---

CASE NO 730/91
/ccc

IN THE SUPREME COURT OF SOUTH AFRICA

(APPELLATE DIVISION)

In the matter between:
GREGORY DAVID HARPUR N.O. APPELLANT

and

GOVINDAMALL
and
THE MASTER OF THE SUPREME COURT RESPONDENTS

CORAM: CORBETT CJ, VAN HEERDEN, EKSTEEN JJA et NICHOLAS, KRIEGLER AJJA

DATE HEARD: 14 MAY 1993

DATE DELIVERED: 6 SEPTEMBER 1993

JUDGMENT NICHOLAS, AJA:

This is an appeal against a judgment given by
2

Howard JP in application proceedings in the Durban and
Coast Local Division. The applicant was one
Govindamall, the widow of the late Perumal Pillay ("the
testator") who died on 26 May 1986. She sought an order
declaring invalid the will executed by the testator on
25 January 1985 in terms of which the sole beneficiaries
were his five children. She cited as first, second and
third respondents respectively the three major children,
as fourth respondent Mr G D Harpur in his capacity as
curator ad litem to the two minor children and as fifth
respondent the Master of the Supreme Court. Howard JP
granted an order in the following terms:

"(a) The will of the late Perumal Pillay dated 25 January 1985 is declared to be invalid. (b) The costs of all parties, including the curator ad litem, shall be paid out of the estate of the late Perumal Pillay No 5132/86 on the scale as between attorney and client."

Leave was granted to appeal to this court but only the

3

curator ad litem pursued the appeal.

The will read as follows:

"LAST WILL AND TESTAMENT
This is the Last Will and Testament of Me PERUMAL PILLAY (BORN 5th AUGUST 1930), married by Antenuptial Contract to GOVINDAMALL, presently of 42 Dubrugarth Road Merebank, in the Natal Province of the Republic of South Africa.
1. I hereby revoke, cancel and annul all previous Wills, Codicils and other testamentary writings, heretofore made or executed by me.
2. I give and bequeath my entire Estate including my immovable property wheresoever situated to my five children, in equal shares, share and share alike:-

1. KUMARI MUNSAMI
2. PARIMALA REDDY
3. ENDRASEN PILLAY
4. RONALD PILLAY
5. NISHALIN PILLAY

3. I nominate, constitute and appoint my daughter KUMARI MUNSAMI, to be the Executor of this My Will and the Administrator of my Estate and Effects, giving and granting unto her all such Powers as are allowed by Law and especially the Power of assumption.
4. I hereby direct that my Executor and Administrator

4

either appointed or assumed, shall not be required to file security with the Master of the Supreme Court or any other official for the due fulfillment of their duties.
5. IN WITNESS WHEREOF I HAVE hereto set my hand at Durban this 25/1/85 day of Jan 1985 in the year of our Lord, One Thousand Nine Hundred and Eighty Two (1982) [sic] in the presence of subscribing witnesses.
SIGNED BY PERUMAL PILLAY,
the Testator of this Last Will
and Testament in the presence
of us who in his presence
and in the presence of each
other, all being present
at the same time have affixed
our signatures hereto as Witnesses.
[SIGNED] P PILLAY TESTATOR WITNESSES:
1. [SIGNED] S R PILLAY
2. [SIGNED] SOOBRAMONEY"

The will consisted of two pages, the first of which ended with para 4. The second page was signed by P Pillay as testator and by S R Pillay and Soobramoney as witnesses. The full signature P Pillay appears three

5

times on the first page: against an alteration in one of the names in clause 2, and on the left and right sides at the foot of the page. Next to each of these signatures appear what are said to be the initials SRP for S R Pillay and the single letter S for Soobramoney.
The applicant submitted that the will was invalid in that it did not comply with para (iii) read with para (iv) of s 2(1)(a) of the Wills Act, 7 of 1953 ("the 1953 Act").

The 1953 Act was the culmination of a long

process of development. See the discussion by Prof B

Beinart in an article published in 70(1953) SAW 159 and

entitled Testamentary Form and Capacity and the Wills

Act, 1953. He stated (at 159):

"Our law relating to the execution of wills has retained many forms which are relics of its Roman and Dutch past. Legislation in the various Provinces has added or partly substituted the English forms of will. As regards the older forms

6

of will each provincial statute has manifested a different attitude. The result has been rather a jumbled pattern, and the law has long stood in need of clarification, revision, simplification and uniformity."

The 1953 Act recites that it is an act to consolidate and amend the law relating to the execution of wills. It repealed in whole or in part the pre-Union statutes of the former colonies, including the Cape Execution of Wills Ordinance No 15 of 1845, the Natal Execution of Wills and Codicils Law No 2 of 1868, the Orange River Colony Execution of Wills and other Testamentary Instruments Ordinance No 11 of 1904, and the Transvaal Wills Ordinance No 14 of 1903.

The Cape Ordinance was modelled on s 9 of the

English Wills Act, 1837, which provided

"No will shall be valid unless it shall be in writing and executed in manner hereinafter mentioned; (that is to say), it shall be signed at the foot or end thereof by the testator, or by some

7

other person in his presence and by his direction, and such signature shall be made or acknowledged by the testator in the presence of two or more witnesses present at the same time, and such witnesses shall attest and shall subscribe the will in the presence of the testator, but no form of attestation shall be necessary."

So far as it is relevant for present purpose, sec 3 of

the Cape Ordinance read:

"3. And be it enacted that no will or other testamentary writing...shall be valid unless it shall be or shall have been executed in the manner hereinafter mentioned: that is to say, it shall be or shall have been signed at the foot or end thereof...by the testator or by some other person in his presence and by his direction...and such signature shall be or shall have been made or acknowledged by the testator...in the presence of two or more competent witnesses present at the same time, and such witnesses shall attest and subscribe or shall have attested and subscribed the will in the presence of the person executing the same; and where the instrument shall be or shall have been written upon more leaves than one the party executing the same and also the witnesses shall sign or shall have signed their names upon at least one side of every leaf upon which the instrument shall be or shall have been written."

