Conditions, Warranties and Descriptions of Quality in Sale of Goods—I
| Date | 01 October 1952 |
| DOI | http://doi.org/10.1111/j.1468-2230.1952.tb00248.x |
| Published date | 01 October 1952 |
| Author | Samuel J. Stoljar |
CONDITIONS, WARRANTIES
AND
DESCRIP-
‘I’IONS
OF
QUALITY
IN
SALE
OF
GOODS-T
1.
INTRODUCTORY.
Where a seller delivers defective goods, two
questions immediately arise: what
is
the precise legal basis of
his
liability, and what rights has the buyer against him. Although
these questions appear quite simple, the answers to them are diffi-
cult enough. Taking
a
broad preliminary vicw, thc existing law
may, somewhat paradoxically, be criticised as follows:
(1)
it
has
too much theory,
(2)
it
has not enough theory, and
(8)
it
has
no
theory whatever. There is too much theory because
of
the ovcr-
elaborate legal symbolism of
“
conditions,”
“
warranties,”
“
snlc
by description,”
‘‘
salc of specific goods
”
and
so
on. There
is,
on
the other hand, too little theory because certain concepts, not being
properly defined, have been allowed to fluctuate. There
is,
finally,
no theory at all, because had there been
a
general
or
basic theory
the existing complications and contradictions could iicvcr have
arisen.
For
this
purpose
I
shall analyse the main legal principles and concepts and
(where necessary) retrace thcir historical career.
2.
Tm
WARRANTY.
It
is
with the warranty that we must
begin.
For
the warranty, properly understood, explains almost
the whole range and scope
of
the seller’s duties with rcgard to the
quality of the goods. And it
is
only through an analysis
of
the
warranty that we can best understand thc conlusing (because inter-
fering) development
oE
certain other important concepts such as
‘‘
caveat
cmptor
”
(including its exceptions), the
‘‘
condition,” and
“
sale by description.” Indecd,
it
can
be
truly said that without
a grasp
of
the law of warranty, the central aspects
of
the law of
sale
of
goods cannot rcally be mastered.’
In this paper
I
shall attempt to clarify the lam.
1
The best oxpositions
of
the Inw
of
warranty are those by l’rofessor Williston,
and
I
have learned moat from him; see especially,
1
Williston
on
Sales
(2nd
ed., 1948), heroafter referred
to
as Williston,
Sales,
$5
178-2213; 4
Williston
and
Thompson
on
Contracts
(2nd ed., 1936), hereafter Williston,
Contracts,
55
968-973, and 5 Williston,
Contracts,
$5
1461-1462.
For
relcronces to
Professor Williston’s several importaqf articles,
BCC
notes,
infra.
See also,
Burdick,
”
Conditions and Warranties (1900)
1
golumbia
L.R.
71; Prosscr,
*’
Tho Implied Warranty
of
Merchantable Quality (10.13)
21
Can.13.R. 446
ct
seq.
(reprinted
froin
(1913) 27 Mitnesoln L.R. 117
et
seq.).
For
a broader vicw
of
the subject:
I<.>
N. Llowellyn, On Warranty
of
Quality, and Society
”
(1936)
36 Co1.L.R. 699; (1937) 37 Col.L.11. 341
;
Bogcrt and Fink,
*‘
Business Practice
egarding warranties in Sale
of
Goods
”
(1930) 25 Illinois L.R. 400; Morrow,
Warranty of Quality:
A
Comparative Study,’,’ (1940) 14 Tulanc L.R. 327, 527;
Rnbel,
“
The Nature of Warranty o!,Quality
As
Professor Llowellyn has said, [the warranty
of
quality] prcsents thn
sweep of
sales
law in perhaps its most dramatic
forin.”
Cases
ntid
Maferials
071
the
hto
of
Sales
(1930)
204.
(1950) 24 Tulane
lilt.
273.
425
426
THE
MODERN
LAW
REVIEW
VOL.
16
How then, to begin with, is the warranty to be defined?
According to section
62
(1)
of the Sale of Goods Act,
1893,
it
is
"
an
agreement
with reference to the goods which
are
the subject
of a contract of sale but
collateral
to the main purpose
of
such
contract, the breach of which gives rise to
a
claim for damages
but not to a right to reject the goods and treat the contract
as
repudiated."
This definition, we are told,5 was adopted by the draftsman
"
after much consideration," and, moreover, its main purpose
was to bring out
"
the distinction between
a
condition precedent and
a
collateral undertaking
or
promise.'' Something will later be
said about this alleged distinction; for the present we are mainly
concerned with the extremely misleading terms
&&
agreement
))
and
"
collateral."
'
Let us begin with the vital word
"
agreement.') At first sight
it
would seem that, according to the statutory definition, a war-
ranty is some sort of special, separate contract referring to the
quality of the goods and ancillary to the
"
main purpose
)'
of the
contract of sale, which purpose is simply the transfer of title.8 But
it is also clear that it cannot be a special
or
separate contract in a
strict sense of those words, because of the rule of law that no frcsh
consideration is required for such a contract," provided the war-
ranty
is
not made
after
the contract
of
sale.'"
It
follows that the
warranty is just one undertaking
or
promise in the bundle
of
promises which a seller must perform in consideration
of
the buyer's
promise to pay the price." that
"
a
reprcsentation may amount to a warranty.)' But what exactly
did the draftsman mean?
If
a warranty
is
defined as an
I'
agree-
ment
')
(and, indeed, as some special
or
&&
collateral
"
agreement),
3
I'
There is no inore troublesome word in the law than the word
'
warranty
I,''
1
Williston,
Sales,
5
181.
4
The italics are mine. It will be noticed that the latter part
of
the statutory
definition states only the
eflect
of
breach of warranty and does not elucidate
the first part of the definition. For the American definition of warranty,
eee
5
10,
itifra.
5
Chalmers,
Sale
of
Goods
(12th cd.,
1045),
224.
See
itifra,
5
10.
5
In
connection with
all
this, Professor Williston has said
:
"
This is something
which tlie English law and the English lawyer
has
never been willing or able
to
define entirely,"
1
Williston,
Sales,
5
182.
8
In
s..1
of the Act the contract
of
sale
is
defined
as
the transfer of the property
for bhe price. But if
the
transfer
of
property
is,
bu
definilion,
the main
parpose
of
a contrnct of sale, do not the implied undertakings a8
to
title in
8.
12 become redundant?
10
In such a case the wurranty is a
"
suppleiiieiital
"
contract, Chalmers,
OP.
oit.,
46.
11
The bundle of promises may, rather scholastically, be enumerated as follows:
(1)
promise to
deliver
the goods; (2) promise to give
title
to
the goods;
(3)
promise
to
deliver the right
qualitv
of the goods;
(4)
promise to deliver
the
r!ght
quartlily
of the goods; and
(6)
promise to deliver the goods at
the
right
117i1
e.
This, however, is not all.
For
the draftsman also tells us
See also
$5
3,
4,
infra.
9
Chnlmers,
OF.
eil..
4G.
12
Chnliiiers,
01,.
cit.,
46-47.
OCT.
1952
CONDITlONS AND WARRANTIES
ON
SALE
427
how can
a
representation
amount to a warranty as well
?
For
need-
less to say,
"
representation
"
and
"
agreement
y'
arc different legal
concepts, the former giving rise, if at all, to a tortious and the
latter to
u
contractual liability.
The simple truth
of
the matter is that a warranty may be both a
promise and a repre~cntation.'~ Its dual nature, however, became
obscured by
a
rather complicated historical development. Although
the warranty started out as an agreement strictly
so
called,i'
requiring special words for its creation,15 an entirely new chapter
began in the eighteenth century.
It
was then first established by
Holt
C.J.,lG
that a vendor's statement as.to title would be an action-
able warranty even though
it
was not made in special words
or
was
made without knowledge that it was actually false." Thus arose
a novel type of warranty,
i.e.,
the warranty implied in fact, -and
its
practical importance lay in this:
(1)
it
extended the contractual
liability
of
the express warranty by making unnecessary special
words of undertaking and
(2)
it also created (what we now would
regard as) a purely tortious liability where the seller had made
affirmations
or
representations as to title which induced the buyer
to buy. This development proceeded the more easily since the
appropriate form
of
action
for
breach of warranty was the action
of
deceit. But in the latter part of the eighteenth century, breach
of
warranty began to be declared in assumpsit
la
until, eventually, the
warranty was
"
transferred almost bodily into the domain of con-
tract."
lo
Whilst,
as
a result of this procedural development, the
strictly contractual aspect
ot
the warranty was very properly
brought into relief, its delictual nature became very blurred.20
13
cf.
Williston,
"
Representiitions and Wof;rantics
"
(1913) 27
Harvard L.R.
1;
14
?!
begau as
a
covennut real in tlic early la,nd law: see
S.
J.
