often Alzheimer’s disease”.^{[note: 64]} The symptom that MCI and dementia have in common is a deterioration in memory; and the main difference between the two is that MCI sufferers generally manifest little or no other symptoms in addition to compromised memory – any deficits in cognitive function are “subtle” – whereas a range of other symptoms may be observed in those with dementia.

166    Both the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (“DSM-IV”) and the tenth revision of the International Classification of Diseases (“ICD-10”) require the manifestation of symptoms other than memory impairment before a clinical diagnosis of dementia may be made. The DSM-IV requires at least one of the following: (i) aphasia, which is deterioration in language function, (ii) apraxia, which is impaired ability to execute motor activities, (iii) agnosia, which is failure to recognise or identify objects despite intact sensory function, and (iv) disturbances in executive functioning. Moreover, the memory impairment and the other symptoms “must be severe enough to cause significant impairment in social or occupational functioning”, and “must represent a decline from a previous level of functioning”. The DSM-IV adds that an associated descriptive feature of dementia is that “[d]elusions are common, especially those involving themes of persecution”, and the example given is the belief that “misplaced possessions have been stolen”.^{[note: 65]} As for the ICD-10, this requires, in addition to memory deficits, a “decline in other cognitive abilities characterised by deterioration in judgment and thinking, such as planning and organizing, and in the general processing of information”, as well as a “decline in emotional control or motivation, or a change in social behaviour” which may appear in different forms, including apathy or “coarsening of social behaviour”.^{[note: 66]}

167    In our judgment, it is apparent that the third respondent’s symptoms are not limited to the deterioration of her memory. We are satisfied, on the basis of the results of the cognitive tests and her clinical interviews and cross-examination, that there has also been a decline in her executive function. For a number of reasons, we were unconvinced by the third respondent’s experts’ assertions that her executive functions were not compromised. First, one of them, Dr Fones, had acknowledged a “mild degree of impairment of executive function (planning, organizing)”^{[note: 67]} on the part of the third respondent. Second, the cognitive tests they had administered to the third respondent demonstrated clear imperfections in her executive function. Third, the reliability of their diagnoses arising from the cognitive tests was undermined by their over-generous scoring and the “sympathy” or “empathy” which they felt for her. Fourth, and finally, they tended to dismiss her poor performance under cross-examination by saying that the conditions of cross-examination were not conducive to an assessment of capacity, but, as we have said, we take the view that her responses in court cannot be disregarded and must be given weight.

168    There remains the matter of the paranoid and false beliefs that the third respondent labours under. Unfortunately, this received very little attention in the medical evidence. The appellants’ experts did not delve into this in any detail: Dr Chan simply stated that there “appears to be evidence of paranoid delusions” and “delusions of theft”, and thereafter, having opined that the third respondent had dementia, explained that dementia causes “abnormalities in thoughts like delusions and paranoid behaviour”.^{[note: 68]} The third respondent’s experts dealt with it in a similarly perfunctory manner: Dr Fones stated without elaboration that she “did not suffer from any delusions that would account for her views” as to the persons she wanted to be involved in managing her estate and affairs for her, while Dr Ung contented himself with a one-liner, which was that he “could not elicit any delusions”. It might be that Dr Ung takes the view that there are no delusions because the medical definition of a delusion is more stringent than what a layman might understand it to be: according to Dr Ung, a delusion is a false imaginary belief “held with abnormal conviction and tenacity despite evidence to the contrary”.

169    There is no denying the fact that the third respondent does manifest a number of paranoid beliefs that appear to be without basis. Given the dearth of expert evidence on the issue, we do not think we are in a position to make a finding on whether these beliefs were “delusions” in the strict sense of the word as adopted by Dr Ung; but what is material, in our view, is that the third respondent clings to these paranoid beliefs with a substantial degree of firmness and persistence, so much so that they affect her actions in tangible ways. Her belief that family members had on various occasions misappropriated her things was strong enough that she broadcast it openly, and her belief that hitherto-trusted people – family members and professionals – were turning against her reached such a level of conviction that she proceeded to distance herself from them. We accept the evidence of the first appellant, CK, NG and the bankers that these paranoid beliefs arose without any apparent trigger, starting from around the time when the deficiencies in her memory were first noticed by those around her. In such circumstances, we think it more likely than not that the third respondent’s paranoid beliefs are properly to be considered a symptom of her mental impairment, which is to say that those beliefs were caused by her mental impairment just as much as the deterioration in her memory was caused by it.

170    On a consideration of the totality of the third respondent’s symptoms, it is not clear that she fulfils the criteria for a diagnosis of dementia under either the DSM-IV or the ICD-10. As to the DSM-IV, although she suffers from memory impairment and a disturbance in executive function, it may be going too far to say that these have resulted in “significant impairment in social or occupational functioning”. The fact is that she remains capable of a degree of independent living – for instance, as her experts pointed out, she remains able to choose attire appropriate to the occasion and to attend medical appointments on her own. As for the ICD-10, although the decline in her cognitive activities such as judgment and thinking appears to have caused “impaired performance in daily living, but not to a degree making the individual dependent on others” such as would fulfil one criterion for a diagnosis of “mild” dementia, there is insufficient evidence that there has been a “decline in emotional control or motivation, or a change in social behaviour” on her part.

171    However, turning to the Global Deterioration Scale, we think that the third respondent’s symptoms take her beyond the stage of mere MCI. She manifests symptoms that at least go towards a diagnosis of mild Alzheimer’s disease, including (i) an inability to recall “seemingly major recent events”, such as the giving of banking instructions involving large sums of money, (ii) “overt mistakes in recalling the day of the week, month or season of the year”, and (iii) a “decreased capacity to manage personal finances”.^{[note: 69]} She even manifests a symptom that would go towards a more serious diagnosis of moderate Alzheimer’s disease, namely, compromised orientation as to time “to the extent that the correct year may not be recalled”, which happened twice in the MMSEs administered to her.