8

It seems that Natal Law 2 of 1868 in turn was modelled cm the Cape Ordinance. Sec 1 provided, however, "that nothing herein contained shall be deemed to prevent a mark being a signing for the purposes hereof".

The Transvaal Wills Ordinance 14 of 1903 also appears to have been modelled on the Cape Ordinance, but it too provided in s 1 that "nothing herein contained shall be deemed to prevent a mark being a sufficient signature".

Sec 1 of the Orange River Colony statute was

in similar terms, except that it provided that -

"The signature of a testator to a will or other testamentary instrument shall be valid whether made by way of a signature or of mark, provided only that in the latter case the mark shall be made in the presence of and attested by a Justice of the Peace and two witnesses."

It will be seen that this Ordinance differed from the

9

laws of Natal and the Transvaal in two important
respects: only the signature of a testator (and not
that of a witness) was valid if made by way of a mark,
and it was then valid only if there was compliance with
the proviso.

The 1953 Act deals in s 2(1) (a) with the

formalities required in the execution of a will.
As amended by s 20 of the Act No 80 of 1964, but before
amendment by Act No 43 of 1992, s 2(1)(a) provided -

"2.(1) Subject to the provisions of section three [which is not here relevant] -

(a) no will executed on or after the first day of January, 1954, shall be valid unless -(i) the will is signed at the end thereof by the testator or by some other person in his presence and by his direction; and (ii) such signature is made by the testator or by such other person or is acknowledged by the testator and, if made by such other person, also by such other person, in the presence of two or more competent witnesses present at the same time; and (iii) such witnesses attest and sign the will in the presence of the testator and of

10

each other and, if the will is signed by such other person, in the presence also of such other person; and (iv) if the will consists of more than one page, each page other than the page on which it ends, is also so signed by the testator or by such other person and by such witnesses anywhere on the page; and (v) if the will is signed by the testator by the making of a mark or by some other person in the presence and by the direction of the testator, a magistrate, justice of the peace, commissioner of oaths or notary public certifies at the end thereof that he has satisfied himself as to the identity of the testator and that the will so signed is the will of the testator, and if the will consists of more than one page, each page other than the page on which it ends, is also signed, anywhere on the page, by the magistrate, justice of the peace, commissioner of oaths or notary public who so certifies."

Sec 1 of the 1953 Act provided in para (iv) before

amendment that -

"(iv) 'sign' includes in the case of a testator the making of a mark but does not include the making of a mark in the case of a

11

witness, and 'signature' has a corresponding meaning;"

This definition affirms that a testator may sign a will

by way of a mark, but the legislature followed the model
of the Orange River Colony statute by providing in s
2(1) (a) (v) for safeguards to ensure that the mark was
genuinely that of the testator. The definition further
restricts the meaning of the word "sign" in the case of

a witness. For the rest the words "sign" and

"signature", which are not technical or legal terms,
must be given their ordinary, popular meaning.
The short question for decision in this appeal is whether initialling by a witness is a signing for the purposes of paras (iii) and (iv) of s 2(1) (a) of the 1953 Act.

In ordinary usage the word signature, used without qualification, means signature by name or

12

signature by mark. That was said in Goodman v J Eban
Ltd [1954] 1 Q.B. 550 (CA). The question there was
whether a signature by means of a rubber stamp was a
good signature for the purposes of s 65(2) of the

Solicitors Act, 1932. In his dissenting judgment
Denning LJ said at 561:

"In modern English usage, when a document is required to be 'signed by' someone, that means that he must write his name with his own hand upon it. It is said that he can in law 'sign' the document by using a rubber stamp with a fascimile signature. I do not think this is correct... [A facsimile] is the verisimilitude of his signature but it is not his signature in fact.
If a man cannot write his own name, then he can 'sign' the document by making his mark, which is usually the sign of a cross.."

Evershed MR (with whose judgment Romer LJ concurred)

said at 555:

"...I confess that, if the matter were res Integra, I should be disposed to think, as a matter of common sense and of the ordinary use of language, that when Parliament required that the bill should

13

be 'signed' by the solicitor, it was intended that the solicitor should personally 'sign' the bill or letter in the ordinary way by writing his name (or, where appropriate, the name of his firm) in his own hand with a pen or pencil."

He went on to add, however, that the matter was not free
from authority. He referred to a number of decided
cases, and concluded (at 557) that -

"In my judgment, therefore, it must be taken as established from the citations which I have made, that where an Act of Parliament requires that any particular document be 'signed' by a person, then, prima facie, the requirement of the Act is satisfied if the person himself places upon the document an engraved representation of his signature by means of a rubber stamp."

The importance of Goodman's case for present

purposes lies in the recognition in both judgments that

the ordinary, popular meaning of the verb sign is sign
by name or sign by mark. That accords with the relevant

definitions given in the Shorter Oxford English

Dictionary:

s.v. Sign

14

"2. To place some distinguishing mark upon (a thing or person)."
"4. To attest or confirm by adding one's signature; to affix one's name to (a document, etc)."

"6. To write or inscribe (one's name) as a signature." s.v. Signature

"1. The name of a person written with his or her own hand as an authentication of some document or writing."

Black's Law Dictionary 5th ed, p 1239, gives as the

definition of Sign,

"To affix one's name to a writing or instrument, for the purpose of authenticating or executing it, or to give it effect as one's act. To attach a name or cause it to be attached to a writing by any of the known methods of impressing a name on paper. To affix a signature to... To make any mark, as upon a document, in token of knowledge, approval, acceptance, or obligation"

of Signature,

"The act of putting one' s name at the end of an instrument to attest its validity; the name thus written... And whatever mark, symbol or device one may choose to employ as representative of himself is sufficient"

and of Mark

15

"A character usually in the form of a cross, made as a substitute for his signature by a person who cannot write, in executing a conveyance, will or other legal document."

