Lecture 9 – Testate Succession in Kenya

Construction of Wills
9.1 Introduction
It sometimes becomes necessary for the court not only to determine whether or not the document
alleged to be a will is a valid will, but also what the meaning and effect of the words and phrases
used by the testator in the will are. First, it must emerge from the words, phrases and expressions
used that the document was made in contemplation of death i.e. that it is testamentary. The role
of the court is therefore, to decide what meaning should be attributed to any disputed clauses in a
will.

Issues relating to construction of wills arise out of poor drafting. The main objective of the
construction of a will is to ascertain the testator‟s intentions as expressed in the will. In Perrin vs.
Morgan (1943) AC 399 Lord Simon LC said that „the question is not, of course, what the testator
meant to do when he made his will, but what the written words he uses mean in the particular
case- what are the “expressed intentions” of the testator‟. (National Society for the Prevention of
Cruelty to Children vs. Scottish National Society for the Prevention of Cruelty to Children
(1915) AC 207).

Where a will uses words and phrases which are capable of two or more meanings and does not
show in which sense the testator intended to use them, the court is faced with two alternatives
either to declare the will void for uncertainty or decide on which of the available interpretations
is to be given to the disputed clause. The latter course is known as the benevolent approach to
the construction of wills. This approach has given rise to the so-called rules of construction of
wills.

Under section 22 of the Law of Succession Act wills are construed according to the rules made
under the 1st Schedule to the Act. There are 78 rules of construction under the 1st Schedule.
These rules are based on some basic general principles of construction of wills.

9.2 The Court construes Wills, it does not Remake Them.

The duty of the court is to interpret the words as used by the testator in the will regardless of
whether they produce an unfair result, provided that was the intention of the testator. Even
where a testator has not made a provision for his lawful dependants it is not for the court in
interpreting the will to seek to make provision for these survivors. The court interprets the will as
it stands and pronounces that the survivors are not provided for. Thereafter the survivors may, if
they so wish, file the necessary application under section 26 asking the court to make a
reasonable provision for them out of the estate. Thus the court‟s business is to construe the
testator‟s will, not to make a new will for him.

Page 1 of 10
Note, however, that a mechanical application of this principle can sometimes produce absurd
results and obvious injustice. In Scale vs. Rawlings (1892) A.C. 342 a testator devised three of
his houses to A (his niece) for life and provided that should A die leaving no children, those
houses were to go to his nephews. “A” died leaving some children surviving her. It was argued
on behalf of A‟s children that the testator intended, although his will did not expressly say so that
the testator intended, although his will did not expressly say so, that if A died leaving children,
the houses should go to them. Interestingly, both the Court of Appeal and House of Lords
decided that the will clearly gave A no more than a life interest in the houses and gave nothing to
her children. Consequently, when A died leaving children, the houses devolved upon the
testator‟s residuary devisees i.e. the nephews. Both courts held that if the testator had desired
A‟s children to benefit he would have said so.

Despite the obvious injustice that can result from such mechanical application of the principle
the rationale behind it is to guard against the tendency to impute a meaning to a will that was
never intended by the testator and thereby defeat his intentions. It also acts as a caveat against
sloppy drafters, who must be warned against using words and phrases carelessly without
considering whether they express the testator‟s true intentions.

9.3 Words are Construed in their Ordinary Natural Sense.

Words in a will are attributed their primary meaning, regardless of whether the construction will
produce a capricious meaning (see Rule 9)( Rashida Begum vs. Administrator General and
another (1951) 18 EACA 102 (Sir Barclay Nihill P, Sir Newnham Worley VP and Pearson Ag.
CJ)). In Gorringe vs. Mahlstedt (1907) AC 225, it was said that there is a presumption in
construing a will that the „ordinary and usual meaning of the words‟ should be applied. In Re
Raphael Public Trustee vs. Raphael (1972) EA 522 (Simpson J) the court was asked to construe
two wills containing the phrase „or of us dying together‟. Both testators were found dead in a
locked room with bullet wounds, the pathologist was unable to determine who died first. It was
held that the testators did not mean dying at precisely the same instant. The court concluded that
the testators died together within the meaning of the phrase in their will.