172    Thus it is difficult to state conclusions with any certainty as to the identity of the third respondent’s mental impairment; the answer may differ depending on the criteria employed. But ultimately we do not think this is unduly problematic: we see no objection to remaining agnostic in the final analysis as to what is the exact name to be given to the third respondent’s mental impairment so long as we know that it is situated somewhere between MCI and dementia on the continuum or scale on which they both exist. That amounts to “an impairment of, or a disturbance in the functioning of, the mind or brain” within the meaning of s 4(1) of the MCA. The next step in the analysis is then to consider the effects that the mental impairment has on the third respondent’s cognitive functions, so that we may determine whether she has the ability to make decisions relating to her property and affairs.

173    In summary, we are satisfied that the third respondent has an impairment or disturbance in the functioning of her mind, and that it was something between MCI and dementia. We are satisfied that this has caused: (i) a significant decline in her memory, (ii) a less severe but observable deterioration in her executive functions, including and especially in her ability to understand information and to weigh countervailing considerations against one another, and (iii) the emergence of paranoid and false beliefs that can and do affect her actions. Our task now is to examine whether the cumulative operation of all these things renders her unable, in the actual circumstances of her life, to make decisions relating to her property and affairs. As we have indicated, we shall do this by subjecting to scrutiny two decisions that she actually made, namely, her decision to set up the Trust and her decision to transfer her UBS assets to DBS.

The decision to set up the Trust

174    From an objective viewpoint, it is not easy to see any good reason for establishing the Trust. So far as the Trust is meant to provide for the third respondent’s material needs during her lifetime, there was already in existence a trust with JP Morgan that had been created to do just that; and furthermore, prior to the execution of a deed of understanding on 27 July 2012, some 21 months after the Trust was created, it appears that there was no legal obligation on the trustee to apply the money in [B] Ltd’s bank account towards the third respondent’s needs. So far as the Trust is meant to (i) ensure that a substantial portion of her wealth is put to charitable purposes, (ii) make a gift of $10 million to the first respondent and (iii) disinherit her other children, CK and NG, after her passing,^{[note: 70]} all this could have been achieved by making the relevant bequests in a will, and there is nothing to suggest that this would have been a less satisfactory device than establishing the Trust and may in fact have been more satisfactory from the point of view of securing the third respondent’s wishes. In cross-examination, the third respondent said that a will would require her to “go through [a] lawyer and spend a lot of money”.^{[note: 71]} But this is hardly an adequate explanation because it would appear that legal assistance was similarly necessary in establishing the Trust. In short, the Trust does not appear to advance the situation of the third respondent in any discernible way and might even have left her worse off than had she created a will. The important thing though is that she did not seem alive to the possible differences or to the need to choose the appropriate course between them.

175    We are mindful of the principle in s 3(4) of the MCA that a person “is not to be treated as unable to make a decision merely because he makes an unwise decision”. Even if the creation of the Trust might be characterised as an unwise decision, that by itself does not warrant a finding that the third respondent lacked the ability to make it. The test is whether she is able to understand the information relevant to the decision and, if she does, whether she is then able to use and weigh all that information in the process of coming to her decision.

176    In our judgment, the evidence does not establish that the third respondent completely lacked the ability to understand the information relevant to the decision to set up the Trust. Such information would include: how the Trust works, what benefit she stands to gain from it, and what costs and risks are involved. She understands that the Trust exists only for her benefit for as long as she is alive. She understands that even if it is her wish that the Trust monies go to charity after her passing, the actual application of the funds might be at the first respondent’s discretion. She also understands that under the terms of the Trust, her passing will benefit the first respondent in the amount of $10 million – and indeed she says that she intended this to be an expression of gratitude to the first respondent for taking care of her.

177    But it is less clear whether she has the ability to use and weigh the information relevant to her decision to set up the Trust. What this requires of her is an ability to engage with the countervailing considerations relevant to the decision – pros and cons, costs and benefits – and to measure them one against another in a non-arbitrary manner. This is not to suggest that feelings and intuition ought never to play a part in the process of decision-making; taking for instance a man who spends a great deal of money participating in a lottery which he knows he has a miniscule chance of winning because he has a strong intuition that it will be his lucky day, we would not necessarily conclude that such a man lacks the ability to use and weigh the information relevant to the decision he has made. Human experience shows that we do make decisions on the basis of rational reasons as well as irrational impulses or instincts or feelings; and even though these rational and irrational factors may be incommensurables, the reality is that we have nonetheless to reckon with them in the same decision-making equation. What matters is the ability to engage with all these factors, rather than allowing one or some of them to dominate the decision-making process such that the other relevant factors are effectively excluded from that process.

178    In order to determine whether the third respondent was able to use and weigh the information relevant to her decision to set up the Trust, we consider whether she was able to provide reasons for making the decision. Here, we found her ability was deficient. She was unable to explain why it was necessary or even desirable to create the Trust if her objectives were merely to provide for herself during her lifetime and to make donations to charity after her passing. She did not seem capable of (i) considering whether one of these objectives was already being met by other trust arrangements already in existence, and (ii) assessing the relative merits and demerits of setting up the Trust versus drawing up a will.