The dictionary definitions are reflected in

the cases. In Hindmarsh v Charlton (1861) 8 H.L. Cas.
160[1861] EngR 382; , (11 ER 388) Lord Campbell L.C. observed at 167:

"I will lay down this as to my notion of the law: that to make a valid subscription of a witness there must either be the name or some mark which is intended to represent the name."

In the same case Lord Chelmsford said at 171:

"The subscription must mean such a signature as is descriptive of the witness, whether by a mark or by initials, or by writing the full name.."

I do not apprehend that Lord Chelmsford was here saying
that there were three categories of signature, viz by
name, by mark and by initials. As will appear, the
trend of the English decisions is that initials are a
form of mark. In the case of In the goods of Chalcraft,

16

deceased [1948] P. 222, Wilmer J said at 233:

"It seems to me that one ought to give a broad interpretation to the words used by the Lord Chancellor in the case of Hindmarsh v Charlton in the passage which I have read. There must either be the name or some mark which is intended to represent the name."

In the case of In re Colling, deceased, (1972) 1 WLR

1440 (Ch.D.) at 1442, Ungoed-Thomas J described the last
sentence as "the crucial sentence".
Although they may be used to identify the person affixing them, initials are not a signing in the ordinary sense of the word. The Shorter Oxford English dictionary gives under the noun Initial "B2. An initial letter; esp. (in pl) the initial letters of a person's name and surname" and under the verb Initial "v...trans. To mark or sign with initials, to put one's initials to or upon."

In the Afrikaans version of the 1953 Act, the
17

words "onderteken" and "handtekening" are used for "sign" and "signature". In Afrikaans usage "handtekening" is distinguished from "paraaf". HAT gives under "paraaf", "Handtekening met voorletters, o.m. om veranderinge of byvoegings in 'n dokument te waarmerk." Van der Merwe, Die Korrekte Woord, gives under "paraaf": "Wanneer jy parafeer, is jou paraaf jou naamtekening, maar eintlik net met die voorletters, want as jy ten voile teken, praat ons van jou handtekening of naamtekening." Die Afrikaanse Woordeboek gives under "handtekening", "Iem. se naam, met sy eie hand geskryf, dikw. as waarmerk om te bewys dat die geskrif wat daaraan voorafgaan van horn is of met sy sienswyse, begeertes of bedoelings in ooreenstemming is; eiehandige ondertekening, naamtekening; outograaf." Cf. Van Dale, Groot Woordenboek der Nederlandse Taal, which gives under "paraaf", "2. handtekening door middel der

18

beginletters, inz. ter waarmerking van inlassingen of bijvvoegingen in akten", and under "ondertekenen" "...zijn naam zetten onder, iets met zijn handtekening bekrachtigen".
For purposes of identification, authentication, or execution etc, in practice either a signature or initials may sometimes be used. That does not mean, however, that initials are a signature in the ordinary usage of the word. In the ordinary use of language the words "sign" and "signature" do not, in addition to their primary meaning, signify "initial and "initials". In his contribution on the Law of Succession in the 1981 Annual Survey of SA Law, Mr Ian B Murray (who was an eminent attorney and for many years a distinguished contributor to the South African Law Journal and the Annual Survey of South African Law - see the preface to the 1988 Annual Survey), pointed out at 288 that

19

initials and signatures not only differ in structure,

"but are regarded, both in business and in legal practice, as performing different functions. Initialling is the writing by a person of the first letter of one or more of his forenames and the first letter of his surname. Signing, in contradistinction, is (a) the indication by a person of one or more of his forenames either by writing its first letter or writing it out in full (or a recognized abbreviation of it, as 'Geo' for 'George'), and (b) (most importantly) the writing of the letters constituting his surname, or, at any rate, a writing or flourish representing or intended by him to represent the letters making up his surname - even although, in the words of John C Tarr Good Handwriting (Pan Books 1957) 7-8: 'Many signatures are an indecipherable scrawl...' Documents are normally initialled by the parties (often being also initialled by the witnesses) on all pages except the last, which is signed by the parties and the witnesses... In a notarial deed the notary initials each page and signs the last one. When it enacted the Wills Act the legislature must have been well aware of the difference between initials and signatures, and the practice regarding documents other than wills..."

No doubt there are cases where what purports

to be a signature is illegible, and it may sometimes be
difficult to decide ex facie a document whether a

20

signing is a signature or initials, but any practical difficulties which may arise do not affect the principle.
If the noun "signature", like the verb "sign", is ambiguous inasmuch as it may mean signature by name, or it may also comprehend signature by initials, recourse should be had, in order to resolve the ambiguity, to the object and policy of the Act. Cf Steyn, Die Uitleg van Wette, 5e uitgawe, 22-24.

In Erasmus v Erasmus' Guardians and Executors

1903 TS 843, Innes CJ said at 851:-

"Our law, like that of every civilised state, is very jealous that satisfactory proof of its genuineness should be given by anybody who files what purports to be the testamentary disposition of a deceased person. And the lines on which vigilance is generally exercised are in the direction of strict regulations with regard to the attestation of testamentary instruments."

And in Ex parte Sewnanden: In re Estate Poolbussia

21

1948(1) SA 539(D & CLD) Selke J, after referring to
the fact that at least in the earlier Roman Law the
prescribed rules and formalities seem to have been
concerned as much with religious as with other objects,
said at 543-5:

"But it is hardly conceivable that, at a very early stage, it was not recognised that the making of provisions of the post mortem disposition of a man's property furnished exceptional opportunities for chicanery and fraud. At all events, it is apparent that in the course of time, the religious significance of the matter tended to fade more and more into the background, and that the rules and formalities became more and more directed to curtailing opportunities for malpractice and fraud, and to securing that, so far as possible, the will reflected the genuine and freely made dispositions of the testator... Thus, it seems, the formalities enjoined by Law 2 of 1868, represent the precautions considered necessary and adequate to protect the testator, and to safeguard the validity of his dispositions."

The requirement for signatures of witnesses to

a will provides a main safeguard against the

perpetration of frauds, uncertainty and speculation.

