//
© 2026 Zoe. All rights reserved.
Zoe is a legal information platform. Always consult the Federal Register of Legislation for authoritative text.
[2008] WASC 210
mixed. Grant of probate of the will only; probate of the first and second codicils refused.
The deceased made a clear will in early May 2004 and then, within a fortnight, two documents described as codicils: an 18-page boilerplate 'testamentary trust' of confusing construction and a short codicil tying the residue to that trust. Because the codicils were complex, depended on a non-existent schedule, permitted broad distributions beyond the testator's stated 'primary beneficiaries', contained provisions likely unknown to the testator (such as trustee commissions), and appear to have been prepared or procured for the applicant's benefit, the Court found 'suspicious circumstances' existed. That displaced the usual presumption that a duly executed instrument was known and approved; the applicant did not provide clear and satisfactory proof of the deceased's knowledge and approval. The will alone was admitted to probate; both codicils were refused.
AI-generated legal information, not legal advice. Zoe can make mistakes. Check the cited source, and for advice about your situation consult a qualified Australian lawyer.
1,806 words · generated 12/07/2026
1 cited instrument linked from this judgment.
Esther Helen Carpenter (born 20 February 1907) died on 14 November 2005 in her 99th year. She executed a straightforward will on 1 May 2004 leaving all her real and personal property to her daughter, Joan Margaret Gorman, who was appointed executrix [2]–[6]. Within a fortnight she executed two further documents described as codicils. The first codicil (said to be dated 10 May 2004) is an 18-page standard-form document of the kind used to create testamentary trusts, heavily reliant on a separate schedule that was not present, nominating Joan as trustee and gifting $100 to the trustee to hold on trust for broadly defined 'General Beneficiaries' [7]–[8], [16]–[21]. The second codicil (15 May 2004) was written on the second page of the will form and purported to leave certain shares to a granddaughter and to give the residue to the Carpenter Testamentary Trust formed by the first codicil [9].
Because of the deceased's advanced age, the close timing of the documents, the complexity and apparent incoherence of the first codicil, and the possibility that the applicant benefitted, the Court issued detailed requisitions seeking evidence about the deceased's knowledge and approval of the codicils, the deceased's knowledge of her assets, who arranged the documents and why, and execution details [13]. Two affidavits (by Joan and by Peter Gorman) sworn 13 April 2007 were filed on 25 January 2008; no further evidence was offered [32]–[33], [42]–[43]. On the papers, Registrar C Boyle concluded that suspicious circumstances existed as to the codicils, that the evidence did not positively establish that the deceased knew and approved the first codicil, and therefore refused probate of both codicils; the will itself was admitted to probate in common form [40]–[52].
The decision turns on the doctrine that the presumption a duly executed testamentary instrument was known and approved by the testator is displaced where 'suspicious circumstances' attend the instrument. The Court relied on the classic statement of Isaacs J in Nock v Austin, set out in the judgment, that where suspicious circumstances exist the presumption does not arise and the proponent must affirmatively prove knowledge and approval by clear and satisfactory proof; failing that, the instrument should not be admitted to probate [15], [38]. The Registrar expressly applied that doctrine here: the codicils were executed very shortly before death, were complex and apparently pointless given the prior will, and appear to have been executed at the applicant's instance for her perceived benefit, all factors that excite suspicion [13], [11]–[12], [41].
On analysis of the first codicil the Registrar found it confusing and defective: boilerplate trust terms intended to be read with a schedule were used without any schedule; definitions and clauses produced an expansive class of 'General Beneficiaries' (including many relations and charities) so that the instrument did not secure the purported benefit to the deceased's great-grandchildren and in fact allowed distribution to an almost unlimited range of persons; references in the document to 'parties' or a 'deed' were hollow because the document had not been executed as a deed and there is no evidence of authorship or legal advice [16]–[31]. In short, to give the first codicil any coherent operation required a most generous construction that assumed facts not in evidence [40].