179    Moreover, the evidence suggests that those were not in fact her objectives. On numerous occasions not only in court but also in the course of clinical interviews with her experts, she said that she had created the Trust in order that she might hide her money from other people, and in particular, from her son, CK. This, she said, was for her “survival”;^{[note: 72]} she recounted having “felt so relieved” when he was “shocked” at being unable to find her money.^{[note: 73]} It became her refrain that she had moved her money into the Trust “so that nobody knows”^{[note: 74]} and “nobody can find”.^{[note: 75]}

180    It is apparent that the third respondent believes that, had she not established the Trust, CK would have come after it and eventually left her bereft. This belief is problematic for two cumulative reasons. The first is that it seems to have arisen suddenly in her mind without any apparent cause, and the second is that it does not appear to be a reasonable belief in that there is no perceptible factual basis for it. This raises the possibility that the third respondent’s decision to set up the Trust was largely if not wholly impelled by a paranoid belief that was caused at least in part by her mental impairment.

The third respondent’s dealings with her banks

181    We turn now to the third respondent’s dealings with UBS and JP Morgan. Of particular concern is her decision to transfer her UBS assets to DBS and the inconsistent instructions that she gave in connection with it. To recapitulate, the sequence of events was as follows: on 8 November 2010, she gave instructions to UBS to transfer all her assets to DBS, but revoked these instructions two days later; on 19 November she renewed her instructions to transfer, only to revoke them again on 23 November; and finally, she gave instructions to transfer on 29 November and 2 December and confirmed them at a meeting on 15 December. It is evident that the volte-face was partly a result of her poor memory – on 10 November 2010, she had no recollection of instructions she had given just two days earlier. This was by any measure a very significant lapse in her memory.

182    Another incident worth mentioning in the third respondent’s dealings with UBS is the meeting that she had on 15 December 2010 with her erstwhile UBS banker, Ms B, the Singapore lawyer, Mr L, and other bankers from UBS and DBS. Ms B was the only person who gave any useful evidence as to what occurred at this meeting, and in the absence of any conflicting account or substantial attack on her credit, we accept her testimony as true. Hence, we accept that, when the third respondent was brought through the series of inconsistent instructions she had given starting from 8 November 2010, she was “unable to recall anything”.^{[note: 76]} Moreover, even after the third respondent said that she did want to close her accounts with UBS, she continued to give Ms B instructions as if those accounts were to remain open after all.^{[note: 77]}

183    In our judgment, the events of the 15 December 2010 meeting point to the third respondent’s inability to make the decision to transfer her UBS assets to DBS even as she was apparently making it. Her inability to recall her previous inconsistent instructions constitutes a failure of her ability to retain information relevant to her decision, and her persisting in giving Ms B instructions in respect of her UBS accounts despite having confirmed the closure of those accounts casts serious doubts on whether she understood what was happening at that meeting, and hence suggests that she lacked the ability at the time to understand the information relevant to her decision.

184    There is also evidence that her ability to use and weigh information relevant to her decision to transfer her UBS assets to DBS was compromised. In court and in her clinical interviews, she explained that she wanted the security of placing her assets in a bank that was backed by the Singapore government. She said also that she trusted Ms D, the senior banker at DBS with whom she interacted fairly frequently. But the fact is that she already had a banking relationship with DBS at the time, except that she also had relationships with UBS and JP Morgan. She agreed that she had chosen to open accounts at more than one bank because she wanted to “diversify” and so that the banks would “compete and do best” for her.^{[note: 78]} She did not satisfactorily explain why her desire for security at DBS and her trust in Ms D would translate into a decision to transfer all her assets in UBS to DBS, nor why such a decision had to be made at that particular time.

185    We note that the third respondent did furnish a reason for wanting to sever all ties with UBS, which was that she did not trust her UBS banker, Ms B. She opined more than once that Hong Kong people – Ms B is of Hong Kong nationality – were “all double-faced” and could not be trusted.^{[note: 79]} She also described Ms B as “double” in the sense that she was “sweet tongued but don’t mean it”, and when asked what evidence she had of this replied, “No need evidence, I can feel she is not that good after a while”.^{[note: 80]} Yet Ms B’s testimony was that the third respondent showed no signs of distrust or animosity towards her when they met on 23 November 2010, and as we have already described, as late as 15 December 2010, the third respondent continued to ask Ms B to purchase renminbi products for her. It is difficult to avoid the sense that the decision, if indeed it was that of the third respondent, was an arbitrary one, and the inference that might be drawn is that she lacked the ability to use and weigh the information relevant to that decision, if she even understood or retained that information at all.

186    Thus far, we have discussed the third respondent’s banking relations with UBS rather than JP Morgan because the evidence focussed very much on the transactions with UBS. This is not surprising because the UBS transactions demonstrate more starkly her compromised decision-making ability. But we would add that the evidence on her banking relations with JP Morgan did fortify the view that she lacks the ability to make the sort of banking-related decisions that someone of her substantial wealth would have to make. At the meeting on 13 December 2010 between the third respondent, Mr L and bankers from JP Morgan and DBS, according to a contemporaneous attendance note prepared by the JP Morgan bankers, Mr L did most of the talking at this meeting, with the third respondent’s participation confined to nodding in assent at certain points and making small talk unrelated to the transfer of assets. The limited extent of her participation, and her tendency to stray into irrelevant matters, does hint at an inability to understand the information relevant to her decision or to use and weigh it.

187    Earlier in this judgment we made mention of internal notes prepared by JP Morgan bankers in June and July 2009 in which they noted that the third respondent’s memory was deteriorating and that there were times when she could not recall transactions that she had placed with them. The fact that her memory was already in such a state at that time does tie in with the evidence that she had a diminished ability to retain information relevant to her banking decisions in November and December 2010. We mentioned also the testimony in court of a former JP Morgan banker that the third respondent (i) instructed him to send her bank statements to the first respondent but was later upset that he had done so and (ii) would ask him to carry out a transaction shortly after that very same transaction had been completed, as if it were a fresh idea; and to this we would add the banker’s evidence that she would agree to any transactions he recommended without asking “insightful questions” or even asking “what’s the risk”.^{[note: 81]} In this way, the evidence as to her transactions with JP Morgan does strongly corroborate the evidence as to her banking relations with UBS.