22

Disputes regarding the validity of a will can arise only after the death of a testator, which may occur many years after it was executed. Ordinarily the only persons other than the testator who are likely to have knowledge of the circumstances of the execution of a will are the witnesses who, being present, personally saw or perceived it, and can testify in that regard. That purpose fails when the witnesses cannot be identified. It may be impossible to identify a witness who has signed by initials only. In the present case, if the signature of Soobramoney had been the letter S as written on the first page of the will or the signature of S R Pillay had been the hieroglyph on the first page, it seems clear that there would have been difficulty in identifying these witnesses.

The virtue of a signature lies in the fact that no two persons have the same handwriting, with the

23

result that signatures are difficult to forge. Initials, by contrast, can often, with a little practice, be readily and convincingly copied. In my opinion, therefore, if there is a doubt as to the meaning of the word "sign" it should be interpreted so as to exclude signing by initials.
The time-honoured form of a mark is a cross, but it may be some other character, or may even be a description, if it is clear that the signatory intended it to be his signature, that is, a substitute for his name. The Irish case of In the goods of Kieran, deceased [1933] I.R. 222 was cited in Chalcraft (supra) at 231. The facts were that the testator, who was in bed very ill, tried to write his name but did not succeed in doing more than write two more or less indecipherable initials. In answer to a question by his solicitor who was present, the testator accepted what he

24

had written as his mark, and it was so endorsed on the will by the solicitor, and was so attested by the witnesses. In that case it was decided that although the mark did not take the usual form of a cross, nevertheless it was a mark acknowledged by the testator as his own in the presence of witnesses and, therefore, sufficient to amount to a signature under the Act. In the case of In the Estate of Cook (deceased) (1960) 1 All ER 689 (Probate Divorce and Admiralty) a testatrix had drawn a holograph will which was duly attested by two competent witnesses. After making certain dispositions of her property, she concluded: "Please Leslie be kind to Dot. Your loving mother". ("Leslie" was her son and "Dot" referred to one of her daughters.) In his judgment Collingwood J referred to cases which laid down that the making of a mark was a sufficient signing under the Statute of Frauds. He also referred

25

at 691 to Hindmarsh v Charlton (supra) and to In the
goods of Redding (otherwise Higgins) (1850) Rob Eccl
339. In the latter case probate was granted of a will
executed by a testatrix under an assumed name, as the
court considered that the assumed name might be regarded
as the mark of the testatrix. Collingwood J concluded
that, applying those principles to the case before him,
he was satisfied that the words "Your loving mother"
were meant to represent the name of Emma Edith Cook, the
testatrix, and he accordingly held that the will had
been properly executed.

The rationale for recognizing initials as a

signature is that they are, or are in the nature of, a
mark. See Jarman on Wills, 8th ed, Vol I p 126:

"Signature by Testator:
The next condition prescribed for the validity of a will is that it should be signed, which suggests the inquiry what amounts to a 'signing' by the testator. It has been decided that a mark is sufficient, even if the testator is able to

26

write...A mark being sufficient, of course the initials of the testator's name would also suffice..."

and p.134

"VII Attestations and Subscription by
Witnesses:
A mark has been decided to be sufficient
subscription... The initials of the witnesses also
amount to a sufficient subscription, if placed for
their signatures, as attesting the execution..."

According to Halsbury's Laws of England, 4th ed, Vol 50,

para 265, p 136:

"To make a valid attestation a witness must either write his name or make some mark intended to represent his name. A will may be subscribed by marks even though the witnesses are capable of writing. The initials of an attesting witness may be sufficient, unless placed on the will merely for the purpose of identifying alterations."

In the case of In the Goods of Blewitt 5 P.D. 116 the

question was whether the signature and subscription by

initials only were sufficient. The President said (at

117):

"A mark is sufficient though the testator can

27

write: Baker v Dening.
Initials, if intended to represent the name, must be equally good... In Christian's Case the initials of the witnesses were held sufficient, although if merely placed to attest the alteration they will not serve as an attestation of the will itself: Re Martin, deceased."

Jarman (op cit at 134) states that a mark by an attesting witness is a sufficient subscription, but adds that "it is never advisable, where it can be avoided (and, now that the art of writing is so common, seldom necessary), to employ marksmen as witnesses."

The matter has received judicial attention in South Africa.
In Van Vuuren v Van Vuuren (1854) 2 Searle 116, a will was written upon two leaves (that is, four pages covered with writing). The testator and witnesses signed their names at the bottom of the will and wrote their initials on the first leaf, except that one of the witnesses wrote his signature in full on the first leaf.

28

Wylde CJ and Ebden J felt constrained to hold that the solemnities required by law were wanting. Bell J dissented in a carefully considered judgment. In regard to the word "signed" as used in the 1845 Ordinance he said that it was impossible to avoid giving to it a popular interpretation. At 122 he cited inter alia Baker v Deninq (1838) 8 A and E 94 [1838] EngR 548; (112 ER 771) and Harrison v Harrison (1803) 8 Vesey 185 [1803] EngR 504; (32 ER 324) as showing that a will attested by the mark of a witness was validly attested and subscribed. He said at 124 that if a mark will do, he did not see why initials should not be sufficient either at the end or on the other leaves of the will.

The view of Bell J prevailed in In re Trollip (1895) 12 S.C. 243, which overruled the decision in Van Vuuren's case. The testatrix and witnesses duly signed at the foot of a will written upon more pages than one,

29

but the only signature of the testatrix to the first

leaf was made by means of her initials just above the

initials of the witnesses. De Villiers CJ said at 246
that the only question was whether the requirement of
the 1845 Ordinance in regard to signing was complied
with by a signature by means of initials. He said that

the requirement was that the testator and witnesses

should "sign", not write, their names.

"What is the original meaning of the term 'sign'? It is a 'mark' from the latin signum. To sign one's name, as distinguished from writing one's name in full, is to make such a mark as will represent the name of the person signing the document. For that purpose it is no more necessary to write one's surname in full than it is to write one's Christian names in full."