The proponent's affidavit evidence was unhelpful. The applicant's affidavit contained cryptic assertions that the deceased knew 'in general' about family trusts and wanted to provide for great-grandchildren, but it did not reconcile that asserted understanding with the concrete content of the codicil (for example, trustee remuneration rights in cl 13) or explain why the deceased would have signed such a boilerplate unrelated to her apparent intention [43]–[49]. On that foundation the Registrar concluded the proponent did not discharge the onus of proving that the testator knew and approved the first codicil; if the first codicil is invalid, the second codicil, which depends on the trust the first purported to create, cannot operate and must also be refused [49]–[51]. The will, by contrast, was straightforward and duly executed and raised no suspicion; it was admitted to probate [11], [34]–[36].
Before this decision the governing legal principles were well established: a duly executed will ordinarily attracts the presumption of capacity and approval; but where suspicious circumstances attend the making of the instrument, the presumption of approval is displaced and the proponents must prove knowledge and approval (Nock v Austin and subsequent authorities) [15], [34]–[41]. This case applies those pre-existing principles to a fact pattern increasingly common in modern estates work: the insertion of mass-produced boilerplate trust documents or complex tax-planning instruments into testamentary documentation, especially where executed close to death and in circumstances suggesting self-interest by a beneficiary who arranged them.
After this decision the law as applied is unchanged in principle; what Re Carpenter emphasises is practical: courts will scrutinise boilerplate testamentary trust documents assembled without the schedule or legal advice, will treat execution at the instance of a beneficiary as suspicious (though not disqualifying in itself), and will refuse probate where the proponent fails to prove the testator's understanding of complex instruments. Practitioners should therefore expect that codicils or testamentary trusts prepared in a non-professional way and signed shortly before death will attract rigorous proof requirements.
Quoted doctrinal passage from Nock v Austin: "In general, where there appears no circumstance exciting suspicion... the mere proof of ... due execution ... creates an assumption that he knew of and assented to its contents" [15].
Plain English: Normally, if a will is properly signed and the testator had capacity, the court will assume the testator knew and approved what was written. But that assumption only stands if nothing about the circumstances makes the court suspicious.
Registrar's statement of concern in the requisitions: "In the light of circumstances that include the very advanced age of the deceased ... the complexity of them and their apparent pointlessness, and the fact that the deceased executed them very shortly before she died..." [13].
Plain English: The court noticed several worrying signs, extreme age, complicated and odd documents, and the close timing to death, and asked for detailed evidence because these facts raise doubts.
Registrar's conclusion on suspicion and burden: "There are circumstances that require that the deceased's knowledge and approval of the codicils cannot be assumed. It must be proved. The first codicil is complex and confused." [40].
Plain English: Because of the surrounding facts, the court will not assume the testator knew and approved. The people seeking probate must prove it, and here the first codicil was too messy to assume the testator understood it.
Disposable holding: "I am thus not persuaded that the deceased knew and approved ... For those reasons, the first codicil should not be admitted to probate." [50]; and "the second codicil cannot be admitted" [51].
Plain English: The court found the evidence insufficient and rejected the codicils; it accepted only the original will.
This decision will be engaged where:
Practitioners should expect close scrutiny and should advise clients arranging late-in-life testamentary changes to obtain independent legal advice and to provide clear evidence of the testator's instructions and understanding.
This decision is a Registrar's decision of the Supreme Court of Western Australia and applies the long-standing principles from higher authorities such as Nock v Austin. As a decision at the Registrar level, it is persuasive within WA courts for similar cases dealing with suspicious circumstances and incoherent testamentary trust instruments. The judgment is notable for its detailed textual analysis of a boilerplate testamentary trust form and for practical guidance about evidential expectations. It demonstrates that courts will not strain to find a testator's intention where the document's mechanics and affidavit evidence do not support it. Any later appellate treatment would be case-specific; practitioners should treat the reasoning as a rigorous application of established doctrine rather than a change in principle [15], [40]–[51].
The limits of permissible construction: The Registrar adopted a generous construction approach to see if the first codicil could be made to work, but concluded that even a generous approach did not achieve a sensible operation. How far should courts go to construe mass-produced boilerplate in favour of probate where a schedule is missing? This remains a contextual question but Re Carpenter signals caution and reluctance to infer precise testamentary intention where key components are absent [19]–[21], [40].