Bailey,
(1944) 8
Cambridge
L.J. 271;
(1945)
9'Cambridge
L.J.
82,
192.
15
Clrandelor
v.
Loprrs
(1603)
Cro.Jnc.
4.
;his case makes it quite clear that
such
words
ns
"
warrantizando actididit,
etc., were then alwys neccssory
to create on express warranty. The case
is
fully discussed (and defended) from
the point
of
view of thc law
of
pleading in
(1887)
1
Hkrvard L.R.
191;
see
also
Hamilton,
"
Tbc Ancient hIasim of Caveat Emptor
(1031)
40
Yale
L.J.
1133,
at
11GG-9;
but see
1
Williston,
Sales,
5
195,
n.
6.
16
Lord Holt's decisions were:
Crosz
V.
Cnrdner
(1689)
1
Show.
68;
ibid.,
Carthzw
90
;
3
Mod.
261;
Mcdino
v.
Stouglilon
(1700)
1
Ld.Raym.
698
;
1
Salk.
210.
How far these decisions advanced be ond the earlier law is not perfc7:ly
clear
. .
.
but Lord
Holt
at least made clenr what :as doubtful before,
1
Williston,
Sales,
J
106,
n.
7;
see also Williston, What Constitutes an
Express Warranty
"
(1908)
21
Harvnrd L.R.
555,
at
556.
17
If
the seller knowingly made a misstatement. deceit would of course, lie. See
Selwyn,
Nisi
Prius
(1820)
vol.
1, 640
el
seg.
18
Stuart
v.
Wilkins
(1778)
1
Doug.
18
is
the first reported case.
But
the
practice of declaring in assumpsit seems somewhat older; see
ibid.,
Buller
J.
at
21.
See also
Jones
v.
Bright
(1829)
5
Bing.
638; 8
Holdsworth,
History
of
English
Law,
67-70.
The warranty ,,preceded assumpsit by almost
a
century, sce Amcs,
"
History
of
Assumpsit
19
Street,
Foundations
of
Legal Liability,
vol.
1, 390.
20
It must be noticed that
the
implied-by-law worrnnty
of
title was not estab-
lished until the middle of the nineteenth century
(cf.
Moorlei
v.
Attenborough
(1849)
3
Esch.
500),
nlter the
law
hnd alrcndy evolved tie implied-by-luw
Liability
for
Honcst Alisrcpresentntion
Warranties
of
Land in the Thirteenth Century
(1911) 24
Harvard
L.R. 416.
(1888)
2
Hnrvard
L.R.
1.
428
THE
MODERN
LAW
REVIEW
VOL.
16
There
is
another point. Chalmers also noted that an affirma-
tion
or
representation may amount to a warranty provided that
‘‘
it
is
intended
as
such-that is to say, if
it
is intended to form part of
the contract.”
21
Thcse words are particularly perplexing.
For
a
representation intended to form part of thc contract is, in effect, a
contractual term; as
a
contractual term it is a part of the seller’s
promise. In short, the annotation is tautologous,
€or
all it means
is simply that a promise may amount to
a
warranty. But the whole
point about the rcprcsentation is that
it
induces the agreement,
i.e.,
induces the buyer to buy.Z2
It
follows that a reprcsentation
‘(
forming part of the agreement
’)
iulfils no separate function.
Unfortunately, not only have the books dutifully adopted the
Chalmers view,13 but the House of Lords has also given
it
judicial
recognition. Thus in
IIeiZbut
Sgmons
4
Co.
v.
Uuclcleton
(1013)
24
an important amrmation was held not to amount to a warranty
because
it
did
not
((
appear on evidence to be
so
intended,”
25
and
Lord Moulton purported to rc-enunciate
‘‘
the true principle of law
which was laid down in early days by Holt
C.J.”20
In point
of
fact, Holt had never said anything about “intention,” and the
requirement of
“
intent
))
was first mcntioned by Buller
J.,
in
PusZeg
v.
Freeman
(1780)
z‘
and attributed by him to Holt. And
all that Bullen probably meant was not an intention to contract,
but an intention to
induce
the other party. The true meaning
of
“
intent
”
WRS
well stated by Benjamin as follows
:
‘‘
In determining whether warranty
was
intended, a decisive
test
is
whether the seller assumes to assert a
jnct
of which the
buyer is ignorant
or
merely statcs an opinion
or
judgment upon
a mutter about which the seller has no special knowledge, and
on which the buyer may be expected also to have an opinion
warwuty
of
uality.
Dor
(lie latter dcvclopmcnt, see
0
4,
itifro.
It
sliould.
pcrliups,
also%c
noticed that not until the Common
Law
Procedure Act, 1852,
was
a
sharp distinction inade between tort and contract;
cf.
Rullen
and
Leake,
Pleading
(1860)
v, vi.
21
Chalmers,
op.
cit.,
467.
2z
Cf.
4
Williston,
Contracts,
5
071.
23
e.r~.,
Salmond and Williams.
The
Law
of
Contracts
(1045) 23G-7; Clieehire
21
(1013)
A.C.
30.
2;
Per
Lord Moulton,
ibid.,
at
,49.
Rut cornparc this with Lord Moulton’s
?G
Ibid.
cT
3
T.R.
51,
57.
2s
Benjatnin
on
Sale
(7th cd., 1031)
G91.
This passage
appears
in
all
cditioiia
with the two alterations mentioned in notcs (29) and
(30),
infra.
Indeed, the
passago
was adopted by the Court
of
Appeal in
Dc
Lassallc
v.
Cuildford
[1901J
2
1C.U.
215, 221, but the
reporter
failed
to
indicate where this
quotation came
from.
Because
of
this
typographical omission, Lord Moultou
selected it
for
criticism
aa
a
“
serious dcviatiou
from
correct principle.” See
on
all tliis,
1
Williston,
Sales,
5
198, and
ah0
by
(he
same author
(1013)
27
Harvard
L.R.
1.
20
In
the last cdition, the words
“
decisive test
”
are replaced by
”
valuable.
tliongh not dccisive test.”
and Pifoot,
Law
of
Contracts
(2nd cd.)
88-00.
mmarks
in
Schawcl
v.
Rcodc
(1913)
2
Ir.11.
81
at 85,
8G.
Cf.
Williston,
op.
cit.
(1008)
21 Harvard
L.11.
856,
5GO-1.
OCT.
1952
CONDITIONS AND
WARRANTIES
ON SALE
429
and to exercise his judgment.
ranty, in the latter not.” In the former case
it
is a war-
There is another aspect of all this. The alleged distinction
between affirmations intended to form part of the contract and
affirmations not
so
intended has sometimes been taken to mean
that there is a vital difference between conditions and warranties
on the one hand and reprcscntations on the other, the reason being
that whilst the former are
“
integral
”
or
“
substantive
”
parts
of
the contract, the latter are But this meaning of
“
intention
”
seems as mistaken as the one we have previously considered. And
it is respectfully submitted that, what is actually implied in this
distinction is not the difference between
“
substantive
’)
and
“
non-
substantive
”
parts of a contract, but the distinction betwcen
material
and
ininraterial
misrepresentations. That this has not
been seen is probably simply due
to
the fact that a seller’s
representations are usually material inasmuch as they affect the
merchantable quality
of
the goods. Yct there may be immaterial
rcpresentations also. Suppose a vendor represent his bicycle
as
“
practically new
’)
and also aflirms that he bought it only a fort-
night carlier
;
the latter statement, if false, may nevertheless be
regarded as immaterial, provided the bicycle is practically new.
The short point
is
that only material representations amount to
warranties, and in the following discussion we shall assume that the
representations are material.
To summarise. The misfortune is that the draftsman did
not perceive the
dual
nature of the warranty more clearly.
For
not only was it clearly recogniscd in many nineteenth
century decisions that a warranty could be both a promise and a
repre~entation,~~ but it is precisely this duality which makes the
warranty that extremely useful as well as indispensable concept in
the salc
of
goods. Consider, for example, the usual case
where
A
buys goods from B, B stating that the goods are sound
(c.6.)
“
I
can also say that the goods are in perfect condition
”).
In a sense, this statement may be regarded as part of B’s offer
(or
promise) to sell or deliver
“
perfect
’)
goods. But in another sense,
30
The last scntcncc
is
left
out
in the last edilion.
31
This distinction derives
from
Williams
J.’s
statement
of
thc law in
Behn
V.
Burness (1863)
+,
B.
&
S.
877;
3
B.
&
S.
751,
at
753-5.
Representations
Ehich \;ere not substantive parts
of
the
contract, the learned judgc called
32
See especiaty thc following cases:
Saltnotid
v.
Ward
(1825)
C.
&
P.
211;
Cave
v.
Colman
(1828) 3
M.
&
R.
2;
Wood
v.
Smith
(1829)
4
C.
&
P.
45;
Allan
v.
Lake
(1852) 16
Q.B.
560;
Stuclcy
v.