The decision in Trollip accorded with the

English law as it had been developed: the initials

constituted a mark which represented the names of the
person who signed the will.

30

That decision would not, however, be authoritative in a case to be decided under the 1953 Act, which provides that "sign" includes in the case of a testator the making of a mark, but does not include the making of a mark in the case of a witness.
In Dempers & Others v The Master and Others (1) 1977(4) SA 44 (SWA), a will consisting of four pages had been signed by the testator at the end and at the foot of each page, but the witnesses had signed only the last page; at the foot of the other pages they had placed their initials. The question arose whether the witnesses to the will attested and signed it in accordance with the requirements of paras (iii) and (iv) of the 1953 Act. After quoting dictionary definitions, Hart AJP said at 50A that there was a significant difference between the meaning of "to sign" and of "to initial", as also between the meaning of "signature" and

31

"initials". At 56 he found that the answer to "this

perplexing problem"

"...lies in the fact that nor the initials per se of the witnesses do not constitute their ordinary or normal signatures which appear clearly at the end of the will... from which it would logically follow that they did not 'sign' or affix their 'signatures' at any place on any of the first four pages of the will."

This reasoning is fallacious. What followed from the
premise was that the witnesses did not affix their
ordinary or customary signatures on the first four pages
of the will. Furthermore, the premise was false. The

1953 Act does not require that witnesses should sign
with their ordinary or customary signatures, nor does it

require that they should sign in the same way on every
page. (Compare Jhajhbai & Others v The Master & Another

1971(2) SA 370 (D & CLD), where on the first page the
witnesses signed their normal signatures and on the

second page each of them printed his name after the

32

attestation clause because, they said on affidavit, the
testator asked them to do so in order that their names
might be clearly legible). Hart AJP did not decide the
real question, which was whether the subscribing of

initials constituted a signature within the meaning of
the 1953 Act.

In Ex parte Singh 1981(1) SA 793(W) Vermooten

J decided that where a testator or witness signs a will
with his initials only, that is a sufficient compliance
with the requirements of s 2(1)(a)(iv) of the 1953 Act.

In reaching this conclusion, the learned judge relied on

Trollip (supra), saying (at 796 C-D) that there is a

remarkable similarity between the Cape Ordinance, the

1953 Act and the English Wills Act of 1837. He said (at

798 E) that the English decided cases were of

assistance, and concluded by saying -

"In view of the judicial authority both here and in England with which I respectfully agree, I think

33

it can now be stated that where a testator or a witness signs with his initials only, then that will be sufficient compliance with the requirements of the Wills Act 7 of 1953 to sign the will provided he intends it to be his signature."

No doubt English decisions do provide assistance in the
solution of problems arising under the 1953 Act, but
they should be applied with discrimination, because in
South African law, differing in this respect as in other
respects from English law, "sign" does not include the
making of a mark in the case of a witness. In my

respectful opinion, therefore, the conclusion of
Vermooten J was wrong.

The next case is that of Mellvill & Another

NNO v The Master & Others 1984(3) SA 387(C). There

Friedman J (Fagan J concurring) held, after a careful,

comprehensive and critical review of the authorities,

that when the 1953 Act requires a testator and the
witnesses to "sign" the will, what is required is a

34

signature and not initials.

The judgment in Melvill did not persuade

Vermooten J that he had been wrong in Singh. In Ex
parte Jackson NO: in re Estate Miller 1991(2) SA 586(W),
he said at 589 E-I:

"In a recent decision the Cape Provincial Division had occasion to consider the legal issue which also arises in the present case. In Mellvill and Another NNO v The Master and Others 1984(3) SA 387 (C) Friedman J (Fagan J concurring) held that signatures cannot be constituted by initials. In arriving at his decision, Friedman J who gave the judgment, does not regard himself bound by the English cases on the point, nor by the pre-Union cases decided in respect of the pre-Union Cape Ordinance, which were decided when the relevant legislation did not contain the present distinction between a 'signature' and a 'mark'. Thereupon the learned Judge comes to the following conclusion at 396D:

'It is accordingly no longer necessary, as it
was before the Act was passed, and as it was
under the Cape Ordinance and still is under
the English Act, to construe a signature
widely so as to incorporate within its ambit a
mark or initials.'
With great respect to the learned Judge the
conclusion sought by him, in my opinion, does not
follow. The word 'signature' should be construed
35

as widely as the common law prescribes, except insofar as the Legislature expressly restricted such a construction. In s 1 of the Act the Legislature chose to legislate about 'marks' and chose not to legislate about writings on wills other than marks. Such other writings include initials, which are not marks. These writings must be interpreted in terms of the common law applying to such writing before the introduction of the Act."

I do not understand the learned judge's reference to the

common law. As pointed out above, the word "signature"

does not bear a technical or legal meaning, but must be

interpreted in its ordinary, popular sense. In that
sense there are two ways of signing: signing by writing
one's name and signing by making one's mark. To the
discussion on this point which appears earlier in this

judgment may be added a reference to Van Niekerk v Smit

& Others 1952(3) SA 17(T), which was cited in Jackson at

588 C. This was not a wills case. Murray J said at 25

D-E:

"Signature does not necessarily mean writing a

36

person's Christian and surname but any mark which identifies it as the act 'of the party' - Morton v Copeland [1855] EngR 539; 16 C.B. 517 per Maule, J., at p. 535. To sign, as distinguished from writing one's name in full, is to make such a mark as will represent the name of the person signing. (In re Trollip, 12 S.C. 243 at p. 246, per Lord de Villiers.)"

For the reasons given above I do not agree that initials
are not marks, and I do not agree that writings on wills
which are not marks, or signatures in the ordinary

sense, may be regarded as signatures.
In my respectful opinion, the conclusion in Melvill was right. That was also the view of Howard JP in his judgment in the court a quo in the present case. It follows that in my view the appeal should be dismissed.