The evidential value of informal explanations: The applicant's affidavit asserted what the deceased 'knew in general' about family trusts and what she wanted. The Court found that insufficient. What degree and form of lay evidence (for example contemporaneous notes, witnessed instructions, solicitor's file notes, or video/audio evidence) will be sufficient to discharge the onus in borderline cases remains fact-specific; Re Carpenter underlines that mere assertions in self-interested affidavits are unlikely to suffice [42]–[49].
Testamentary instruments prepared by accountants or advisers: The decision critiques the provenance of the boilerplate trust document and the absence of legal competence in its assembly [23]. The interplay between accountant-prepared testamentary trust forms and legal formalities, and whether the presence of competent professional advice from a non-lawyer can assist to discharge the onus of knowledge and approval, will depend on the content and form of that advice and evidence of explanation to the testator.
Treatment of beneficiary-prepared instruments: While acknowledging that a beneficiary's involvement commonly excites suspicion, the Court reiterated it does not produce disqualification as a matter of law [15(4)]. How much corroborative evidence is required to neutralise that suspicion remains an evidential question; this decision provides a cautionary benchmark but not a bright-line rule.
Practical takeaways: when arranging late-life testamentary changes, ensure independent legal advice, document the testator's instructions and understanding contemporaneously, avoid inserting complex boilerplate without a schedule, and be alert that courts will require clear, candid evidence to rebut any suspicion that a document was not fully known and approved by the testator [13], [40]–[51].
JURISDICTION : SUPREME COURT OF WESTERN AUSTRALIA
CITATION : RE CARPENTER; EX PARTE GORMAN [2008] WASC 210
MATTER : THE ESTATE OF ESTHER HELEN CARPENTER (Dec)
Will - Codicils - Complex trusts - Whether deceased knew and approved
Property Law Act 1969 (WA), s 9
1 REGISTRAR C BOYLE: This is an application for a grant in common form of probate of a will and two codicils. The issue is whether the deceased knew and approved of the contents of the two codicils.
2 Esther Helen Carpenter died on 14 November 2005. She had been born on 20 February 1907, and so was in her 99th year at the date of her death.
3 The application made on 20 May 2006 is for probate of a will dated 1 May 2004, a first codicil said to be dated 10 May 2004, and a second codicil dated 15 May 2004. The application is made in person by Joan Margaret Gorman, who is the daughter of the deceased and the institute executrix appointed by the will.
4 The testamentary documents involved need to be described in some detail.
5 The will has been prepared using a standard form of the kind available from newsagents and stationers. It has been completed in black ink in a hand that appears not to be that of the deceased, because it bears no relation to the deceased's signature but does appear on later documents.
6 The will appoints the applicant as executrix. Its sole operative provision gives 'all my real and personal property to my daughter Joan'. The will appears to have been duly executed. The witnesses are a Marie Therese Bliss and a Peter Gorman (whose address is not the same as that of the applicant).
7 What is put forward as the first codicil is a document of a different kind.
8 It is an 18-page standard form document into which there have been inserted (in what appears to be the same hand as that of the body of the will) the various particulars required to make the document a whole, if not particularly comprehensible. Although it is described on its cover as a codicil, it appears to have been designed to function for the limited purpose of creating a testamentary trust. In this case, the testamentary operation of the instrument is to make a gift of a nominal sum to the trustee of a trust also created by it. The nominal sum is $100 and the trustee is the applicant Joan Margaret Gorman. The Trust is termed the Carpenter Testamentary Trust.
9 The second codicil has been produced in a different fashion. The original will form described above provides two pages of space for operative provisions. There is a standard form of execution clause on each page. The will described above takes up only the first page, and the second page was at that point left blank. The second codicil consists of writing inserted on the second page. The execution clause for the codicil is that provided on the standard will form. There is in principle no particular difficulty in effecting a codicil in this fashion.
10 The codicil recites that whereas the will gave the entire estate of the deceased to Joan, the deceased wishes to leave Wesfarmers shares to her granddaughter Deena and the rest and residue to the Carpenter Testamentary Trust formed by her first codicil.
11 Were the will to stand on its own, there would be no hesitation in admitting it to probate. It is on its face regularly executed and completely rational. There is nothing to displace the presumptions that the deceased had testamentary capacity, and she knew and approved of the contents of the will. However, the nature and sequence of the codicils raised questions about whether the deceased understood what these documents did.