The circumstances in which the decisions were made

188    The sum of our reasoning so far is that there is strong basis for saying that the third respondent lacked the ability to make the decision to set up the Trust as well as the decision to transfer all her UBS assets to DBS. But before we draw any firm conclusions on her capacity, we should look at the actual circumstances in which those decisions were made.

(i)   Possibility of undue influence

189    It is vigorously asserted by the appellants and those associated with them, especially CK and NG, that the third respondent is subject to the undue influence of the first and second respondents. One of their main allegations is that the first and second respondents engineered the creation of the Trust and the transfer of the third respondent’s UBS assets to DBS in order that they might purchase the Bungalow for $30 million using the third respondent’s money. It is not disputed that all these events took place around the same time. But it is another thing to say that it is more likely than not that this was indeed the plan of the first and second respondents.

190    The first and second respondents’ version of events is that they made the decision in September or October 2010 to spend more time in Singapore and to that end, they looked for a property to purchase. After a weekend of house-hunting they settled on the Bungalow; the option to purchase was granted on 30 October, the option was exercised on 19 November and on 26 November, the stamp duty, amounting to almost $900,000, was paid. We pause at this point to merge these events into the wider chronology of the other events involving the Trust and the third respondent’s banking instructions:

(a)     30 October 2010: option to purchase granted to the first and second respondents in respect of the Bungalow.

(b)     1 November 2010: the third respondent grants the first respondent a limited power of attorney over her UBS accounts.

(c)     5 November 2010: the first and third respondents travel to Singapore from Hong Kong.

(d)     6 November 2010: the third respondent signs documents relating to the Trust and/or the ancillary banking arrangements, in the presence of the first respondent, Mr L, Ms D and others.

(e)     7 November 2010: the first respondent returns to Hong Kong, leaving the third respondent in Singapore.

(f)     8 November 2010: UBS receives the third respondent’s instruction to transfer all her assets held with the bank to DBS.

(g)     10 November 2010: the third respondent tells Ms B of UBS not to act on the 8 November instruction.

(h)     19 November 2010:

(i)       UBS receives another instruction from the third respondent to transfer all her assets to DBS;

(ii)       The first and second respondents exercise the option to purchase the Bungalow.

(i)     23 November 2010: at a face-to-face meeting with UBS bankers and in the presence of the first appellant, the third respondent revokes her 19 November instruction to transfer her UBS assets to DBS, as well as the limited power of attorney granted in favour of the first respondent on 1 November.

(j)     24 November 2010: an officer from DBS turns up at the third respondent’s apartment with documents for her to sign, these documents containing instructions to transfer her JP Morgan assets to DBS, but the third respondent declines to sign them.

(k)     25 November 2010: the first respondent arrives in Singapore.

(l)     26 November 2010:

(i)       The third respondent signs further documents relating to the Trust and/or related banking arrangements, in the presence of the first respondent, Mr Z and Ms D;

(ii)       The first and second respondents pay the stamp duty on the Bungalow.

(m)     28 November 2010: the third respondent returns to Hong Kong with the first respondent.

(n)     29 November 2010: UBS receives yet another instruction from the third respondent to transfer her assets to DBS.

(o)     2 December 2010: the third respondent reiterates her 29 November instruction to UBS.

191    These events lend themselves to certain observations. First, the events surrounding the acquisition of the Bungalow and the creation and the execution of the Trust and the associated documents were closely connected in time and even interwoven with interactions between the first and the third respondents and with the instructions that were being given to UBS to transfer the third respondent’s assets to DBS. Second, the third respondent changed her position rather dramatically at various points within this brief period of four weeks or so in relation to the operation of her UBS account, and in general, these reversals were in favour of the first respondent when the third respondent was in her presence, but against the first respondent when she was not in her presence and was instead in the presence of the first appellant.

192    The first and second respondents say it was pure coincidence that the Trust was created shortly before the grant of the option to purchase and that the third respondent’s first instruction on 8 November to transfer money from UBS to DBS was given around that period. There is some evidence which supports their case that their intention all along was to purchase the Bungalow using their own funds. There is an e-mail which shows that on 18 October 2010, they were working out how much cash they had available,^{[note: 82]} presumably with a view to purchasing the Bungalow. There are also e-mails suggesting that from 1 to 3 November the first respondent was arranging to sell her shares, presumably in order to raise funds for the property purchase.

193    As against this, the second respondent did not come across as having been a truthful witness on the stand. In particular, he put forward a highly suspect explanation for an unusual aspect of the purchase of the Bungalow – which was that, in the caveat lodged in respect of the Bungalow which named him as the caveator, it was also expressly indicated that he was but a trustee for his and the first respondent’s children. It is not disputed that the first respondent, not being a Singaporean, was precluded by Singapore law from owning the property. But the appellants say that it is odd that the second respondent would have taken the trouble to hold the property as trustee and to have broadcast that fact when he could simply have held it in his name, without more.