In Anarali Museraza (a minor by his next friend) Mohamedtaki A. P. Champsi vs. Mohamedali
Nazerali Jiwa and others (1966) EA 117 (Wicks J) the court was called upon to construe the
clause „in addition to this will‟ in a codicil to an earlier will. It was contended that the testator
was attempting to make further dispositions „in addition‟ to the one-third willable property under
Islamic law and as this exceeds that one-third the codicil was therefore void. The court held that
looking at the codicil as a whole by the words „in addition to this will‟ the testator intended that
the codicil was to be read with the earlier will and by the words „the following addition to it„ the
testator intended to add to the provisions of the prior will.

However, if on reading the will as a whole or on investigating the habits and circumstances of
the testator, it is evident that he used a particular word or phrase in some special sense of his

Page 2 of 10
own, the court may interpret it in this special or secondary sense provided that the word or phrase
is capable of carrying such a meaning. There are two ways in which the general principle may be
departed from, through the application of the „dictionary principle‟ and the use of a secondary
meaning where the ordinary meaning does not make sense.

The dictionary principle applies in circumstances where the testator has set up his own dictionary
in the will by defining words he uses in a particular way. The principle would apply if the
testator has a definition clause saying how particular words are used in the will.

Where the ordinary meaning does not make sense, a secondary meaning which makes sense can
be applied. In Re Smalley (1929) 2 Ch. 112 which illustrates the power of the court to interpret a
word in a secondary sense when the surrounding circumstances of the testator show that he used
it with that particular meaning. In this case a testator bequeathed all his property to „my wife E.
A. S‟. The woman named believed herself to be his wife and was generally reputed as such but
in fact, he had commuted bigamy in marrying her, for he was already married to another woman.
On evidence of these surrounding circumstances, the Court of Appeal construed the word “wife”
to mean “reputed” as opposed to “lawful” wife, for the circumstances showed that he had used
the words in this secondary sense. In Thorn vs. Dickens (1906) WN 54, evidence of surrounding
circumstances showed that the testator referred to his wife as “mother” so that a bequest to
„mother‟ in his will was taken to be a bequest to her.

Where a word has more than one meaning the general rule, that words are given their ordinary
meaning, cannot be applied. (Re Everett (1944) 176, Re Barnes’ Will Trust (1972) 1 WLR 587, ,
Re Mellor (1929) 1 Ch 446). Where special or technical words are used in a will, they are
presumed to be employed in their technical sense, unless the context clearly indicates the
contrary (see rule 8) (Re Cook (1948) Ch. 212). Such technical words may also be construed in
a secondary sense if the will provides sufficient evidence that this is the sense in which the
testator used them.

9.4 The Will must be Read as a Whole

The meaning of clauses is to be collected from the entire will (Rashida Begum vs. Administrator
General and another (1951) 18 EACA 102 (Sir Barclay Nihill P, Sir Newnham Worley VP and
Pearson Ag. CJ)). Since the paramount purpose of construction is to give effect to the testator‟s
intention as expressed in the will, the meaning of any clause in a will is to be ascertained from
the entire document and not in isolation (Abdulla Rehemtulla Walje vs. Alibhai Haji and another
(1943) 10 EACA 6 (Sir Norman Whitley CJ, Mark-Wilson Ag. CJ and Hayden J). The
provisions of the will must be construed in relation to each other.

However, where two clauses or provisions are irreconcilable or mutually inconsistent to the
extent that they cannot possibly stand together the last one prevails (see rule 2) (Re Hammond
(1938) 3 All ER 308). The rationale for this rule lies in the notion that the later clause in the last

Page 3 of 10
expression of the testator. The rule has been described by Lord Greene MR in Re Potter’s Will
Trust (1944) Ch 70 as a „rule of despair‟. Case law shows that the courts try not to apply the rule
at all (Re Alexander’s Will Trusts (1948) 2 All ER 111).