Bailey
(18G2) 1
H.
&
C.
405;
Cowdy
v.
Tlionaas
(1877) 36
L.T.(N.~.)
22.
The decision in
Hopkins
?.
Tanqucray
(1854)
15
C.B.
130; 23 L.J.C.P. 163,
seems to contradict all this,
but it is submitted that the better explanation
of
that case lies
in n
remark
by Maulo
J.,
15
C.B.,
at
140,
to the eflect that the seller’s assurance was
so
anteccdcnt
to
the actual sale that it could only be regarded as an expression
of
opinion arid not as a misrepresentation
of
fact inducing the buycr
to
buy
tho
horse.
mere re rcsentations.
Voi..
15
26
430
THE MODERN
LAW
REVIEW
VOL.
15
B
only represents the goods as sound, which representation consti-
tutes an inducement. Yet to try to classify
B's
statement as either
promise
or
representation would be
a
rathcr sterile and unnecessary
undcrtaking. All that we need usefully say
is
that we have
here
a
dual liability,
so
that
B's
statement must constitute a war-
ranty, whether it amounts to
a
promise
or
to a repre~entation.~~
The other
misleading word
in
the statutory definition is
"
collateral."
This term has, unfortunately, very many meanings which (if only
for the sake of comprehensive discussion) may perhaps be fully listed.
(1)
A
warranty may be
"
collateral
)'
for the purposes of the law
of evidence,
i.e.,
it
may operate as an exception to the parol evidence
rule and may therefore be given in evidence although the agreement
is reduced to
(2)
A
warranty may be
"
collateral
')
from the point of view of
pleading,
i.e.,
a seller suing for the price of goods need not declare
that he has performed all warranties;
it
is for the buyer to com-
plain of their breach by way of defence
or
counter-claim. Thus,
the meaning of collateral is simply that it shifts the burden
of
proof.35
(8)
A
warranty may be regarded as
"
collateral
')
in the sense in
that
it
is auxiliary to the
"
main purpose
"
of the sale, which main
purpose is the transfer of title.J'
(4)
A
warranty may similarly be regarded as
"
collateral
')
in
that
it
is a speFial agreement accompanying the sale of specific
goods and not an integral part of the
description
of the goods.S7
(5)
A
warranty may be called
"
collateral," again in the sense
of a special, separate undcrtaking, the
"
sole effect
)'
of which is
'(
to vary
or
add to the terms of the principal contract." Stated
differently, the warranty is collateral in that it is a specially agreed
exception to the general rule of
caveat emptor.
(G)
A
warranty may be regarded as
"
collateral
)'
where collateral
means less important: a collateral warranty is a less important
33
Yet the notion that the warranty is
a
purely
"
contractual concep!, seeme
still to persist; see, for example, Rabel,
"
The Nature
of
the Warranty
(1950)
24
Tulane
L.R.
273,
at
279.
It
is submitted that this
is
on1
true in the
sense that the warranty will, in tho nature of the case,
on&
arise in
n
contract situation; which does not mean, however, that the
basis
of
1is;Pility
for breach of warranty
is
also always contractual. See also Prosser, The
Implied Warranty
of
Merchantable Quality
"
(1949) 21
Can.B.R.
446,
at
447-8.
De Lassalle
v.
Guildford
[1901] 2
W.B.
216.
For
n
full discussion
of
this
problom, see
J.
Williams,
The
Stnttcte
of Frauds,
s.
4
(1932) 186-193;
Benjamin,
op.
cit.,
696-7.
35
Parlter
v.
Palmer
(1821)
4
B.
&
Ad.
387,
at
391.
The decision was fore-
ehadowed
in
Clark
v.
Gray
(1805)
6
East
564.
See, generally,
6
Williston,
Contracts,
5
1461.
36
See the statutory definition,
8.
62
(l),
quoted
5
2,,fupra.
3'
See Bullen and Leake,
op.
cit.,
267,
note
(a).
[For]
a
warranby
of
quality
is
not an essential element
of
a
sale, it is a collateral en$agement to be
attached to or omitted from it
at
the pleasure of the parties," Williams,
Vendor and Purchaser
(4th
ed.,
1933),
vol.
ii,
768.
8.
Tm
''
COLLATERAL
))
ASPECT
OF
THE
WARRANTY.
38
Per
Lord Moulton,
IIeilbrct
v.
Buckleton
[I9131
A.C.
30,
at
47-8.
OCT.
1952
CONDITIONS
AND
WARRANTIES
ON
SALE
431
obligation than the
(‘
condition ”-the condition denoting the seller’s
basic duty to deliver such goods as answer the de~cription.~~
(7)
A
warranty may be called
‘‘
collateral
”
from the point of
view of legal effect, inasmuch as, according to the statutory defini-
tion,
it
gives rise to a right to damages, but not to a right to
repudiate the goods.do
(8)
A
warranty may be “collateral
”
in that it is an “inde-
pendent
”
and not a
‘‘
dependent
”
promise.d1
(9)
A warranty may, finally, be
“
collateral
”
from the point of
view of pure form
or
verbal sequence.
It
may be pointed out that
(1)
supra
may still be useful with
regard to the admissibility of parol evidence;
(2)
supra
has no
longer any modern importance in sale of goods; and that
(8)
to
(8)
supra
illustrate various aspects.of the warranty which resulted from
the distinction that was drawn between sale by description and sale
of specific goods. This will become more explicit in the course of
this discussion. On the face
of
it, the ninth and last meaning
of
cc
collateral
’)
should also be the least important, but, unfortunately,
this is not
so.
Now what is meant by collateral in “form
”
or
“verbal
sequence
”?
A simple example will make this clear. Suppose A
says to
B:
cc
I’ll sell you my car,” and
B
replies:
‘‘
I’d take
it,
but is
it
all right
?
”
And
A
replies
:
‘‘
I
can guarantee that
it
is
in
excellent condition.” Thus
A
‘‘
sells
”
the car, and subsequently,
separately as
it
were, guarantees
or
warrants its excellent condition.
The warranty is therefore collateral in form
or
verbal sequence,
because it is separate from,
or
subsequent to, the seller’s promise
to sell the car.42
In the early law of sales, most warranties were collateral in form.
The reason was that the typical situation of buying and selling was
the horse-sale in the market-place combined with the stern rule of
caveat empto~.~~
For,
because
of
caveat emptor,
the early buyer
could only protect himself against the purchase
of
a latently defec-
tive horse by requiring a special warranty of quality which because
of the accidents of the situation was therefore collateral in form.
This had,
it
is submitted, two far-reaching effects. In the first
place
it
gave rise to the idea that
a
warranty was
a
special agree-
ment quite distinct from the bargain and sale of the specific article
:
in other words, a bargain and sale
plus
an express warranty seemed
to represent two contracts instead of one. Indeed, the more
so,
30
Per
Fletcher-Moulton
L.J.
(as
he then was) in
Wallis
v.
Prnft
[1910]
2
K.B.
1003,
at 1014.
40
This follows from the statutory definition and from Chalmers’ distinction
between conditions and warranties, see Chalmers,
op.
cit.,
pp.
49,
162,
176
and
226.
41
See
5
10,
infra.
42
nor
a
fuller discussion
of
this, see
5
9,
infra.
43
The history
of
caaeat
emptor
has been brilliantly, dealt with
by
Profeeeor
(1931)
40
Yale
L.J.
1133.
Hamilton,
I’
The Ancient Maxim of Cavent Ernptor
43
2
THE MODERN
LAW
REVIEW
VOL.
16
because the bargain and sale was executed on the spot
(e.g.,
the
instantaneous delivery of the horse for the immediate payment of
the price),
so
that the only contract that remained over was the
“special contract of warranty.” In the second place
it
also pro-
moted the bifurcation of the law of sales into the sale of specific
goods (with
or
without warranties) on the one hand and the sale by
description on the other. This distinction, artificial and inefficient
though it was, became the cornerstone upon which, in the nineteenth
century, the whole conceptual frame-work
of
the law of sale of goods
was built. In the early nine-
teenth century the law of sale of goods underwent a profound
transformation. The courts developed the theory that
“
sale by
description
))
constituted some sort of special type of sale; and also
substantially restricted the
rule
of
caveat emptor
by creating thc
warranty
of
quality implied-by-law. These developments took
place concurrently and overlapped, and because of
it
created much
confusion. To explain this we must consider two well-known, yet
also very misleading, cases both decided by the Court of Exchequer
in
1838.
In
Burr
v.
Gibson44
A
bought a ship from
B
which at the
time of the sale was supposed to be at sea but which in fact had got
ashore and become a near-wreck.
Her
value had dropped from
E.E,200
to
210,
the jury having found that she was just a cheap
bundle of timber. In the second case,
Chanter
v.
Hopl~ins,~~
X
ordered a
“
smoke-consuming furnace
’)
from
Y
to be fitted in
X’s
brewery.