194    In his evidence in court, the second respondent explained that the reason why he made it clear in the land records that he was holding the Bungalow on trust for his children was simply that the Bungalow had been purchased for their benefit. But he also claimed that his intention all along had been that the names of his children should not appear in the records – and this was patently at odds with the contents of an e-mail that was in evidence. This e-mail was from him to his lawyer dated 10 November 2010 in which he sought advice as to what the ownership structure over the property should be. He wrote:

a. [The first respondent] would like to know, if we use a shell company to buy rather than in my personal name, what is the likely additional one - off set up and recurrent costs? She wants to consider this option because she is sensitive to how the open land title will name me – will it show me as owner, but “holding on trust for….”? Otherwise, she is prepared to pay extra for the shell company to be named as owner [ even though I have said that it will be very easy to prick the veil of secrecy, as I shall still have to be the sole shareholder/director of the company, unless we set up a trust company with corporate trustees etc ?] [emphasis added]

195    The second respondent said that his concern here was that he did not want the names of his children to be on the title document because he did not want people to know that his children were holding a valuable property.^{[note: 83]} But this cannot be correct – a plain reading of the e-mail demonstrates that he in fact wanted it to be shown that he was holding the property on trust for others, “[o]therwise” he might be willing to pay extra for a shell company to be named as owner. He testified that when he used the word “Otherwise” he meant “If so”^{[note: 84]} but this is inherently improbable since “otherwise” ordinarily means “if not”. His true concern was plainly to keep his own name out of the title documents. Indeed, the fact is that he wanted the children’s names on the records because in an earlier e-mail to his lawyer dated 28 September 2010,^{[note: 85]} he had asked her if it was possible for the property to be purchased in the names of the children.

196    What this comes to is that the second respondent has not been candid as to his reasons for stating explicitly in the land records that he holds the Bungalow on trust for his children. The inference which the appellants urge is that he was holding it on trust for the children so that the third respondent would not mind if her money was used to purchase the property. The undisputed background to this is that she is generally suspicious of in-laws and would not be willing to give her money towards a property if that property was to be in the name of an in-law. As unsatisfactory as the second respondent’s evidence was, however, we do not think that we are able to make the logical leap and conclude that in fact the first and second respondents intended to use the third respondent’s assets in this venture. The evidence shows that the first and second respondents purchased the Bungalow using their own money, and that the third respondent gave them no financial aid. Moreover, there is no evidence that the first and second respondents had any means to gain access to the third respondent’s assets in the Trust and in her DBS accounts, save by persuading her to turn such assets over to them, and there is no evidence that in fact they did this. We are therefore unable to conclude that the first and second respondents engineered the creation of the Trust and the transfer of assets to DBS with the specific objective of funding the purchase of the Bungalow.

197    Having said that, we think that there is a strong case for inferring that the third respondent was acting under the undue influence of the first and second respondents when she established the Trust and transferred her assets to DBS, even if we are unable to conclude that this undue influence was in fact directed towards the acquisition of the Bungalow. As we have found, the establishment of the Trust and the transfer of assets to DBS do not appear to bring any discernable benefit to the third respondent; they occurred without any evident prompt or reason; the Trust was set up in secrecy and the banking instructions were given with unusual urgency and were, as we have noted above, subject to inexplicable (and indeed unexplained) reversals. Given the manner in which the transactions were carried out, it is difficult to believe that they were the considered expressions of the third respondent’s volition. The fact is that, on both occasions when the third respondent signed the documents relating to the Trust, the first respondent was with her in Singapore. In contrast, when the first respondent was away, the third respondent (i) revoked on 10 November 2010 and 23 November 2010 her previous instructions to transfer her UBS assets to DBS, and (ii) declined on 24 November to sign instructions to transfer her JP Morgan assets to DBS. In short, the choices made by the third respondent varied drastically depending on whether she happened to be with the first respondent, and this strongly suggests that the third respondent was subject to the undue influence of the first respondent. This conclusion is buttressed further by the evidence on the first and second respondents limiting the access that others could have to the third respondent, and it is to this we now turn.

(ii)   Cutting off access

198    The evidence shows that, after the third respondent went to live with the first and second respondents in Hong Kong at the end of November 2010, she became isolated from the rest of her family members, including the appellants and her children, CK and NG. It is not disputed that CK and NG were unable to see their mother when they visited the first and second respondents’ home on 1 December 2010; given the second respondent’s poor performance on the witness stand as noted by the SDJ, we are inclined to believe CK’s and NG’s evidence that the second respondent barred them from entering the house without good reason. Furthermore, the unanimous testimony of the first appellant, CK and NG – which we accept as reliable – is that they were unable to get through to the third respondent on the telephone because it was constantly on fax mode.

199    The first and second respondents say that they have not cut off access to the third respondent; they claim that it is her own wish to have nothing to do with the appellants and her other children. They allege that CK and NG have harassed her and that she desires to be left alone. The third respondent did express agreement with that characterisation of things in court and in clinical interviews, but this should not be given much weight since the appellants’ case is that any hostility towards them, CK and NG on her part is the product of undue influence and does not reflect her true feelings and attitudes.

200    In our judgment, the evidence strongly suggests that, if the third respondent were completely at liberty to make her own independent decision, she would not want to cut off access and would instead wish to meet and interact with the appellants and her children, CK and NG. By her own admission in court the third respondent was “quite happy” to see CK at the Hong Kong crafts shop on 17 December 2010; it is telling that she was not with the first and second respondents on this occasion. There is also the fact that the third respondent’s entire family met for lunch on Christmas Day just over a week later with all the grandchildren in attendance – that this lunch took place at all suggests that the third respondent was not intent on keeping herself away from CK and NG. Then there is NG’s testimony that, after this lunch, her mother asked her to call “any time”; this is corroborated by an e-mail dated 27 December 2010 in which NG informs the first respondent that their mother had left a message on NG’s voicemail asking her to call back, but when NG tried to call, the phone was engaged or on fax mode.^{[note: 86]} What this shows is that the third respondent was in fact desirous of speaking to NG at least. Yet, despite this, NG had no access to her mother.