Apparently a rule has developed to the effect that where, looking at the will as a whole, it looks
like that the testator intended the first clause to apply, the presumption that the latter clause
prevails should not be applied (Re Bywater (1881) 18 Ch 17).

9.5 The Will must Speak for Itself.

As general rule, courts must ascertain the testator‟s intention from the words of the will itself
(Colclough vs. Cocker (1917) 7 EALR 120 (Hamilton CJ, Murison CJ and Pickering J), Rustomji
Kersasji Khursedji Sidhwa vs. Dinshwa Ruttonji Mehta and others (1934) 1 EACA 38
(Abrahams CJ, Lucie-Smith Ag. CJ and Horne J)). This is because the construction of wills is
about ascertaining the testator‟s intention, as expressed in the will. In Perrin vs. Morgan (1943)
AC 399, Lord Atkin remarked that „the sole object is…to ascertain from the will the testator‟s
intentions‟. In
Re Raphael, Public Trustee vs. Raphael (1972) EA 522 (Simpson J), the court was asked to
construe the phrase „or of us dying together‟ contained in the wills of two testators who were
found together in a locked room and who had died from bullet wounds. The court found that the
deceased died together within the meaning of that phrase in their wills. In the opinion of the
court, the phrase „dying together‟ did not mean dying precisely at the same instant of time, what
the deceased had in mind was death in the same air crash, road accident, ship collision or similar
calamity.

It is not about what the testator intended to do when they made their will. Where there is an
ambiguity or a deficiency on the face of the will, no extrinsic evidence as to the intentions of the
testator may be admitted e.g. of the description of a person or property in the will is so vague that
there is no person or property to whom it can apply, the bequest will be void for uncertainty (see
Rule 25).

In Re Feather (1945) Ch 343, a testator bequeathed $2000 to his servant “if still in my employ”.
The servant was conscripted into the army and was still serving when the testator died. Evidence
was adduced that shortly before he died, the testator had affirmed to one of his executors that he
wished this legacy to stand and regarded the servant as still in his employment. It was held that
such evidence was not admissible to prove that the testator intended the servant to have the
legacy whether still in his employment or not.

There are, however, circumstances where extrinsic evidence of the testator‟s intention is
admissible in construing wills. These circumstances are where the armchair rule applies, the
words are ambiguous on the face of the will, there is latent ambiguity or any part of the will is
meaningless.

Page 4 of 10
In addition to ascertaining the intention of the testator from the will, the court also has to see
whether the will can be carried into effect consistently with the rules of law. In Wakf
Commissioner of the Colony and Protectorate of Kenya vs. Alimohamed Ali Nahdi Executor of
the Will of Aisha binti Shafi, deceased (1951) 18 EACA 86 (Sir Newnham Worley VP,
Lockhart-Smith JA and De Lestang J) the testatrix, a Muslim woman, gave a house by her will to
a mosque apart from the land on which it stood.. It was held that although Islamic law permits
house and the land it stands on to be legally in different ownership, the same was inconsistent
with the Land Titles Ordinance. The disposition of the house alone was found to be invalid and
intestacy resulted. It was further found that the express provisions of the Land Titles Ordinance
had ousted the Islamic position on the matter1.

(a) The armchair rule

In construing the will the court can put itself in the testator‟s position at the time he or she made
their will, in order to understand the words of the will itself (Boyes vs. Cook (1880) 14 Ch D 53).
The objective of the exercise is so that the court can make itself aware of the facts that were
known to the testator at the time of the execution of the will.

The armchair rule is used most commonly to identify the beneficiary or the subject matter of the
gift. It is applied by the courts by construing the will without reference to the surrounding
circumstances and applying the apparent effect of the will to the surrounding circumstances to
ascertain that the will is being construed in accordance with the circumstances which prevailed at
the time when the will was made.