X
had carefully indicated what he needed, but the
furnace proved to be totally unsuitable.
In both these cases the essential question was whether a war-
ranty of quality would be implied by law since the seller had neither
promised nor affirmed merchantability
(i.e.,
the scaworthiness
of
the ship)
or
fitness
(i.e.,
fitting
X’s
brewery). But it was hcld that
no such implication could be made becausc the chattels were specific
and ascertained.
“In thc bargain and sale of an existing chattel, by which
property passes, the law does not, in the absence of fraud, imply
any warranty of the good quality
or
condition of the chattel
so
sold. The simple bargain and sale, therefore, of the ship does
not imply any contract that it is then seaworthy,
or
in
a
ser-
viceable condition;
. . .
But the bargain and sale of a chattel,
as being of a particular description, does imply a contract that
And in
Chanter
V.
Hoplcins,
Lord Abingcr delivered himself
of
“
A
good deal of confusion has arisen in many of the cases
on this subject from the unfortunate
use
made of the word
4.
THE
WARRANTY
AND
“DESCRIPTION.”
Thus Baron Parke said in
Burr
v.
Gibson
the article sold is of that description.
.
. .
YY
the followilig famous passage
47
:
44 3
M.
k
W.
390;
7
L.J.Ex.
124.
“6
3
M.
k
W.
300.
at
300-400.
45
4
M.
&
W.
390;
8
L.J.Ex.
14.
47
4
hi.
k
IT.,
at
404.
Ow.
1962 CONDITIONS
AND
WARRANTIES ON
SALE
433
‘
warranty.’ Two things have been confounded together.
A
warranty is an express
or
irnplicd4’ statement of somcthing
which thc party undertakes shall be part of a contract, and
though part of thc contract, yet collateral to the express object
of it. But in many of the cases, the circumstance of a party
selling a particular thing by its
proper
description has been
called a warranty, and the breach of such a contract the breach
of warranty; but it would be better to distinguish such cases as
a
non-compliancc with a contract which a party has cngagcd to
fulfil; as
if
a man offers to buy peas of another, and he sends
him beans, hc docs not perform his contract; but that is not a
warranty; there is no
warranty that he should sell
him
peas,
the
contract
is
to sell peas,
and
if
he sells him anything else in their
stead, it is a non-performance
of
it.”
48
Several points arise from these judicial statements. In the first
place, both Baron Parke and Lord Abinger unfortunately miscon-
strued the effect of several earlier decisions.
For
these had clearly
established not only that a warranty of merchantable quality would
be implied, especially in such situations where the buyer had had no
opportunity to inspect the goods, but also that this implication did
not depend upon the distinction betwecn the sale of specific chattels
on the one hand and the sale by description on the other. This is
clearly brought out by the two earliest lcading cases concerning the
implied-by-law warranty of quality. In
Gardiner
v.
Gray
(1815)
50
the buyer bought
12
bags of
‘‘
waste silk
”
which turned out to be
unmerchantable. Lord Ellenborough said
:
“
Where there is
no
opportunity to inspect the commodity, the maxim of
caveat emptor
does not apply. He
[i.c.,
the buyer] cannot, without
a
warranty,
insist that it shall be
of
any particular quality
or
fineness, but the
intention of both parties must be taken to be that it shall be sale-
able in the market under the denomination mentioned in the con-
tract between them. The purchaser cannot be supposed to buy
goods to lay them on a Similarly, in
Jones
v.
B~ght
(1829)
52
the buyer bought copper from a manufacturer for the
‘particular purpose
of
sheathing a ship. But the copper proved
defective in that it lasted only four months instead of the normal
period of four years.” “The law, then,” said Best
C.J.,53
“
resolves itself into this: that if a man sells generally, he under-
takes that the article is
fit
for some purpose; if hc sells for
a
particular purpose, hc undertakes that
it
shall bc
fit
for a particular
purposc.”
Clearly, the
12
bags
of
“waste silk
’)
were as specified as
the wrecked ship; and the copper for the ship was as identified as
48
The
word
“
implied
”
could
obviously only
mean
I’
iniplied in
foct.”
49
My
italics.
50
4
Camp.
144.
fil
Ibid.,
nt
p.
145.
s2
5
Bing.
633.
53
Ibid.,
at
546.
Cj.
also
Laing
V.
Pidgeon
(1815) 4
Camp.
169;
OIceZl
v.
Smith
(1815)
1
Stark.
107;
Bridge
v.
Wain
(1816)
1
Stark.
504.
434
THE
MODERN
LAW
REVIEW
VOL.
16
the
“
smoke-consuming furnace.”
J4
And whilst it is true that in
making the implication
of
merchantable quality, both Lord Ellen-
borough and Best
C.J.,
pretended that they were merely giving
effect to the
“
contractual intention
’)
of
the parties, it is equally
obvious that, whether
or
not such was the actual contractual inten-
tion, the warranty implied-by-law constituted a completely new
obligation upon the seller,
i.e.,
an obligation to delivcr an article
that was merchantable
or
fitted the particular purpose indicated,
irrespectively of whether the seller had actually either promised
or
affirmed that the article in question would be merchantable
or
fitting. Indeed, in
Shepherd
v.
Pybus
(1842)
55
this obligation was
extended. In this case a warranty
of
quality was implied in a sale
of
an admittedly specific barge which the buyer had inspected and
in which property had passed to him. In other words, a warranty
was implied by law not because the buyer had not had an oppor-
tunity
for
inspection, but because the buyer had not been able to
exercise
a
judgment of his own.
“
[And] this,” as Mellor
J.
so
rightly observed in
Jones
v.
Just
(1868),“
“
appears to us to be at
the root
of
the doctrine of implied warranty, and that in this view
it makes no difference, whether the sale is
of
goods specially appro-
priated
to
a particular contract,
or
to goods purchased as answering
a particular description.”
57
The
whole development just traced completely contradicts %he
rationnle in
Burr
v.
Gibson
and
Chanter
v.
IIopkins,
and both these
decisions plainly reveal themselves as inconsistent and incompatible
with the cssential pattern
of
the nineteenth-century law.
It
there-
fore
follows that their authority stands almost fatally impaired, at
54
This point
was
first made by Professor
Prosser,
“Tho Implied Warranty
of
Mcrchantablc Quality
”
(1948) 21
Can.B.R.
444,
at
470.
65
3
Man.
&
S.
868;
11
L.J.C.P.
101.
Bart
v.
Gibson
was not cited at all, but
this does not affect the soundness of the decision; see especially the reasoning
of
Tindal C.J.,
3
hlan.
&
S.,
at
880-1.
This decision, which must also be
taken
to
ovcrrule
Blucft
v.
Osborne
(1816)
1
Stark.
384,
was foreshadowed by
Brown
v.
Edgingtoit
(1841) 2
Man.
&
G.
279;
10
L.J.C.P.
66.
In
Brown,
howover, the seller was nq.! a manufacturer, but a dealer, and it is therofore
the
first
dcfinito “dealer -case. See also
La
Ncuuille
v.
Noursc
(1813) 3
Camp.
351,
and
Burnby
v.
Bollctt
(1847) 1G
M.
&
W.
644.
Mellor
J.,
however, at the beginning of his judsinent also enunciated
obitcr
five rules
according to which the cases concerning
caucat ctnptor
should be classified
(see
(1868)
L.R.
3
Q.B., at
201
ct scp.).
The trouble with these rules is two-
fold:
(1)
Ilia first three rules make nonsense of his actual
ratio
dccidcndi,
for
they state and restate thc principle that
caucat
ctiiptor.np,plies in
full
if the
sale is for a SpCCific chattel which
may
be inspcctcd. lhis, however, was
01~0
the cnse
in
the facts before tho learned judge.
(2)
The last two
rules
date the principlo that where the seller contracts to
stimly
goods, a warranty
of quality will be implied if either the buyer has slated his particular purpose
or
there is no opportunily
of
inspecting the goods. Bat the seller may
contract to
supply
chattels which
may
be inspected. Whst then,
is
the
decisive
trst
:
the potential opportunity
to
insprct
or
the actual reliance by
the buyer? Actually, Mollor
J.
attempted to reconcile too much. In formu-
lating hin rulcs he seems
to
have forgotten that he would havc
to
mnko a
choice betwecn some contradictory decisions. One cannot build a principle
a8
cne cannot uiukc
UII
omelette without brealiing
(at
least some) eggs.
56
(18G8)
L.R.
3
Q.B.
197,
at
207.
57
This, it is submitted, is the essential
ratio dccidcndi
in that case.
OCT.
1952
CONDITIONS AND WARRANTIES ON SALE
435
any rate with regard to the general proposition that in the case of
specific chattels no warranty of merchantable quality
or
fitness
would be implied by law. But even wrong
or
obsolete decisions
sometimes preserve a petrifying vitality all their own, especially
if
Baron Parke happened to be their champion.