12 The sequence is that, within the space of a fortnight, the deceased:
13 On 29 June 2006 the Court issued requisitions. Again because of the particular circumstances, it is worth setting out in full the body of those requisitions.
In the light of circumstances that include the very advanced age of the deceased when she executed the documents described as first and second codicils, the complexity of them and their apparent pointlessness, and the fact that the deceased executed them very shortly before she died, the following questions require to be addressed:
The preceding questions indicate that the history of the codicils raises serious doubt about the separate issues of the deceased's testamentary capacity and her knowledge and approval of the contents of the Wills. Those questions must be addressed by frank and complete evidence on affidavit. The applicant should not attempt to answer these requisitions without obtaining competent legal advice.
14 The answer to those requisitions was not filed until 25 January 2008. It consists of affidavits of Peter Gorman and Joan Gorman, each sworn 13 April 2007.
Knowledge and understanding; and the doctrine of suspicious circumstances
15 In Nock v Austin [1918] HCA 73; (1918) 25 CLR 519, 528, Isaacs J observed,
The relevant law is not doubtful. It may be thus stated:
(1) In general, where there appears no circumstance exciting suspicion that the provisions of the instrument may not have been fully known to and approved by the testator, the mere proof of his capacity and of the fact of due execution of the instrument creates an assumption that he knew of and assented to its contents (Barry v Butlin; Fulton v Andrew).
(2) Where any such suspicious circumstances exist, the assumption does not arise, and the proponents have the burden of removing the suspicion by proving affirmatively by clear and satisfactory proof that the testator knew and approved of the contents of the document (Baker v Batt; Tyrrell v Painton; Shama Churn Kundu v Khettromoni Dasi).
(3) If in such a case the conscience of the tribunal, whose function it is to determine the fact upon a careful and accurate consideration of all the evidence on both sides, is not judicially satisfied that the document does contain the real intention of the testator, the Court is bound to pronounce its opinion that the instrument is not entitled to probate (Baker v Batt; Fulton v Andrew).
(4) The circumstance that a party who takes a benefit wrote or prepared the will is one which should generally arouse suspicion and call for the vigilant and anxious examination by the Court of the evidence as to the testator's appreciation and approval of the contents of the will (Barry v Butlin and Fulton v Andrew; per Lord Shaw in Low v Guthrie).
(5) But the rule does not go further than requiring vigilance in seeing that the case is fully proved. It does not introduce a disqualification (per Lord James in Low v Guthrie).
(6) Nor does the rule require as a matter of law any particular species of proof to satisfy the onus (Barry v Butlin).
Citations have been omitted. For the sake of clarity, I have placed his Honour's numbered propositions in separate paragraphs; they are not tabulated in the authorised report.
16 In order to apply correctly the legal principles I have identified, it is necessary to return to the first codicil and examine it in more detail.
17 On the first page of the document the 'primary beneficiaries' are identified as the 'grand-children of Joan Margaret Gorman' and the 'specified beneficiaries' as 'Joan Margaret Gorman' or persons related to her by blood and marriage and all those identified in the 'Trust Terms'.
18 The 'Trust Terms' begin with a clause of definitions and interpretations. Various sub-clauses of that set out who are to be the beneficiaries of the trust. Sub-clause 1.8 defines 'General Beneficiaries'. They include the 'Specified Beneficiary' or the 'Specified Beneficiaries' as the case may be and various relations of those and certain other entities. In fact the definition of 'General Beneficiaries' contains four sub-sub-clauses, one of which in turn broken further into four sub-sub-sub-clauses and is followed (with no respect for the ghost of Coode) by two lengthy provisos, so that the assemblage stretches for the better part of a page. It is built around the notion of the 'Specified Beneficiary'. That term is defined in sub-clause 1.17 to mean 'the person or persons named described or defined as such in the Schedule'. There is no Schedule.
19 The Trust Terms have clearly been drafted so as to form part of a document of a type mass-produced by or at the instance of accountants for tax planning purposes. The usual arrangement of these documents is that they will contain a dozen or more pages of boilerplate text, followed by a schedule that contains variable information such as the names of the relevant parties and the holders of the various offices created by the document. The Trust Terms are heavily dependent upon there being such a schedule. For example, there are references to the 'Schedule' in cl 1.2; cl 1.8.4; cl 1.9; cl 1.17; cl 1.20; cl 11.6; cl 11.7.12; and cl 20.5.