This rule can only be used to confirm the apparent effect of a will or to shed light on vague
terms. It cannot be used to alter the effect of the words used in the will if those words are clear
and unambiguous. The armchair rule could be used to explain an unclear term (Ricketts vs.
Turquand (1848) 1 HL Cas 472). (Kell vs. Charmer (1856) 23 Beav. 195, National Society for
the Prevention of Cruelty to Children vs. Scottish National Society for the Prevention of Cruelty
for the Prevention of Cruelty to Children).

(b) Ambiguous words

Where words are ambiguous on the face of the will, either direct or circumstantial evidence is
admissible to explain the words used. Words are said to be ambiguous on the face of the will
where the words used have more than one normal meaning, as in the terms „money‟ and „my
effects‟, or where the words used are equally applicable to two or more persons or items of
property.

(c) Latent ambiguity

1
See also Khatijabai vs. Kassam Sunderji Samji and others (1955) 22 EACA 301 (Sir Barclay Nihill P, Sir
Newnham Worley VP and Sir Hugh Holmes J)

Page 5 of 10
A latent ambiguity occurs where a will is ambiguous not on its face, but only in the light of the
surrounding circumstances, for example where a testator gives property to „my nephew
Onyango‟, and the testator had several nephews by that name. Another example is where he
testator leaves his „Volkswagen beetle to my daughter‟, and the testator had more than one
Volkswagen beetle. This type of ambiguity is also called equivocation. Equivocation not only
occurs where the description fits two persons or things exactly, but also where the description is
not in all respects totally accurate (Bennett vs. Marshall (1856) 2 K & J 740). If equivocation
cannot be solved with the aid of extrinsic evidence, the gift will fail for uncertainty.

(d) Meaningless provisions

A meaningless provision is one where the court cannot without extrinsic evidence give any
meaning to the word or phrase (Kell vs. Charmer).

A provision of a will cannot be said to be meaningless, simply because the provision seems
pointless, in the sense that it has no effect. So if a testator was to provide „ I give nothing to my
son‟, the clause would not be meaningless and extrinsic evidence would not be admissible to
suggest that the testator meant anything other than to make no provision in her will for his son.

Extrinsic evidence is not admissible on the basis that a provision is meaningless, in order to
complete a blank space in a will. For instance, if a legacy provided „I give my son Kamau…‟
The rationale for the rule is that the purpose of admitting extrinsic evidence is to assist in the
interpretation of the will, and arguably one cannot interpret a blank space. Interpretation should
be of a phrase as a whole.

(The only exception to this principle is that evidence may be led as to the circumstances in which
the testator was situated at the time he made a will, so as to help the court to ascertain the
meaning he intended to impute to his chosen wording e.g. it is admissible to adduce evidence as
to the state of his property at the time he made the will or at his death in order to ascertain to
what property the will refers e.g. if he refers to this house No. 029 in Langata estate but there is
a clear evidence that the only house he had in Nairobi is number 092 in Langata estate, this
evidence is admissible – Ref: Re Smalley Supra).

9.6 Ascertaining the Subject Matter of Gifts

Paragraph 3 of the First Schedule provides that, as regards property, a will speaks from the date
of death unless a contrary intention appears by the will. A gift of „all my shares‟ would be taken
to refer to all the shares owned by a testator at the date of his death, rather than being confined to
the shares which he owned at the date on which he executed his will. Paragraph 3, however,
states that with reference to specific gifts or legacies, the presumption is that the will speaks as at
the date of its actual execution.

Page 6 of 10
In In the Matter of the Estate of Ivo Murray Murton, Deceased (1938) 18 (1) KLR 65 (Sir Joseph
Sheridan CJ)) where the testator directed the trustees to hold property upon trust for his son until
the son attained majority age, it was held that the son took a vested or contingent interest in the
property at the testator‟s death and on attaining majority age was entitled to a conveyance and
transfer of the property to his name.

A contrary intention refers to where the property is described in a very specific way (Re Gibson
(1866) LR 2 Eq 669, Re Sikes (1927) 1 Ch 364). In deciding whether there is a contrary intention
to Paragraph 3, difficulties occur where the testator uses terms such as „now‟ and „at present‟.
Whether such words amount to a contrary intention to the operation of Paragraph 3 depends on
whether the reference to the present time is construed as an essential part of the description of the
subject matter of the specific gift. If it is, this will operate as a contrary intention to Paragraph 3.
(Re Fowler (1915) 139 LT Jo 183, Re Willis (1911) 2 Ch 263, Hepburn vs. Skirving (1858) 32
LTOS 26).