Barr
v.
Gibson
was followed in a few later cases
;
it was alsq given a new
lease of life through its incorporation in Mellor
J.’s
rules concerning
caveat ernpt~r.~”
From there
it
passed into the Sale of Goods Act
and Chalmers quotes Parke B., with apparent approbation.60 The
total confusion can best be seen in the peculiar drafting of section
14
of the Act, which will later be discussed in somewhat greater detail.61
In
short,
Barr
v.
Gibson
has been responsible for
so
much harm that
one is chary of according it even decent burial,
In
view of all this, Lord Abinger’s statement
of
the law assumes
renewed importance.Ga
So
full of assurance was his style, and
so
impressionably apt his peas-and-beans analogy, that
it
has hardly
occurred to anyone that, in certain respects, he might have been
mistaken.B3 For implicit in his statement is the distinction between
(1)
a buyer buying
‘‘
peas
”
which are specially promised
or
repre-
sented as being “merchantable,” and
(2)
a buyer buying
ccmerchantable peas
”;
and, according to Lord Abinger, in
(2)
the word
cc
merchantable
”
is not a warranty but part of the defini-
tion
or
description of the goods. But the distinction between the
buyer buying
‘‘
merchantable peas
”
and
‘‘
peas warranted mer-
chantable
”
is purely formal and accidental,
so
that this difference
in verbal sequence can make no difference to what the seller
promises to sell
or
the buyer is induced to buy. The short point
is
that
‘‘
merchantable
”
amounts to
a
warranty, whether collateral
in form
or
not, and the delivery of
‘‘
unmerchantable peas
”
is as
much a breach of warranty as the non-delivery of
cc
peas warranted
merchantable.G4 Furthermore, Lord Abinger also overgeneralised
a*
See,
e.g.,
Emmerton
v.
Mathews
(1862)
7
H.
&
N.
586; 31
L.T.Ex.
139;
Smith
v.
Baker
(1878) 40
L.T.
261.
..
89
See note
57,
supra.
6o
Chalmere,
op.
cit.,
60.
61
See
5
10,
infra.
62
See
p.
,432,
supra.
63
It
must be noticed that Lord Abinger was not
‘I
protesting
”
as
Chalinere
seems to have thought
(op.
cit.,
175),
against the confusion between conditions
and warranties, because
he
was simply not concerned with conditions in that
case. Nor was he concerned with layin down*
a
special rule relating to
goods
Sold
under
a
patent name.
For
&is rule, now incorporated in the
proviso to
8.
14 (l),
Chanter
v.
Hopkins
is
of
course still the leading case
though subsequent decisions have modified it greatly
;
see especially
Bn’stol
Tramways
Co.
v.
Fiat
Motors,
Ltd.
[t9101 2
ILB.
8!:;
Baldry
v.
hfarshall
[1925]
1
K.B.
260.
See
IvIechern, Trade Names
(1927) 11
Minnosota
L.R.
485.
64
Professor Williston points out that at one Ffme it was doubtful whether words
of
description were warranties or not.
The
law
is
now, however,
con-
vincingly settled that descriptive statements do constitute
a
warranty,“
1
Williston,
Sales,
5
205.
See also
Parsons
V.
Seston
(1847)
4
C.B.
899;
16
L.J.C.P.
181;
Allan
v.
Lake
(1852) 18
Q.B.
560.
486
TIIE
MODERN
LAW
REVIEW
Vor..
16
the
''
peas-beans
))
analogy. Whcrc thc scllcr merely sells
cc
peas
)'
and delivers unmerchantable peas, the question is not whether the
seller has performed his contract
or
whether
'(
unmerchantable
peas
"
answer the definition of
"
pcas."
For
the seller only sells
peas,
he does not promise to deliver
merchantable peas.
The
crucial question therefore
is
whether
caveat emptor
will apply,
or
whether, on the other hand, the law will impose an additional
duty to supply
merchantable
peus,
whether thc sale is for specific
goods
or
not.
5.
TIIE DOCTRINE
IN
STREET
v.
BLAY. Thc law, howevcr,
had from early on also been troubled by another dificulty which
for present purposes may bc called the doctrine in
Street
v.
BLiy
(1881).''
In
that case the court of King's Bench enunciated the
broad doctrine that a buyer of a specific chattel, in which property
passes to him immediately, may not rejcct the specific chattel for
breach of warranty,
on
the ground that thc buyer may not unilater-
ally revest the title in the seller.06
For
this doctrine very little authority previously cxisted, and
what there was was rather confused and inc~nclusive.~'
In
fact,
the better available authority was overwhelmingly against it.Gs
Perhaps the thcn most recent authority against the doctrine was
the decision in
Poulton
v.
Lattimore
(182Q),G0
in which the buyer
bought certain seeds which the vcndor had warranted to be
cc
good
growing seed
))
but which later turned out to be completely worth-
less. Not only was it clearly assumed that the buyer could reject
the goods in all cases where they did not correspond to warranty,
but the strange question was raised whethcr a buyer could both
escape liability for the price and not return the article at all.70
Unfortunately, the doctrine in
Street
v.
Blay
became an imme-
diate success with Baron Parke, who even referred to it on some not
65
2
1).
k
Ad.
456.
66
Lord Tenterdcn
C.J.
did not actually lay down
a
very definite doctrine. He,
in fact, expressly denied such an intention (at
p.
4GO),
and he merely said
(at
p.
463)
that in
an
executed sale, distinct from an executory one, the
vendee
may
not have
a
right to return tho article. However this may be,
this case
has
subsequently been treated
as
laying down
a
prccisc
rule.
67
C/.
Wcslon
v.
Dontrcs
(1778)
1
Doug.
23;
I'owcrs
v.
Barratt
(1786)
1
T.R.
133;
Paytic
v.
Wtale
(1806)
7
East
274;
Powcr
v.
Wells
(1778)
1
Doug., n.
24;
Dr.
Coitipfott's
Care
(1786)
1
T.R.
133,
at
136;
1
H.131.
20,
n.
(b).
6"
The right
of
rejection
seems
rccognised in
Potucr
v.
Wells
(1778)
2
Cowp.
818;
J'iclder
V.
Stadtiti
(1788)
1
H.131.
17.
Stronger aulhorities are:
Bucknnan
v.
PariuRaw
'(1788) 2
T.R.
745,
at
746,
per
Lord I
Curlis
V.
Hannay
(1800)
3
E8p.c.
82;
Paync
v.
talc
(1806) 7
East
274,
at
278-9;
.
Caswell
v.
Loarc
(1809)
1
Taunt.
566;
Oltcll
v.
Smith
(1815)
1
Stark.
107;
Yules
v.
Pim
(1816)
6
Taunt.
446; 2
Margh.
141;
Parker
v.
Paher
(1821)
4
B.
k
Ald.
387,
per
Best
C.J.,
at
395;
Ccrrnainc
v.
Burton
(1821)
3
Stark.
32;
Toultnin
v.
Hcdlcy
(1845) 2
Car.
k
IGr.
tp7.
For
a
full discussion
see
Williston,
"
Rescission for Brcncli
ol
Warranty
(103)
16.Harvard L.R.
465.
Williston's views were strongly opposed by Professor Burdick in
(1904) 4
Col.
L.R.
1;
ibid.,
at
105
(Williston's rcply)
and
Burdick's
"
Codification of the
Doctrine
of
Rescission,"
ibid.,
at
264.
'9
9
I).
L
C.
257; 7
L.J.K.B.(o.s.)
225.
70
See
the argument by Sergeant Andrews,
9
13.
&
C.
257,
at
260.
OCT.
195’2
CONDITIONS AND WARRANTIES ON
SALE
437
very relevant
occasion^.^^
However, the high-water mark of thc
new doctrinc was reached in
Dawson
v.
Collis
in
1851.“
The
buyer bought goods by sample but the goods did not correspond
to samplc. I-Ie refused to acccpt the goods and the vendor then
brought an action for the purchase-price. Apart from an interest-
ing point
of
pleading,73 thc court applied the principle in
Street
V.
l3lay
and disallowed the right of rejection. Although
it
is not
quite ccrtain whether they regarded
R
sale by sample as a sale
of
specific
it
is certain that the court wished to extend the
doctrine in
Street
v.
Bla~,~~
even to executory contracts. Indeed,
the court was trying (if onc may rationalise its reasoning) to cstab-
lish some kind of compromise: let
caveat emptor
continue
to
apply
to the extent that the buyer’s right of rejection would completely
disappear; on the other hand, let the buyer havc a right to
damages in every ease whcre the goods prove to be unmerchant-
able. Fortunately, such a rule did not become part of English
law, even though
it
had the advantage of simplicity.
Only one exception the court would allow: where the goods
were wholly worthless, the right of rejection should continue to
exist.
For
this exception, the cases of
l’oulton
v.
Lattimore
70
and
Young
v.