20 It is only possible to make sense (of a kind) of this document by reading references in the Trust Terms to the 'Schedule' as in fact being references to the opening half a page or so, where it is provided, after referring to the Carpenter testamentary trust, that:
Such trust shall have the conditions as per the Trust Terms Deed and the parties to such Deed will be: -
21 Of course, despite the language used, Joan Margaret Gorman and Peter Andrew Gorman are not parties to any deed. Although Peter Gorman has signed the document at its conclusion, he has clearly done so in his capacity as witness. The document is not executed as a deed. There are clearly no parties to it in a sense that there are parties to a contractual instrument or to a deed of the kind that was once technically described as an indenture. Nor is this a deed poll. Even applying the most generous interpretation of the requirements of s 9 of the Property Law Act 1969 (WA), there is no indication that this document was executed as a deed.
22 I will return to the question of exactly what this instrument achieves when I consider the evidence that is offered in support of the proposition that the deceased knew and approved of its contents.
23 There is no indication on this document of its authorship. It is understandable there might be nobody willing to assert such a claim. Experience and a detailed analysis of the kind commenced above lead to the conclusion that, while individual components of it may have some respectable provenance, the ultimate assemblage is the product of someone with no legal qualifications, or at least no legal competence.
24 If the document is to be construed making the generous concession that references to offices whose occupants were said to be set out in the non-existent Schedule were references to those positions and persons identified in the preliminary parts of the document, the next step is to consider whether the benefits were for the great-grandchildren of the deceased. That is a question that requires consideration of the eligibility for distribution of income and capital under the proposed trust.
25 The principal provision dealing with distribution of the income of the trust fund is cl 3 of the Trust Terms. This requires the trustee, at the conclusion of each accounting period, to determine either to distribute the income 'for any one or more of the General Beneficiaries', or otherwise to accumulate it.
26 The term 'General Beneficiaries' is defined, as already mentioned, in cl 1.8. It includes the specified beneficiaries: cl 1.8.1. The specified beneficiaries, making the concessions indicated above, are Joan Margaret Gorman 'or persons related to her by blood or marriage and all those identified in the Trust Terms'.
27 It is thus immediately apparent that the distribution of income under the trust is not limited to the great-grandchildren of the deceased, nor of course does it guarantee that they would receive anything.
28 Clause 1.8.2 further widens the class of general beneficiaries to include,
the parents step-parents grandparents brothers sisters spouses de facto spouses widows widowers children grandchildren step-children uncles aunts and cousins and any relatives by blood or marriage or de facto marriage of the specified beneficiaries or the parents of the specified beneficiaries.
29 It is to be remembered that the only specified beneficiary is Joan Margaret Gorman. Sub-paragraphs 1.8.3 and 1.8.4 further widen the class of general beneficiaries.
30 If, as paragraph 1(b) of the supplementary affidavit of the applicant has it, it was the intention of the deceased by the first codicil to provide a mechanism for benefiting her great-grandchildren, subject to the exercise of some control by the applicant, then the codicil did not achieve that aim. It neither ensures benefits for the great-grandchildren nor restricts benefits to them. The boilerplate definitions allow distributions of income to an almost unlimited range of natural and legal persons, provided only that they have some connection with the trustee, or are charities (cl 1.8.3.4 provides for distribution to charities).
31 I think the analysis of the document need be carried no further. However, I would add one more point. The opening part of the codicil, before the commencement of the trust terms, identifies as 'primary beneficiaries' the grandchildren of Joan Margaret Gorman. However, the term 'primary beneficiaries' does not appear anywhere in the trust terms. A reader (such as, for example, the testator) who assumed from the mention of the grandchildren of the applicant as primary beneficiaries that the instrument secured some benefit for them would be misled.
32 Because of the contents of those affidavits a further requisition was issued on 5 February 2008 in the following terms:
Are the affidavits of Joan Margaret Gorman and Peter Gorman both sworn 13 April 2007 but filed 25 January 2008 all that is to be offered in reply to the requisitions of 29 June 2006?
33 On 28 February 2008, the Court received a letter from the applicant advising that there was no further evidence to offer. The application therefore falls to be considered on the evidence to date, without the benefit of submissions.