201    It is not only the third respondent’s family members who were cut off from her – at least one other person encountered a similar fate. This is the former JP Morgan banker whose testimony we have referred to at various points. He gave evidence that, on a day at the end of October or the beginning of November in 2010, he had, in the morning, a friendly conversation with the third respondent in which she asked about renminbi investments and told him to call her again in the afternoon. When he did call that afternoon, the first respondent answered the phone and told him that he should not call the third respondent in the future because she no longer needed his services. The first respondent then put the third respondent on the phone to confirm what she had said.^{[note: 87]}

202    We are satisfied on the evidence that the first and second respondents did endeavour to cut the third respondent off from the appellants, CK and NG, as well as from professionals whom she had trusted and relied on all along; we are also not satisfied that this was done in accordance with her true wishes and desires. This fortifies our conclusion that the first and second respondents are exercising undue influence over the third respondent.

203    Given the fact that the third respondent’s decision-making varies quite drastically depending on whether or not she is in the company of the first respondent, we have real concerns that this state of affairs may be maintained with a view to inducing the third respondent into making the decisions they want her to make, without any interference from any person who might otherwise put a stop to that. In our judgment, it is more likely than not that the first and second respondents have exercised undue influence over the third respondent and will continue to do so for as long as other people are prevented from seeing her.

(iii)   Assistance from professionals

204    Following from our earlier discussion on the “practicable steps” contemplated in s 3(3) of the MCA, in considering the third respondent’s actual circumstances we should pay attention to the likelihood of her receiving assistance in making her decisions. Counsel for the third respondent argues that she has been and will continue to be given assistance from professionals such as the lawyer, Mr L, and the banker, Ms D, and that, whatever the defects in her memory or executive function, she is able to make decisions when such assistance is rendered to her.

205    We accept that the third respondent has been receiving assistance from those professionals and that this assistance is likely to remain available to her. However, the fact that such assistance was apparently made available to the third respondent at the relevant time did not prevent her acting to establish the Trust and to transfer her UBS assets to DBS. We have found that the third respondent lacked the ability to make those decisions, notwithstanding the assistance she apparently received. It follows that the “practicable steps” taken by her present professional advisors are not sufficient to help her retain her decision-making ability.

Our findings

206    Drawing together the threads of our analysis above, on a consideration of the third respondent’s observable medical symptoms, we need not and do not make a definitive finding as to whether she had MCI or dementia. We find that that she did have a mental impairment and whatever the precise diagnosis of this mental impairment might be, it has resulted in a significant deterioration of memory, a decline in executive function – in particular the ability to understand and use information and to weigh countervailing considerations against one another – and the harbouring of paranoid beliefs. This might not deprive the third respondent of her ability to live independently and to make simple decisions, but we are concerned here with relatively complex decisions concerning her property and affairs. In that connection, we find that the third respondent lacked the ability to make the decision to set up the Trust as well as the decision to transfer her UBS assets to DBS. In relation to the Trust, although we consider that she understood at least some of the information relevant to the decision to set up the Trust, we are satisfied that she lacked the ability to use and weigh the information relevant to the decision because of the decline in her executive function together with the fact that her mind was overborne by a paranoid belief caused by her mental impairment. In relation to the transfer of assets, we are satisfied that the problems with her memory and her difficulties in the area of executive function caused her to lack the ability to retain the information relevant to the decision or even to understand it. Nor was she able to use and weigh the information relevant to this decision.

207    Further, in our judgment, the actual circumstances of the third respondent’s life do not furnish her with such help as would enhance her ability to make decisions, and in fact supply positive hindrances to her decision-making independence in that she is cut off from people who would otherwise be able to give her advice and is subject to the undue influence of the first and second respondents. We are satisfied that her circumstances contribute to her inability to make decisions; in other words, she lacks capacity because of a combination of mental impairment and the circumstances in which she lives. Therefore the statutory test for lack of capacity under the MCA is met in her case.

Conclusion and orders to be made

208    In the light of our conclusions on the third respondent’s capacity in relation to specific decisions, we are satisfied that those decisions must be set aside and also that deputies should be appointed. We take the latter view because, given the third respondent’s considerable wealth, there will likely be many decisions she will have to make in the future in relation to her property and affairs that involve substantial sums of money. It does seem that there may be practical difficulties with specifying a threshold beyond which she needs a deputy, while leaving her free to act in respect of sums below that, but we will hear submissions on this before coming to a decision.

209    We will also hear the parties as to who should be appointed as her deputy or deputies. Any such person should not be in a position of conflict of interest, and should be persons who are trusted by the third respondent or who are people with whom she is not uncomfortable. We may also consider the appointment of an independent legal advisor accountable to the court and we wish to hear submissions on that too. A final point on which we would like submissions is the precise scope of the deputyship – in particular, how best it can be kept within the narrowest possible bounds.

210    We accordingly allow the appeal subject to hearing the parties on the matters set out in the preceding two paragraphs and on costs.

Postscript on procedure and practice in MCA proceedings

211    We conclude this judgment with some remarks on the procedure and practice that should apply in proceedings under the MCA. In the present case, we found several aspects of the proceedings that were unsatisfactory. The most significant one was that the third respondent effectively had her capacity assessed under cross-examination. This was not ideal; it would have been much better to have had her examined by an independent medical expert. That indeed was what the appellants sought but this was opposed by third respondent, who then applied for an order that would necessarily leave her exposed to cross-examination; and when her cross-examination did not speak well of her capacity, counsel sought to persuade the court that little or no weight should be given to it. This was unsatisfactory and it illustrates the need for a sensible approach to be taken by parties, especially in the conduct of such cases. We also found that the evidence, and in particular the medical evidence, was not often helpful in dealing with the central issues in the case. Time and costs could have been saved had the evidence been adduced in a more targeted manner.