Cole
(1837)
77
wcrc cited. But the latter case was not
really in point, for it dealt with a completely different situation.
A
stockbroker had sold,
on
behalf of the defcndant, certain bonds
which later turned out to have no value. The stockbroker re-
imbursed the defendant to whom he had previously handed ovcr
the purchase-price, but who now refused to return it.
It
WRS
held, that the stockbroker was entitled to recover the money paid
over as there had been
a
total failure of consideration. Yet
Williams
J.,
in
Dawson
V.
Collis
78
interpreted
Young
v.
Cole
as
establishing
a
distinction bctwcen condition and warranty,
thus confusing the condition with total failure of consideration
and thc quasi-contractual action to which it gave rise. But
once the fateful word “condition
’)
was thus introduced, it was
71
Diron
V.
Yates
(1833)
5
B.
B
Ad.
310;
Sycrs
v.
Jonas
(1848) 2 Ex. 111, at
117;
Murray
V.
Mann
(1848) 2 Ex.
538,
at 538-9; scc also
Allen
v.
Cutneroir
(1833)
1
Cr.
B
M.
832;
Patesliall
V.
Troitler
(1835)
3
Ad.
B
E.
103,
at
104;
Gompertz
V.
Denton
(1832) 1 Cr.
&
M.
209.
72
10 C.B. 523; 20 L.J.c.1’.
1113.
Streel
v.
Blay
had prcvioosly already been
applied in
Parsons
V.
Sezton
(1847)
4
C.B.
899;
16
L.J.C.1’. 181.
73
‘l’he
point was an exceptionally
good
example
of
an
I’
argumentativc traverse,”
thc buyer pleading non-assumpsit in an action for the pricu for goods sold
and dclivered.
7.1
Jcrvis C.J., 10 C.B., at 531, seems
to
have regarded it
as
a sale
of
specific
chnttcl, whereas Creswell J. scums to have held diffcrcnlly,
ibid.,
at 533.
‘5
I‘
But it
sccrns
to
me
that
the
principle
of
Street
v.
Ulay
ought
to
be
cxtcnded, and that thc just and wnvcnicnt thing is, that the
vendee
should
have an action
for
breach
of
warranty
. . .
,
per
Maule J., at 532.
76
‘J
B.
dc
C.
257;
7
L.J.X.B.(o.s.) 225.
This
ca6e
is
now the illustration for
77
(1837)
3
Bin N.C.
724
:
4
Scott 489.
78
(1851)
10
C.g.
623,
at 530.
8.
53
(1)
(a).
438
THE
MODERN
LAW
REVIEW
VOL. 15
but a very small step to the further generalisation that a brcach
of
warranty would not entitle rejection
‘I
unless there be an express
condition to that effect.”
7u
This new and mistaken distinction
was later taken up by Lord Blackburn (as he became) in the very
influential decision
of
Kennedy
v.
Panama, etc.
(1867),”
and thence
passed into the Sale
of
Goods Act,81 and indeed beyond.82
Before going further
it
is necessary to consider the doctrine in
Street
v.
Blay
as such, assuming that
it
is confined to specific
chattels only. Two things may be briefly said about
it.
First, the
doctrine is unjust for it unduly penalises the buyer.
**
He
does not
want, as Professor Williston has put
it,83
an unsound horse, worth
half the money and the difference in damages.” Secondly,
it
is
entirely artificial to argue that the buyer must lose his right of
rejection because he cannot unilaterally divest himself
of
“
property
”
or
‘‘
title.” For the notion of
‘‘
passing ownership
”
is really only useful where the question arises as to who, between
buyer and seller, is to bear the risk
of
loss
or
where third parties
become involved. Furthermore, the buyer
is
entitled to revest
the property in the case
of
fraud; and the seller has even the two
further rights
of
(i) retaining and (ii) reselling the goods, even
though
“
title
”
may have left him.84
It
follows that the doctrine in
Street
v.
Blay
creates legal com-
plications which have
no
intrinsic merit.“ Above all,
it
poses the
unnecessary problem of what is a sale
of
a specific chattel as
distinct from a sale of description. Actually, as we shall later
see,se many
“
specific
”
sales were treated as sales by description
and rejection was allowed. But
if
the
‘‘
passing
of
the property
’)
had been
so
insuperable a theoretical hindrance, such a develop-
ment could hardly have been possible.
6.
“SALE
BY
DESCRIPTION” RECONSIDERED.
One
of
the
important results of (especially) Lord Abinger’s judgment in
Chanter
v.
Hopkins
(1838)”
was the introduction
of
the theory
of
sale by description. This theory had (and indeed still has)
one advantage and one disadvantage. The advantage was that
thc
“
sale by description
”
limited the applicability of the doctrine
59
Seo, for examplq; the headnote to
Dawson
.v.
Collis,
10 C.B. 623. Observe
that
”
condition is hero used
in
two completely different. fieuses: (i) total
failure
of
consideration; and (ii)
a
stipulation
in
the contract expressly pro-
viding for a right
of
rejection.
80
(1867)
L.lt.
2
Q.B. 680, at 687.
81
Chalmers,
op.
cit.,
GO.
82
See
5
10,.injra.
83
\Villiston.
I’
Rescission for Breach
of
Warrantv
”
(1903) 1G Harvard
L.R.
465,
...
at 472.
by
K.
N.
Llewellyn, “From Title to Contract&?nd
a
Bit Beyond
New York L.Q.R. 159.
47 Northwestern University L.R. (formerly Illinois
L.R.)
1 at
3-6.
*4
Seo
6s.
41, 44, 48 (3)
of
the Act.
85
The whole concept
of
title in the contract
of
sale has been brilliant12 discussed
(1938) 15
The Ambiguity
of
Yromiee,” See
nlso
my article.
86
See
5
8,
infra.
a7
4
M.
&
W.
399. See
5
4,
supra.
OCT.
1958
CONDITIONS AND WARRANTIES
Oh’
SALE
439
in
Street
v.
Blay,
simply because
‘‘
sale by description
’)
was never
clearly defined. The disadvantage was that
‘‘
sale by description
”
emerged as a special category of sale wholly distinct from the sale
of specific chattels.
To illustrate the last point, we must deal with a trilogy of cases
which established the description-theory in its modern form, and
upon which section
18
of the Sale of Goods Act is based. The
first is
Nichol
v.
Godts
(1854),””
in which the plaintiff sold to the
defendant
“
foreign refined rape oil
.
.
.
warranted only equal to
samples.” The defendant buyer refused to take
it
because what
the plaintiff delivered was not “foreign refined rape oil
))
but a
mixture of hemp and rape oil. Since the seller’s warranty extended
only to the sample which he had shown to the buyer, and since the
oil he actually supplied corresponded to that sample, he argued
that he had fully performed his contract
so
that the buyer had
no
option to reject, although he admitted that the sample was not
“foreign refined rape oil.” The court held for the buyer on the
grounds, first, that the purchaser was only bound to take the
goods as described, and, secondly, that the samples referred
‘‘
only
to
quality.” Thus, Pollock
C.B.
argued that the oil must not
only correspond with the samples in quality, but must also
“
agree
with the description of the contract of
it
as to its character.”89
And Parke
B.
said:
“
The warranty affects only the quality, but
not the
nature
of the
article itself.”
The next case is
Wieler
v.
Schilizzi
(1856)
91
where the plaintiff
had bought a specific cargo of
“
Calcutta linseed
tale quale.”
It
was then discovered that the linseed contained a large admixture
of foreign seeds which rendered the linseed unmarketable as
Calcutta linseed. The jury was asked whether this admixture
had altered the “distinctive character
’)
bf
the linseed and they
returned a verdict for the plaintiff. On a motion
for
a new trial,
the court rcfused to disturb the verdict.
“
The purchaser had a
right,” Willes
J.
said,DZ
“
to expect, not a perfect article, but an
article which would be salcable in the market as Calcutta linseed.
If
he got an article
so
adulterated as not reasonably to answer that
description, he did
not get what he bargained
for.”
D5
In
Josling
V.
Ringsford
(1868)
94
the facts were rather similar.
The buyer bought oxalic acid, the samples and the bulk of which
his servant had examined, but it was later found to contain a
10
per cent. admixture of sulphate and magnesia. This, however,
88
10
Ex.
191;
29
L..J.Ex.
314.
The decision seeins to have been
so
revolutionary
that its correctness was
nt
first
donbtcd,
6ce the remnrks
by
counsel in
Josling
v.
Kingsford
(1863)
13
C.B.(h..s.)
447,
at
45G.
89
Ibid..
at
193.
90
ibid.:
at
1%.
u1
17
C.B.
619.
92
Ibid.,
nt
624.
9:’
My
italics.
O4
13
c.B.(N.S.)
447.
My
italics.
440
THE
MODERN
LAW
REVIEW
VOL.