34 In order for a will or codicil to be admitted to probate, it must be shown that the testator had testamentary capacity, and that she knew and approved of the content of the testamentary instrument. Those are two distinct tests that are not to be conflated, although there can at times be a relationship between them: Hoff v Atherton [2004] EWCA Civ 1554.
35 Testamentary capacity will usually be presumed from due execution: Browne v McElhone (1885) 15 LR (NSW) B&P 154; Bull v Fulton [1942] HCA 13; (1942) 66 CLR 295; Western Australian Trustee Executor and Agency Co Ltd v Holmes [1961] WAR 144.
36 In certain circumstances, knowledge and approval may also be presumed. As Hannen P observed in Parker v Felgate (1883) 8 PD 173:
If a person has given instructions to a solicitor to make a Will, and the solicitor prepares it in accordance with those instructions, all that is necessary to make it a good Will, if executed by the testator, is that he should be able to think thus far, 'I gave my solicitor instructions to prepare a Will making a certain disposition of my property. I have no doubt that he has given effect to my intention and I accept the document which is put before me has carrying it out'.
37 That is a presumption of law that operates in particular circumstances that do not apply here. In this case, there is no suggestion that the will or either of the codicils was professionally prepared.
38 I have already referred to - but it is worth setting out in full - the passage in Nock v Austin (528), in which Isaacs J observed:
In general, where there appears no circumstance exciting suspicion that the provisions of the instrument may not have been fully known to and approved by the testator, the mere proof of his capacity and of the fact the due execution of the instrument creates assumption that he knew of and assented to its contents.
39 It is the qualification at the beginning of that passage that is critical here. The question is whether there are suspicious circumstances, as that term is to be understood in light of the case law, so as to trigger a requirement that knowledge and approval must be established positively rather than just being presumed.
40 In my view, there are circumstances that require that the deceased's knowledge and approval of the codicils cannot be assumed. It must be proved. The first codicil is complex and confused. To give it any operation at all requires a most generous approach to its construction.
41 The deceased received no competent advice to explain it. It does not do what the applicant swears the deceased believed it to do. It appears to have been executed at the instance of the applicant and for her perceived benefit.
The evidence of the deceased's knowledge and approval
42 In response to a requisition, two affidavits were filed. This first is that of Peter Gorman, sworn 13 April 2007. It provides no assistance in discerning whether the deceased knew and approved of the contents of the first codicil.
43 The other affidavit, sworn the same date, is that of the applicant. The first three paragraphs of this affidavit set out her evidence of the deceased's knowledge and approval of the contents of the first codicil:
44 While needless verbosity is a frequent vice in affidavits, this errs in the other direction: it surpasses the merely laconic and attains the cryptic.
45 It is impossible to decipher what paragraph 1(a) of the affidavit means, and therefore whether it has any relevance to the question under consideration.
46 It is paragraph 1 (b) that comes closest to an explanation of what the deceased may have understood the effect of the first codicil to be. The terms of the document have to be examined in light of the proposition that she thought she was enabling her great-grandchildren to benefit, 'provided that [the applicant] was able to exert some manner of control of it'.
47 Finally, there is I think a further significant point. That is, that cl 13 provides for the trustee to charge and retain reasonable commissions out of the trust fund. That is a significant right would not have accrued to the applicant in her capacity as executor, and there is no evidence the deceased was aware of the provision.
48 As with cl 1(a) I find cl 1(c) to be of no assistance in addressing the deceased's knowledge and approval of the content of the codicil. In fact, I find it difficult to make any sense of it at all.
49 I am thus not persuaded that the deceased knew and approved of the content of the first codicil.
50 For those reasons, the first codicil should not be admitted to probate.
51 It follows, I think, that the second codicil cannot be admitted. Its operation must have depended upon assumptions by the deceased that were not sound.
52 There will be a grant of probate of the will only. The application is refused in relation to the codicils.
(1918) 25 CLR 519
(1942) 66 CLR 295
Grant of probate of the will only; probate of the first and second codicils refused.
(7) The doctrine that suspicion must be cleared away does not create 'a screen' behind which fraud or dishonesty may be relied on without distinctly charging it (Lord Loreburn L.C. in Low v Guthrie).