212    In our judgment, such cases under the MCA may be better dealt with if the court were to direct the inquiry, if need be, with an assessor. To the extent that this signals a departure from the usual adversarial mode of litigation and a shift towards a more inquisitorial model, it is not unprecedented. In the English case of Cheshire West and Chester Council v P and another [2011] EWHC 1330 (Fam), Baker J said (at [52]):

… The processes of the Court of Protection are essentially inquisitorial rather than adversarial. In other words, the ambit of the litigation is determined, not by the parties, but by the court, because the function of the court is not to determine in a disinterested way a dispute brought to it by the parties, but rather, to engage in a process of assessing whether an adult is lacking in capacity, and if so, making decisions about his welfare that are in his best interests.

213    One manifestation of the inquisitorial nature of Court of Protection proceedings is the way in which the Court of Protection Rules 2007 bestow on the court wide powers to control the evidence, with r 95 providing thus:

Power of court to control evidence

95.    The court may—

(a)    control the evidence by giving directions as to—

(i)    the issues on which it requires evidence;

(ii)   the nature of the evidence which it requires to decide those issues; and

(iii)   the way in which the evidence is to be placed before the court.

(b)    use its power under this rule to exclude evidence that would otherwise be admissible;

(c)    allow or limit cross-examination; and

(d)    admit such evidence, whether written or oral, as it thinks fit.

214    We think that this is an entirely sensible approach for the courts in Singapore to adopt. In MCA proceedings, the interests of P are paramount; the court’s role is a protective one and it should not shy away from taking control of MCA proceedings and directing parties on the evidence that it requires in order to reach its decision.

215    This applies with greater force in the context of expert evidence, which can be very technical and at times difficult to grapple with, and which thus has greater potential to distract the court from the key issues. In the UK, the Court of Protection Rules expressly provide that expert evidence “shall be restricted to that which is reasonably required to resolve the proceedings” (r 121). The Practice Directions supplementing this part of the Rules – specifically Practice Direction A – provides in para 1 that “where possible, matters requiring expert evidence should be dealt with by a single expert”. Thus, no person is permitted to file expert evidence “unless the court or a practice direction permits” (r 120). Moreover, where more than one party wishes to submit expert evidence on a particular issue, the court “may direct that the evidence on that issue is to be given by one expert only”: r 130(1). If parties cannot agree on who this single expert should be, the court may either select from a list prepared by the parties or direct the manner in which the expert is to be selected: r 130(3). There are in addition rules which grant the court power to restrict or limit the content of experts’ reports.

216    The UK approach to expert evidence in these cases seems to us to be sensible. It does little good to have multiple experts traverse much the same ground, as happened to some degree in the present case. In our judgment, where P’s mental capacity is challenged and there is disagreement between parties as to the appointment of experts, P should be independently examined in consultation with her own doctor, with the court appointing the independent expert if parties are unable to agree. The results of such examination should be made available to the court and the parties, if necessary, with adequate protection of confidential information. Other experts may be called to give evidence on what the independent expert has said, and such evidence may be tested through cross-examination including the concurrent giving of evidence, also known as a witness conference or “hot-tubbing”. We think that this, coupled with a more inquisitorial and court-directed approach to the evidence, might afford a better way of dealing with cases under the MCA. We will refer this to the Rules Committee for further study and if thought fit, for implementation through the making of appropriate amendments to the Rules of Court.

---

^{[note: 1]}ROA Vol 4S, p 78.

^{[note: 2]}ROA Vol 3Z, p 297 lines 2 to 9.

^{[note: 3]}ROA Vol 3AA, p 6 line 22 to p 8 line 7.

^{[note: 4]}ROA Vol 3AA, pp 144, 146 and 147.

^{[note: 5]}ROA Vol 3X, pp 162 to 164.

^{[note: 6]}ROA Vol 3P, p 166 lines 15 to 17; p 188 lines 7 to 9.

^{[note: 7]}ROA Vol 4V, p 42 (cl 4(a)).

^{[note: 8]}ROA Vol 3EE, p 48 line 9 onwards.

^{[note: 9]}ROA Vol 3FF, pp 226 to 249.

^{[note: 10]}ROA Vol 3P, p 224, lines 1 to 4.

^{[note: 11]}ROA Vol 3R, p 86, lines 11 and 12.

^{[note: 12]}ROA Vol 3R, p 63, lines 11 to 14.

^{[note: 13]}ROA Vol 3P, pp 222 to 227.

^{[note: 14]}See R&T’s letter at ROA Vol 4V, p 184.

^{[note: 15]}ROA Vol 3S, pp 97 to 99.

^{[note: 16]}ROA Vol 3FF, pp 233 and 234.

^{[note: 17]}ROA Vol 3FF, p 273.

^{[note: 18]}ROA Vol 3FF, p 231.

^{[note: 19]}See the first appellant’s affidavit at ROA Vol 3A, pp 37 and 38.

^{[note: 20]}ROA Vol 3S, p 35 line 10.

^{[note: 21]}CK’s evidence on this is at ROA Vol 3AA, pp 246 and 247; NG’s evidence is at ROA Vol 3DD, pp 217 and 218.

^{[note: 22]}ROA Vol 3PP, p 141 at para 2.7.2.

^{[note: 23]}ROA Vol 3U, p 47, lines 22 and 23.

^{[note: 24]}Record of Appeal Vol 4G, p 245.