15
could not have bccn detected by mere inspection and was only
discovered aftcr an analysis. The buycr thcn sucd in damagcs
for
breach
of
contract, and (alternatively)
for
breach
of
warranty,
although the seller had expressly refuscd to assume any
respon-
sibility
for
the strength of the oxalic acid. Erle
C.J.
directed thc
jury that thcre
WRS
no evidence
of
any warranty, but that the
seller could only pcrform his contract by supplying such oxalic
acid which might properly be said to come within the description
of oxnlic acid. On
a
motion for a new trial, the court refuscd to
disturb the jury’s verdict in favour of the buyer, and Williams
J.
said
95
:
“
However completely the defepdant may have guarded
himself against the contracting that the thing was of any particular
quality, it is not possible to construe the contract in any other
way than that
it
was a part of the agreement that the subject
of
the
sale should be the oxalic acid of commerce.”
The results of this trilogy wcre profound and far-reaching. In
the first place, they completed and crystallised the bifurcation
of
the law of sale of goods. Whereas in the sale of a specific chattel
only complete worthlessness,
i.e.,
total failure of consideration,
would entitle the buyer to reject, in a sale by description even a
partial failure
of
consideration would give rise to such a right.
This was not merely the corollary to the rule in
Street
v.
131ay,9G
but was also the outcome of the “higher” view that was now
taken of sale by description.
For
it seemed to be assumed that
thcre was some metaphysical connection bctwcen the
“
descrip-
tion
”
of the goods and their
“
kind,”
“
nature,”
“
distinctive
character
’)
or
“
identity
”;
the warranty, on the other hand,
only affected
“
quality
)’
and not the
“
nature
”
of the article
It became therefore quite natural to
look
upon merchant-
ability
as
part of the
“
description,” since
“
unmerchantable
”
goods could be regarded as not answering their true “nature,”
“character” and
so
on.98 In the second place, the bifurcation
of the law also established the completcly wrong antithesis bctwcen
the condition on the one hand and the warranty on the other.
Indeed, Lord Blackburn regarded
Josling
v.
Kingsford
s8
as
‘‘
a
striking illustration of the distinction
”
betmccn the two.’ The
explanation was that in the sale by description, the seller’s state-
mcnts
(kc.,
promises and representations) with regard to quality
were now called
“
conditions,” whilst similar statements became
mere
“
warrantics
”
in the sale
of
a spccific chattcl. But there was
95
Ibid.,
at
457.
96
(1831)
2
13.
k
Ad.
4%;
see
5
6,
supra.
97
See
the stntctnctlt
by
Pnrke
B.
in
Nichol
v.
Codls
(1854)
10
Ex.
191,
at
195;
cited
p.
439,
supra.
98
’’
In
all,
it
seems to me, it
is
either assiimed
or
expressly
stated. that
the
fundnmental undertnking is, that the article
ofTered
or delivered shall answer
the description
of
it contained in
the
contract,”
per
Brett
L.J.
in
Ra?ldall
V.
Ncwson
(1877)
2
Q.E.D.
102,
at
109.
99
(1863)
19
C.B.(N.R.)
447.
1
Blnckburn
on
The
Contract
of
Sale,
2nd
ed.,
210.
OCT.
1953
CONDITIONS AND WARRANTIES
ON
SALE
441
no difference between these two descriptive statements; and, hence,
there was, in this context, no difference also between conditions and
warranties
as
such. The difference only arose because
of
the non-
rejection-rule established by
Street
v.
Blay
as
well as the rule, for
which
Barr
v.
Gibson
(1838)
was responsible,
i.e.,
that in the ease of
specific chattels a special or collateral warranty was required to
exclude
caveat e)rrptor.
In other words, the antithesis between
conditions and warranty resulted from
the
telescoping
of
two
quite different doctrines, one relating to the
encct
of
a
descriptive
statement and the other relating to the absence
of
an implied-by-
law warranty
of
quality in the case
of
specific chattels.'
Thirdly, the result
of
the above
"
description cases was
important also in another way.
It
will be observed that the argu-
ment in these decisions,
i.e.,
that the goods did not answer the
description, was really begging the whole question. For (to take
but one examplc) the Calcutta linseed
did
answer the description
01
"
Calcutta linseed tale quale," the seller having sold not
an
ideal type
of
linseed but
a
cargo
of
so-called linseed or linseed such
as it was. The seller's liability, therefore, could not rest on the
ground that the goods did not correspond to their actual descrip-
tion. The liability could only be explained by saying that the
buyer was supplied with goods which in the given circumstances
happened to be not merchantable at all. This, however, amounted
to no more than to a rather daring and dangerous extension
of
the
implied warranty
of
merchantability combined with
a
pretty
sub-
stantial restriction
of
the seller's freedom to contract.
These eases also raise another difficulty connected with some
aspects
of
the sale by sample which will briefly be referred to
at
a
later stage.
For
we must now deal with
a
rather more funda-
mental problem,
i.e.,
what does the concept
"
description
"
actually
describc
?
7.
DESCRIPTION, ACQUAINTANCE,
IDENTITY
AND
SABIPLE.~
A
buyer may have knowledge
of
the article he purports to buy in two
rather different ways. In the first place, he will know the subject-
matter
of
the sale by
ucquuintancc, i.e.,
by being directly aware
of
it; or he will know the subject-matter by
description,
i.c.,
when hc
is not aware
of
it and when all he knows about it is based on the
description
of
the goods. More particularly, the buyer will know
the goods by acquaintance in the following three instances
:
2
(1838)
3
M.
c>
11'.
300;
7
L.J.Ex.
121.
3
See
on all thls,
9
4,
supra.
4
I
have tried to utilise in this discussion some
of
the principles
of
modern
logic,
i.c.,
the distinction between knowledge by acquaintance and knowlcdgc
by
description. This distinction was first conceived
of
by
Bertrand Russell, see
Problems
of
Pldosophy,
pp.
GS
el
seq.
T
have, however, followed
the
latc
Professor
Srisan Stebbing's sliglitly modified
(as
well as rather simpler) treet-
ment of the ~iibject;
see
her
Modern
ltrlroductiott
to
Logic
(Gth
cd.,
1918),
chaps.
111
nnd
IX.
442
TIIE
MODERN LAW REVIEW
Vor..
15
(i)
the buyer buys
this
object, pointing at it directly;
(ii) the buyer buys an object directly before him, but referring
(iii) the buyer buys an object by way of
a
sample,
i.e.,
he buys
There
js,
in fact, no real difference between (i) and.(ii); furlher-
more, it does not usually happen that a buyer will buy mereIy by
pointing his fingers at what he wants without referring to the
object by means of a namc.
It
.follows that instance (i) is
extremely rare, and may for present
purposes
be disregarded. As
for (iii), the parties must at least
use
some descriptive symbols to
determine the
qiiantit!l
of the quality agreed on; but as far as the
quality of the goods is concerned, the buyer buys (subject to
a
later qualification)
that
quality,
i.e.,
the quality
of
which
he
is
directly aware.
On the other hand the buyer will know the goods he is buying
by description when hc either is not
or
cannot be aware of them,
Le.,
in the following three instances
:
(iv) the buyer buys a definite object,
e.g.,
the horse
"
Constan-
(v) the buyer buys
n
horse, the seller having many horses;
(vi) the buyer buys a horse as yet unborn
or
non-existing.
In these three instances the buyer is not, and cannot be (as in
(vi)
),
acquainted with the horse; and it is therefore clear that what-
ever knowledge the buyer may possess of the objects he is buying
he can only have by way of description.5
Now what does this classification show
us?
It
shows
us
that
the word
"
description
')
is here used principally for the
purpose
of
"
identifying
)'
the subject-matter of the sale. Let
us
first take
instance (ii). Here the buyer
uses
a
name, but merely
fdr
his and
the seller's convenience; he could just as well have referred to
this
or
that.
Although the horse
('
Constantine
))
is in law
D
specific, ascertained and existing chattel
because there is only one Constantine, and no other, to which the
word
"
Constantine
))
applies, the buyer's knowledge of this parti-
cular
horse is based on description, and on description only.
For
he does not, and
ex
hypothesi
cannot, know the horse called
"
Constantine
)'
otherwise than by the description(s) about the
horse which the seller
(or
someone else) has given him.E And what
has just been said about (iv) applies even more strongly to (v) and
(vi).'
5
The point is that
it
is such
a
descriptive stntement as may be understood
even "without knowing whether there is anything that may
be
described
by
it, and
we
could
use
it significantly even
if
we
believed that there
was
nothing
to
which the phrnse applied," Stebbing,
op.
cit.,
chap.
111,
8.
9.
c
As
we
shall Inter
see
(5
8,
infra)
it was never quite certain whether this
situation amounted
to
a
sole
of
a specific chattel, or
to
a
sale by description.
See,
e.g.,
Ihe disciission in
Bonnerrnan
V.
White
(1861)
10
C.B.(N.B.)