The fact that the testator, after the execution of the will, acquires a different interest in the
property that forms the subject matter of a gift does not necessarily prevent the court from
finding that Paragraph 3 operates to make the will speak from the date of death (Saxton vs.
Saxton (1879) 13 Ch D 359).

The effects of the republication of a will need to be considered in the light of Paragraph 3. If
Paragraph 3 operates to make the will speak as to the property from the date of the testator‟s
death, obviously the republication of the will has no effect. If Paragraph 3 does not apply
because a contrary intention is in operation, the will is generally taken to speak from the date of
the codicil. (Re Reeves).

In In the Matter of the Estate of Ivo Murray Murton, Deceased (1938) 18 (1) KLR 65 (Sir Joseph
Sheridan CJ)) it was stated that where property is given to several persons concurrently the
question whether these persons take as joint tenants or tenants in common depends on the
context of the whole will; they prima facie take as joint tenants.

9.7 Ascertaining the Beneficiaries

Paragraph 3 only applies to property. References to people are as a general rule construed to
refer to people at the date the will was made, unless there is a contrary intention. A gift to „my
shamba boy‟ would refer to the testator‟s shamba boy at the date of the will. A gift to the
„youngest child of my niece Kanini‟, and the youngest child of Kanini alive at the date of the will
was Wayua, but Wayua had died by the date of the testator‟s death. The consequences of the
general rule that a will is construed referring to people at the date of the will, is that the gift to
Kanini‟s youngest child will lapse, even if Kanini has other children alive at the testator‟s death.
(Re Whorwood (1887) 34 Ch D 446).

Page 7 of 10
It is, however, possible to have a contrary intention to the general rule that, as regards reference
to people, a will speaks from the date of the will. (Re Daniels (1918) 87 LJ Ch 661, Radford vs.
Willis (1871) Ch App 7).

Where no person fulfils the description at the date of the will needs to be distinguished from that
where a person fulfils the description at the date of the will, but by the date of the testator‟s
death, that description has become inoperative to them (Radford vs. Willis).

A number of difficulties in construction of a will occur in relation to gifts where the relationship
is specified. This is particularly so where a gift is made to children or remoter issue. Regarding
relationships referred to in the making of gifts generally, there is a presumption that the only
persons to take are blood relatives, and not relatives by affinity or marriage. A gift to „all my
nieces‟, does not include any females born to a brother or sister of his wife. In Rashida Begum
vs. Administrator General and another (1951) 18 EACA 102 (Sir Barclay Nihill P, Sir Newnham
Worley VP and Pearson Ag. CJ) the clause „other relations‟ was construed to exclude the
testator‟s adopted daughter. It was stated that the ordinary meaning of the word „relation‟ does
not include an adopted child. The presumption can be rebutted by evidence of a contrary
intention. The position regarding relatives of half-blood is unclear. It was, however, suggested in
Re Reed (1888) 57 LJ 790, that there is a presumption that relatives of half- blood are included.

Difficulties often occur over the use of the word „children‟ or the term „issue‟. Children includes
children en ventre sa mere (Villar vs. Gilbey (1907) AC 139), and there is a presumption that the
term „children‟ refers to immediate children. The presumption may be rebutted by evidence that
grandchildren and remoter issue were intended (Loring vs. Thomas (1861) 1 Drew & Sm 497).
„Issue‟ technically means children, grandchildren and remoter descendants, but in some matters,
the courts have construed the term to refer only to the children of the testator (Re Noad (1951)
Ch 553).

9.8 The Class Closing Rules

A class gift is a gift to be divided amongst individuals who fulfil a general description, where the
amount that each individual gets depends on the number of beneficiaries falling within the class.
A gift of Kshs. 400,000.00 to Mureithi‟s children is a class gift. How much each child of
Mureithi receives depends on the number of children that Mureithi has.