I come to this conclusion with regret because its effect is to defeat the intention of the testator. It appears that the will is genuine, and that the initials on the first page were in fact affixed by the

37

witnesses animo attestandi. Moreover, in consequence of

the amendment effected by s 2(e) of the Law of

Succession Amendment Act 43 of 1992, the definition of

"sign" in s 1 of the 1953 Act has now been amended to

read -

"'sign' includes the making of initials and, only in the case of a testator, the making of a mark and 'signature' has a corresponding meaning."

In his judgment in Ex parte Goldman & Kalmer NNO

1965(1) 464(W), Galgut J held that it was clear that the

testatrix intended to sign the will in that case and

said that it seemed to him that it was proper at that

stage of the proceedings to hold that prima facie the

"sign" made by her on the will was her signature. He

said that in coming to this conclusion he was also

influenced because the facts showed that the will

certainly appeared to be genuine and represented the

real intention of the testatrix, and he referred to Ex

38

parte Nel 1955(2) SA 133 (C) at 136 (See p. 469 G).

In Nel's case Steyn J said that in coming to

the conclusion that the will there was invalid, he had
not overlooked the tendency by our courts to uphold a
will rather than declare it invalid for want of due
execution, adding -

"That is a consideration which weighed with me particularly because this appears to be a genuine case for relief. This consideration is, however, not strong enough to overcome what I conceive to be the directions of the Legislature for the execution of a valid will."

That accords with the approach of the Lord Chancellor in

Hindmarsh v Charlton (supra) when he said at the
beginning of his speech (at 166):

"My Lords, these are very distressing cases for Judges to determine. I may honestly say that we have a strong inclination in our minds to support the validity of the will in dispute, which the parties bona fide made, as they believed, according to law, and where there is not the smallest suspicion in the circumstances of the case. But we must obey the directions of the Legislature, and we are not at liberty to introduce nice distinctions

39

which may bring about great uncertainty and confusion"

and at the end (p.168)

"I regret very much that we are compelled to hold this instrument to be an invalid will, but we are constrained to do so by the Act of Parliament..."

See also the remarks of Lord Cranworth (at 168-9) -

"I concur with my noble and learned friend in having a sort of personal feeling of regret that this will cannot be sustained as a valid will. It appears to be a reasonable will, and a will as to which there is not the least suspicion of anything like fraud or imposition. But for the security of mankind, the legislature has thought fit to prescribe certain forms and rules which are necessary to be complied with, in order to authorize a distribution of property, different from that which the law would make if there was no will; the legislature, in truth, on these forms being complied with, putting into the hands of the party who is making a will, power to dispose of his property in a way contrary to what, but for the will, would be the provision of the law."

Counsel on both sides were agreed that if the

appeal should be dismissed, there should be no order as
to costs.

40

The appeal is dismissed.

NICHOLAS, AJA

CORBETT CJ )
EKSTEEN, JA ) CONCUR
KRIEGLER, AJA )

VAN HEERDEN JA:

2

The late Perumal Pillay ("the deceased") died on 26 May 1989. He was survived by his wife (the respondent in this appeal), their minor child as well as three major children and one minor child born of a previous marriage. On 25 January 1985 and at Durban the deceased had signed a document ("the will") bequeathing his estate to his five children. After his death that document was accepted by the Master as the will of the deceased. It consisted of two pages. Each page had been signed by the deceased and the second page also bore the signatures of two witnesses, but those signatures did not appear on the first and crucial page of the will. It had, however, been initialed by the witnesses.
Because of that alleged deficiency the respondent in this appeal brought an application in the Durban and Coast Local Division. The respondents cited by her were the aforesaid three major children,
3 the appellant in his capacity as curator-ad-litem for the two minor children and the Master. She sought an order setting aside the will. Her interest in the application stemmed from her capacity as an intestate heir of the deceased. The court a quo (Howard JP) allowed the application but granted the major children and the appellant leave to appeal to this court. Subsequently those children withdrew their appeals. Hence the curator-ad-litem is the only appellant before us.
Relying mainly on the judgment in Mellvill v The Master 1984 (3) SA 387 (C), Howard JP found that the writing of initials does not qualify as a signature for the purposes of s 2(1)(a) of the Wills Act 7 of 1953. That subsection provides that no will executed on or after 1 January 1954 shall be valid unless certain formalities are complied with. So, for instance, the will must be signed at the end
4 thereof by the testator (or by another person acting on his direction), and two or more competent witnesses must attest and sign the will. S 2(1)(a) (iv) then prescribes that if the will consists of more than one page, each page other than the page on which it ends, must also be signed by the testator (or by the above person) and by such witnesses anywhere on the page.
These provisions must be read with the definition of "sign" in s 1. In terms of that definition the word " includes in the case of a testator the making of a mark, but does not include the making of a mark in the case of a witness". And if a testator signs his will by making a mark on it, s 2(1)(a)(v) requires that it be certified by a magistrate, justice of the peace, commissioner of oaths or notary.
In setting out the above provisions I have ignored the amendments brought about by the Law of

5

Succession Amendment Act 43 of 1992. I have done so, and shall continue to do so, because s 15 of that Act prescribes that its provisions are not applicable to a will where the testator died before the commencement of the Act.
The question whether the requirements of s 2(1)(a) of the Wills Act are complied with if the testator or a witness places his initials on one or more pages of a will, has given rise to a sharp division of judicial and academic opinion. In three cases the courts have either expressly or inferen-tially answered the question in the affirmative: Ex parte Goldman and Kalmer NNO 1965 (1) SA 464 (W); Ex parte Singh 1981 (1) SA 793 (W), and Ex parte Jackson NO: In re Estate Miller 1991 (2) SA 586 (W). The reasoning in Jhajbhai v The Master 1971 (2) SA 370 (D) also appears to support this view. An opposite conclusion was reached in Dempers v The Master (1)

6 1977 (4) SA 44 (SWA), in Mellvill and, of course, in the judgment of the court a quo.