844.
7
According
to
Professor IVilliston
(1
SaZes,
s.
221),
the
word
description
slioiild
IJ~
confinccl
I(,
identificntion especially when
refercuce
is
made
to
a
to it by a name;
this quality
or
kinds of thc goods.
tine
))
which he has hcvcr seen
;
Now let
us
consider instance (iv).
OCT.
I952
CONDITIONS AND WARRANTIES
ON
SALE
443
There is, however, the important difference between (iv) on the
one hand, and (v) and (vi) on the other. Whereas in (iv) the
description
“
identifies
”
the horse
‘‘
Constantine
))
completely, in
(v) and (vii) the objects are only identified by reference to a class.
For
plainly, you cannot identify
a
horse; you cannot identify an
unascertained
or
non-existing horse; all that we can identify is
the
horse,
e.g.,
the horse
“
Constantine.” Lastly, let
us
consider
instance (iii). There,
it
can easily be seen, the sale is both descrip-
tive and demonstrative; it is demonstrative with regard to quality,
and descriptive with regard to quantity. (iii), Therefore, stands
midway between (ii) and (iv).
.
We have
so
far only been considering the word
“
description
))
in the broad sense of “identification.”
For
in order to make a
promise meaningful, the parties must promise to buy and sell
some-
thing.
And they cannot merely refer to “goods
))
as such; they
must specify the goods to achieve a
“
certainty of terms,” although
“
certainty
))
must here be understood in a very relative sense.
It
is with this aspect alone that the “identification
)’
of the goods is
concerned.
There is, however, a further aspect which does not relate to the
knowledge
of
things
or
objects, but to the knowledge of
truths
about things
or
objects. Suppose
A
sells beans to
B,
adding that
they are
“
good and edible.” These beans the buyer knows either
by acquaintance
or
by description inasmuch
as
he may be aware
of
a
specific bag of beans,
or
only knows by description a specific,
or
unascertained,
or
even as yet non-existing bag of beans.
Although he thus knows to which particular bag of beans in the
world reference is made (and in the case of unascertained
or
non-
existing goods, he will not even know this) this knowledge does not
tell him anything about the quality of the beans.
For
the descrip-
tion
“
beans
))
only denotes the class of beans, including good, bad
and indifferent beans. We do not have another name for “bad
beans
)’;
we do not classify them,
or
think of them, as “no-
beans
”;
we call them beans and add that they are rotten.
Assuming that the buyer does not examine the beans, the only
knowledge of their quality he will have
is
what the seller told him.
But this
is
a different sort
of
“
description,”
i.e.,
a
“
description
’)
which asserts some truth about the beans, which assertion may be
either true
or
false. About certain qualities, the buyer may
of
course have knowledge by acquaintance. Thus he may be directly
class and not to a :pecific thing. Professor Prosser, “The Warranty
of
Mcrchantablo Quality
(1943) 21
Can.B.R.
4A6,
at
470-2.
strongly disagreed
Yith that vie: We
,;hall
later,,eee that doscription” may mean both
identification and warianty.
For,
a rather diflerent analysis
of
the
meaning
of
description, see Montrose, The Operation
of
Description in a
Contract
of
Sale
of
Goods
”
(1937)
16
C8n.B.R.
760,
where it is argued that
descriptive
words
operate either
aR
a definition
of
the contract
goods,
or
as
a
condition,
or
as a warranty. But the criteria
for
distinguishing these operations
are, it
is
submitted, not made very clear.
444
THE
MODERN
LAW
REVIEW
VOL.
15
acquainted with their external features, their shape, size and
colour, though he may know nothing about their edibility. In
other wbrds, a buyer may have knowledge, both by description
and acquaintance of the truths or qualities of thc goods. But
where the buyer buys an article on his own knowledge by acquain-
tance, he cannot complain that the article is not the one he bought;
where, on the other hand, hc buys on knowledge by description,
the responsibility must be the sdlcr’s, the buyer having relied on
the latter’s description rather than on his own knowledge. There
is one exception to this. Even though the buyer may, of a given
article, have neither knowlcdgc by description nor by acquaintance,
the seller
is
still compelled to give an article
of
“
merchantable
”
quality. First,
suppose that a buyer is confronted by a bag of beans about which
the seller asserts that they are edible, but which the buyer has an
ample opportunity to examine for himself,
i.e.,
he has an
oppor-
tunity to get his own knowledge by acquaintance.
Is
he obliged
to get that knowlcdge
?
Or
is he cntitled to rely on his knowledge
by description
?
Although
I
shall refer to this particular problem
at a later stage,’ the bcttcr view sccms to be that mere opportunity
is irrelevant, and what matters is whether the buyer actually has
knowledge by acquaintance
or
not. Secondly, an apparent con-
flict may arise between knowledge by description and knowledge by
iicquaintance. Thus, where a buyer buys a commodity by sample,
the seller describing it as
‘‘
X,”
although, unknown to cithcr party,
the sample represents commodity
‘‘
Y,”
the buyer buys
that
quality (as shown in sample), and the seller should only be liable
for
that quality, whether it is called
“X”
or
“Y.”
For
the
buyer buys on his own knowlcdgc by acquaintancc,
so
that the
responsibility should bc fully his. Nevertheless, certain decisions
which we have previously considered
’
seem to go against this
pro-
position. The explanation, it
is
submitted, is simply that we usc the concept
“
by sample
”
in
a
twofold sense, and that we have, therefore, two typcs of sale by
siimplc: (i) whcre the buyer actually samples, inspects, and is
fully acquainted with the quality he buys, in which case that
quality, and no other is fixcd as the subject-mattcr
of
the sale, and
(ii) whcre the buyer still
“
samples
”
or
“
inspects
”
the commodity
which the seller offers, which process of sampling
or
inspecting,
howcver, reveals to him only the external qualities of the com-
modity in question.
As
regards its latent qualities
or
defects the
biiyer has no personal knowledge by acquaintance, but has to rely
on the knowlcdge by description, if any, given by the seller. This
explanation tics
up,
once the matter is given a little thought, with
Two other situations present somewhat greater difficulty.
What, thcn, is the explanation
?
8
Sco
5
9,
infra.
9
Nichof
v.
Godfs
(1854)
10
Ex.
191;
23
L.J.Ex.
314;
Il’idcr
v.
SclriIizzi
(1856)
17
C.13.
619;
Joslitig
v.
Kitupford
(1863) 13
C.~~.(N.S.)
4.17.
OCT.
1982
CONDITIONS
AND
WARlthNTIES ON
SALE
449
the view previously referrcd to, that mere opportunity of ltnow-
ledge by acquaintance is not enough, and that the decisive factor
is the buyer’s actual possession
of
such knowledge.
In the first place, it
shows that the distinction between sale by description and sale by
specific goods contains a complctely false antithesis.
For
this
dis-
tinction is based upon the further distinction between identified and
non-identified goods. But, as wc have seen, the logical antithesis,
so far as it gocs, is not bctwecri specific and described goods, but
between goods identified, wholly
or
partly by description, and goods
identified by acquaintance. In the second place, the above dis-
tinction also breaks down with regard to the truths
or
qualities of
the goods.
For
knowledge of them does not depend upon whcther
an article
is
specific,
Le.,
“
identified,” but simply upon whether
the buyer has actual knowledge
of
thc quality
(or
qualities) of the
goods concerned by acquaintancc
or
has to rely on that know-
ledge by description.
As
rcgards the lutcnt defects
of
an
article, a buyer is in exactly the same position whether he buys
it
specifically
or
in a sale by description. There is thus
no
necessity of deciding whether description means
“
identification
”
or
‘‘
descriptive statemcnt
”
of
the goods. By limiting the mean-
ing of
“
identification,” every statement about the goods becomes
a
part
of
its description
;
in this way, moreover, it is also no longer
necessary to distinguish between
‘‘
identity
”
and
‘‘
quality.”
lo
The logical analysis here presented pinpoints the exact problems
with which,
so
far as defective quality is conccrned, the law of sale
of
goods constantly has to deal. -4ctually, all these problems the
concept of warranty was (and is) quite cdpable of solving; and,
indeed, had the law of warranty been allowed to develop along
straightforward an$ consistent paths, most,
of
the frightful com-
plications which the law now faccs would never have arisen.
This analysis shows
us
sevcral things.
SAMUEL
J.
STOLJAR.~’
(To
be
concluded.)
10
Tho
analysis
hero
siiggcstcd
rimy
nlso
throw soiiie light
upon
the
Inn
of
A
disciission
or
this,
howcvcr,
would
obviously
11
J,L.M.(r,Ond.)
;
holder
of
a teacli;llg fdlnwship
nl
the
~niversi(.y
of
Chicago
mistake in
the
luw
of
conlrnct.
lead
too
far.
Lnw
SCllOOl,
1950-82.
\‘OL
15
1!J
To continue reading
Request your trial