^{[note: 25]}Record of Appeal, Vol 3D, p 286 (paras 3.2.5 and 3.3.2 of the expert opinion of Dr Chan Kin Ming dated 8 July 2011).

^{[note: 26]}ROA Vol 3T, p 171, lines 15 to 19.

^{[note: 27]}ROA Vol 3V, p 292, lines 15 to 18.

^{[note: 28]}ROA Vol 3T, p 173.

^{[note: 29]}ROA Vol 4T, p 291.

^{[note: 30]}ROA Vol 4T, p 225.

^{[note: 31]}ROA Vol 3V, p 257, lines 20 to 23.

^{[note: 32]}ROA Vol 3V, p 215, lines 7 to 14.

^{[note: 33]}ROA Vol 3V, p 189, lines 12 to 17.

^{[note: 34]}ROA Vol 3O, p 237 line 16 to p 238 line 10.

^{[note: 35]}ROA Vol 3O, p 245 line 6 to p 247 line 23.

^{[note: 36]}ROA Vol 3O, p 263 line 16 to p 264 line 2; p 270 lines 16 to 24.

^{[note: 37]}ROA Vol 3O, p 296 lines 2 to 13; ROA Vol 3P, p 21 lines 13 to 20; p 25 lines 10 to 19.

^{[note: 38]}ROA Vol 3O, p 299 lines 16 to 20; ROA Vol 3P, p 5 lines 7 to 18.

^{[note: 39]}ROA Vol 3P, p 205 and p 214 lines 9 to 11.

^{[note: 40]}ROA Vol 3Q, p 11 and p 22 lines 4 to 8.

^{[note: 41]}ROA Vol 3Q, p 89 and p 95 lines 21 to 24.

^{[note: 42]}ROA Vol 3R, p 120, lines 14 to 16.

^{[note: 43]}ROA Vol 3Q, p 18 lines 7 to 10; p 19 lines 3 to 13.

^{[note: 44]}ROA Vol 3Q, p 99 lines 11 to 16.

^{[note: 45]}ROA Vol 3Q, p 143 line 24 to p 144 line 4.

^{[note: 46]}ROA Vol 3Q, p 250 line 21 to p 252 line 2.

^{[note: 47]}ROA Vol 3O, p 242 line 11 to p 243 line 12.

^{[note: 48]}ROA Vol 3Q, p 212 line 10 to p 213 line 18.

^{[note: 49]}ROA Vol 3S, p 77 lines 1 to 25.

^{[note: 50]}ROA Vol 3S, p 105 lines 12 to 15.

^{[note: 51]}ROA Vol 3S, p 123 lines 8 to 25.

^{[note: 52]}ROA Vol 3P, p 218 line 19 to p 219 line 24; p 225 lines 3 to 17; p 227 line 8 to p 228 line 4.

^{[note: 53]}ROA Vol 4I, p 153 lines 10 to 24.

^{[note: 54]}ROA Vol 4J, p 132, lines 21 to 25.

^{[note: 55]}ROA Vol 3Q, p 180 lines 1 to 7.

^{[note: 56]}ROA Vol 3R, p 234 lines 23 to 25.

^{[note: 57]}ROA Vol 3EE, p155 lines 18 to 24.

^{[note: 58]}ROA Vol 3DD, p 184 lines 21 to 23.

^{[note: 59]}ROA Vol 4J, p 5 line 23 to p 6 line 6; p 18 lines 10 to 15.

^{[note: 60]}ROA Vol 4I, p 226 lines 23 to 25 (Clinical interview, 5 February 2012).

^{[note: 61]}ROA Vol 3P, p 86 line 8 to p 87 line 5.

^{[note: 62]}ROA Vol 3P, p 106 lines 4 to 7.

^{[note: 63]}ROA Vol 4A, p 178.

^{[note: 64]}ROA Vol 4B, p 92.

^{[note: 65]}ROA Vol 4E, pp 226 to 228.

^{[note: 66]}ROA Vol 4F, pp 42 and 43.

^{[note: 67]}ROA Vol 4L, p 213.

^{[note: 68]}ROA Vol 4K, pp 172, 173 and 179.

^{[note: 69]}ROA Vol 4A, p 180.

^{[note: 70]}ROA Vol 3P, p 197.

^{[note: 71]}ROA Vol 3P, p 189 lines 21 to 25.

^{[note: 72]}ROA Vol 3O p 238 line 24 to p 239 line 7.

^{[note: 73]}ROA Vol 3Q, p 203 line 21 to p 204 line 9.

^{[note: 74]}ROA Vol 3S, p 98 line 19 to p 99 line 2.

^{[note: 75]}ROA Vol 4S, p 64 lines 10 to 16.

^{[note: 76]}ROA Vol 3CC, p 235 line 23 to p 236 line 3 (Transcript of 14 June 2012).

^{[note: 77]}ROA Vol 3CC, p 239 lines 5 to 23.

^{[note: 78]}ROA Vol 3O, p 180 lines 7 and 8.

^{[note: 79]}ROA Vol 4S, p 31 lines 7 to 15.

^{[note: 80]}ROA Vol 4R, p 243 line 1 to p 244 line 18.

^{[note: 81]}ROA Vol 3AA, p 76 lines 11 to 15.

^{[note: 82]}Record of Appeal Vol 4S, p 132.

^{[note: 83]}ROA Vol 3HH, p 138 line 9 to p 139 line 8.

^{[note: 84]}ROA Vol 3HH, p 144 line 13 to p 145 line 7.

^{[note: 85]}ROA Vol 4S, p 135.

^{[note: 86]}ROA Vol 3A, p 133.

^{[note: 87]}ROA Vol 3AA, pp 18 and 19.