Class gifts could (if it were not for class closing rules) make it difficult for the personal
representatives of an estate to make an early distribution of the property to be given to the class.
Class closing rules are rules of convenience, designed to allow personal representatives to
distribute the estate at the earliest opportunity. The rule frowns on keeping property out of
circulation or use for a long period of time. The rationale behind the class-closing rule is similar
to the rationale for the perpetuity principle. It is, however, often argued that the rules frequently

Page 8 of 10
act against the intentions of a testator, as they have the effect of excluding from the class
beneficiaries whom the deceased intended to benefit.

The class closing rules originate from the case of Andrews vs. Partington (1791) 3 Bro CC 401.
The rules only apply to gifts made by will and operate so that the class closes at the date on
which the first member of the class becomes entitled. Precisely how the rules operate depends on
the nature of the gift.

There are four types of class gifts.

(a) Immediate class gift

This usually takes the form of a gift say „to the children of Akinyi‟. If Akinyi has a child who is
living at the testator‟s death, the class closes at the testator‟s death and includes all the children
of Akinyi alive at that date (In the Matter of the Estate of W. J. Bellasis, deceased (1919-21) 8
EALR 142 (Barth J). In Rustomji Kersasji Khursedji Sidhwa vs. Dinshaw Ruttonji Mehta and
others (1934) 1 EACA 38 (Abrahams CJ, Lucie-Smith Ag. CJ and Horne J)) a direction in the
will that a share be paid to „to Rustomji‟s children during their lifetime‟ was construed as
meaning both those children who were living at the testator‟s death and those born subsequent to
that event.

In In the Matter of the Estate of W.J. Bellasis, deceased (1919-21) 8 EALR 142 (Barth CJ), a gift
to husband and wife „and their family‟ was construed as giving an immediate life interest in the
gift to the husband and wife with a remainder to any children of the marriage born before or after
the death of the testator.

(b) Deferred class gift

This would take the form of a gift say „to Kwamboka for life with remainder to the children of
Nyakenyanya‟. If Nyakenyanya has no children at the testator‟s death, the class remains open
until Nyakenyanya dies, and includes all children subsequently born to Nyakenyanya. If
Nyakenyanya has a child before the death of Kwamboka, the class closes at Kwamboka‟s death,
and includes all the children of Nyakenyanya who are alive at that date. It, however, will close at
the testator‟s death if Kwamboka predeceases the testator. If Nyakenyanya has no children at
Kwamboka‟s death or the testator‟s death, the class remains open until Nyakenyanya dies. (See
In the Matter of the Estate of W.J. Bellasis, deceased (1919-21) 8 EALR 142 (Barth CJ)).

In Latif Suleman Mohamed vs. K..J. Pandya and others (1963) EA 416 (Sir Ronald Sinclair P,
Sir Trevor Gould Ag. VP and Newbold JA)

(c) Contingent class gift

This would be a gift say „to the children of Kimathi who attain 21 years‟. If Kimathi has a child
who reaches 21 years before the testator‟s death, the class closes at the testator‟s death and

Page 9 of 10
includes all children of Kimathi alive at that date who subsequently reach 21 years. If Kimathi
has no child who has reached 21 years by the date of the testator‟s death, then the class only
closes when the first child of Kimathi reaches 21 years and includes all children alive at that
date.

(d) Contingent and deferred class gift

This would be both a contingent and a deferred gift. It could take the form of a gift; say „to
Mulusa for life with remainder to the children of Luvaga who attain 21 years‟. If Luvaga has a
child who has reached 21 years by the date of Mulusa‟s death, or the testator‟s death if this is
later, the class closes at this date. The class will then include all children alive at the date the first
child reaches 21 years who subsequently reach 21 years. If no child of Luvaga has reached 21
years by the date of Mulusa‟s death or that of the testator if later, then the class remains open
until the first child reaches 21 years and all children alive at that date who subsequently reach 21
years.

Page 10 of 10