Before analysing the reasoning in those cases it is convenient to give a brief resume of relevant pre-Union legislation and decisions.
Ordinance 15 of 1845 (Cape), Law 2 of 1868 (Natal), Ordinance 14 of 1903 (Transvaal) and Ordinance 11 of 1904 (Orange River Colony) all required the last page of a will to be signed by the testator and witnesses. The Transvaal and Orange River Colony enactments moreover prescribed that every sheet of a will had to be so signed, whilst the Cape Ordinance required one side of every leaf of a will to bear the signatures of the testator and witnesses. Unlike that Ordinance, the other enactments expressly provided that a mark was an acceptable form of signature. In terms of s 1 of the Orange River Colony Ordinance, however, only the testator could sign by

7

way of making a mark, and in such a case the mark had to be made in the presence of, and be attested by, a justice of the peace. All four enactments remained in force until the inception of the Wills Act.
Apart from the Cape there were no reported decisions on the meaning of the word "sign" in the pre-1953 enactments. In the Cape Colony there was initially some divergence of judicial views. In Van Vuuren v Van Vuuren 2 S 116, the majority of the Cape Supreme Court held that a will was invalid because the testator and one of the witnesses had written their initials on the first page. However, in Troost v Ross, Executrix of Hohenstein 4 Searle 211, it was held that the making of a mark by a testator constituted a compliance with the requirements of Ordinance 15 of 1845 relating to the signing of a will. A similar conclusion was reached in Re Le Roux 3 SC 56. And in In re Trollip 12 SC 243, the

(
8

majority judgment in Van Vuuren was overruled. The court held that a will had been validly executed although a page of a leaf had not been signed by the testatrix who had, however, placed her initials on that page.
The later Cape decisions were probably influenced by the trend in England. As regards statutory requirements relating to the subscription of a will, the English courts have consistently held that a "subscription" means a signature which is descriptive of the person signing, whether by a usual signature, a mark or by initials. (See the cases referred to in Mellvill at p 391.)
Trollip was decided in 1895. There are no later Cape decisions which are in point, and it would therefore appear that at the time of the enactment of the Wills Act the Trollip interpretation of the word "sign" in legislation pertaining to wills had for a

9

period of nearly 60 years been regarded as the last word on the subject, not only in the Cape but also in the other colonies (and later provinces).
Before dealing with the arguments advanced in support of a different construction of the word "sign" in the Wills Act, it is apposite to consider an aspect of the approach of Vermooten J in Singh and Jackson. In Singh he said (at p 798) that "where a testator or a witness signs with his initials only, then that will be sufficient compliance with the requirements of the Wills Act ... to sign the will provided he intends it to be his signature".
The qualification in the phrase underlined by me was criticized by Friedman J in Mellvill at p 396. He said, rightly in my view, that intention cannot make a signature of something which is not a signature, and that if the intention of the testator - and obviously also a witness - was to be the

10 decisive factor, any mark or sign made by him could be construed as a signature if he intended that mark or sign to be his signature whether or not it complied with the meaning of the word "signature".

Adverting to this criticism in Jackson at pp 589-590 Vermooten, then AJ, explained his use of the above underlined phrase. His approach may be thus summarized:

(a) The word "signature" should be con
strued as widely as the common law permits, except in
so far as the legislature expressly restricted such a
construction.

(b)	Writings on a will, other than marks, must be interpreted in terms of the common law applying to such writings before the introduction of the Wills Act.
(c)	Effect must be given to the presumption

that the legislature does not intend to alter the

11 common law.
(d) Since the intention of a signatory is the criterion in terms of the common law, that criterion must be applied in interpreting the Wills Act.
I have experienced considerable difficulty in attempting to ascertain which rules or criteria of the common law Vermooten AJ had in mind. There is indeed no reference to a common law authority in his judgment. And, as Beinart, Testamentary Form and Capacity and the Wills Act>, 1953, 70 SALJ 159, 171, points out, signatures on a will did not play a significant role in Roman or Roman-Dutch Law. Some support for proposition (d) may, however, be derived from Schrassert, Consultation 4.36. A Dutch statute apparently provided that an antenuptial contract had to be signed by the parties and witnesses. One of the questions raised in the advice given by Hendrik ...

12

Schrassert (presumably Johan Schrassert's father) and one Westenberg, was whether the making of a mark sufficed. In answering in the affirmative the consultants rhetorically asked (at p 180):

"... wat is openbaarder dan dat 't woord ondertekenen, 't geen d.i. art. gebruikt word, in 't generaal genomen het haelen en tekenen van een merkteken includeert".

And (at p 181):

"Oock als men omtrent de woorden niet te veel wil subtiliseren, wat isser bekender, dan dat de waarheid van 't gepasseerde soo wel kan afhangen van de characteren van een handmerk, dat bewijs is, als van de characteren van de naam. Dewyl en het merkteken en de naam beide alleen dienen om te verkonden dat sulks in dervoegen voor die tekenaars gepasseert is."

With reference to Huber 2.12.43 the consultants also pointed out that because the making of a mark was a customary form of signature, the Court of Friesland had approved of wills to which the testator, and even a witness, had appended his mark. It

13

does not appear from Huber, however, why such wills

had been held to be valid. For Huber says (Gane's

translation, vol 1, p 154):

"However, since signature by mark is very customary here, the Court has approved various testaments, in which the mark of the maker appeared alone, with a note in the hand of another to show whose mark it was, and there is now no longer any doubt about this; indeed we notice the same thing in the case of the witnesses themselves to a solemn testament, to the effect that in addition to the writer of the will at least two witnesses should be found, who had signed the same with their names; that, however, is too great a departure from the written law."

It will be observed that the. courts did not merely equate a mark with a signature but required that a testator' s mark had to be authenticated by somebody else.
I have been unable to establish whether a similar rule was applied in Holland, and it would therefore be somewhat rash to deduce from Schrassert and the decisions of the Court of Friesland the

14 existence of a recognized rule of the common law to the effect that a "signature" included a "mark", or that any form of writing intended to be a signature, or to be indicative of a name, was to be equated with a signature.
It may be that when he spoke of a rule of the common law Vermooten AJ had in mind the interpretation of "sign" adopted in the later Cape cases. Those decisions were not, however, based on the common law but solely on a construction of a statutory provision. As such that interpretation can hardly be said to have become part of the common law.

I turn to the main reasons put forward in
Mellvill, Dempers and the judgment of the court a quo
for the conclusion that the word "sign" in s 2(a) of
the Wills Act does not include the writing of

initials. They are as follows:

15

(1) That word does not have a technical
meaning in legal nomenclature, and must accordingly
be construed according to its ordinary, popular
sense. In that sense "sign" means to write one's
initials (or one of them) and surname. By contrast
the word "initial" bears a completely different
meaning; i e to write one or more of one's initials
as well as the first letter of one's surname.
(2) The Act is intended to eliminate as far
as possible the perpetration of fraud. A signature
is less easy to copy than initials, and there will
consequently be less scope for fraud if "sign" is
construed to exclude the writing of initials. Given
the design of the legislature there is accordingly an
added reason to interpret "sign" according to its
popular and normal sense.
(3) Since the Act draws a distinction
between a signature and a mark and makes specials

16 provision for signing by means of a mark, it is unnecessary to construe "sign" widely so as to include the making of a mark or initials. Cases decided in the Cape Colony on the meaning of the same word in s 3 of the Cape Ordinance are therefore no longer of assistance.

I shall deal consecutively with the above considerations: Ad (1)
It is true that when "sign" is contrasted with the writing of initials a tolerably clear distinction can be drawn between the meanings of those words or concepts. Even so, a signature or an initialing can take on various forms. So, for instance, one person may sign by writing his name and only one of his initials, or even only his surname, whilst another may write his surname preceded by all his initials or even his full, forenames. Again, a

17

signature may consist of no more than an illegible scrawl, squiggle or flourish. And should a woman write out a will leaving her whole estate to her only daughter and append at the end thereof, in her handwriting, the words "your loving mother", there would be much to be said for the view that she signed the will (cf Law Reform Commission of British Columbia, Report on the Making and Revocation of Wills, p 29). An initialing, in turn, may consist of the writing of the first letter of one or more of a person's forenames and the first letter(s) of his surname(s), but also of the appending of only the first letter(s) of his forename(s).
The important point, however, is that when the two concepts are not thrown in contrast, "sign" has an extended or wide meaning which embraces the writing of initials. A signature can therefore be either a full or an abbreviated signature. That much

18

is borne out by dictionaries. So, for instance, the Oxford English Dictionary defines "initial" as "to mark or sign by initials". And HAT gives the following meaning of "paraaf": "handtekening met voorlet-ters". (And cf Matanda and Others v Rex 1923 AD 435, 436 and Putter v Provincial Insurance Co Ltd 1963 (3) SA 145 (W) 148-9.)
The distinction between "sign" and "initial" also becomes blurred in the case of a person who, because of some or other temporary affliction, can only write his initials and not also his full surname. In such a case an initialing by him whilst so disabled would surely have to be regarded as his temporary signature.
I am therefore of the view that whilst in its narrow and more usual sense "sign" does not include the writing of initials, it does so in its wider meaning.

19 Ad (2)
This is undoubtedly a weighty consideration since, as a rule, it may be easier to forge a person's initials than his signature in the narrow sense. On the other hand some testators' or witnesses' signatures may well be more susceptible to forgery than the initials of others. Ad (3)

I am by no means convinced that the Wills Act does draw a clear distinction between a signature and a mark. The fact that "sign" is defined in s l(iv) to include, in a case of a testator, the making of a mark, is not indicative of the legislature's appreciation that, absent that definition, the word "signature" would not include a mark. I say so because given the legislature's design that a testator's mark, but not that of a witness, should qualify as a signature, it was obviously necessary to incorporate

20

s l(iv) in the Act.
A more important consideration is this. If a signature does not include an initialing the anomalous result would be that, subject to certain safeguards, provision has been made for a testator to "sign" his will by making a mark on, but not by initialing, a will. And it is indeed not easy to grasp why the legislature would have wished to afford solemnity to a certified mark than to a certified initialing. True, a mark is normally made by an illiterate person, but a semi-literate testator who has been taught to write his initials, but not to sign his name in full, may well prefer to initial his will rather than making a mark on it.

In the final analysis, however, the most
important pointer to the lawgiver's intention is to
be found in the interpretation of the word "sign"
which was eventually adopted by the Cape courts.
21

According to that interpretation "sign" in the Cape Ordinance included the writing of initials and the making of a mark. The same word appeared in s 2 of the Wills Act. Since the legislature must have been aware of the authoritative interpretation adopted in Trollip in 1895 which was not challenged in the other colonies (later provinces) - as could have been done in regard to the initialing of a will - a presumption arises that the word "sign" in the Wills Act was intended to bear the meaning assigned to it in inter alia Trollip (see Ex parte Minister of Justice: In re Rex v Bolon 1941 AD 345, 359-60). Admittedly s l(iv) of the Wills Act does not recognize a mark made by a witness as a valid signature, but, as I have pointed out, this in itself does not throw significant light cm the legislature's appreciation of the meaning of the word "sign". The legislature may well have thought that that meaning was wide

22

enough to include the making of a mark, but that for

reasons of policy it should be curtailed by the

provisions of s l(iv). On the other hand, since the

legislature was aware of the aforesaid authoritative

interpretation it is to my mind hardly conceivable

that if it was intended that "sign" should not

include the writing of initials, this would not have
been made clear. The legislature could have done so

easily enough by casting s l(iv) in the following

form:

" 'sign' includes in the case of a testator the making of a mark but does not include the making of a mark in the case of a witness, or the writing of initials in the case of a testator or a witness...."

Although the matter is by no means free from difficulty, I am consequently of the view that for the purposes of s 2(1) of the Wills Act "sign" includes the making of initials.

I would therefore uphold the appeal, order

23

that the costs of both parties be paid out of the deceased's estate, and substitute the following for the contents of para (a) of the order of the court a quo:

"The application is dismissed."

H J O VAN HEERDEN JA

---