IN THE SUPREME COURT OF APPEAL OF SOUTH AFRICA
CASE NO. 305/96
In the matter between
APPELLANT JACANA EDUCATION (PTY) LTD
AND
FRANDSEN PUBLISHERS (PTY) LTD RESPONDENT
BEFORE: HARMS, SCHUTZ, SCOTT, ZULMAN and
PLEWMAN JJA
HEARD: 21 NOVEMBER 1997
DELIVERED: 27 NOVEMBER 1997
SCHUTZ JA
2
JUDGMENT
SCHUTZ JA:
Two competing maps of the Kruger National Park ("the Park"), each with supplemental matter, are the subject of a copyright dispute. The appellant ("Jacana") was the first in the field. Later it was the unsuccessful applicant for a final interdict before van Dijkshorst J in the Transvaal Provincial Division. The respondent is Frandsen Publishers (Pty) Ltd ("Frandsen").
Jacana's map is printed on both sides of a large foldable sheet.
3
The actual map is broken into three sections, called northern, central and southern. As befits a map of the park, it depicts features such as the perimeter, roads, gates, rivers and camps. A striking feature of it is that pronounced colours are used to illustrate 16 different kinds of "ecozone" or natural area, each having its own combination of underlying rock and soil, rainfall, elevation and slope, leading to differing patterns of vegetation and animal distribution. An illustrated cross-section of each of these areas is provided. Various commonplaces are included, such as a distance and travelling time table and a key to symbols depicting features such as waterholes, look-out points, camps and the like. The map includes a grid, plotting camps on one axis and a variety of services and facilities on the other. I shall describe it more fully later. The grid
4
is also the subject of alleged infringement. The map contains the statements "Created by Jacana Education and Kruger National Park" and "Published by (C) Jacana Education, Johannesburg 1993". It is common cause that the reference to Jacana is a reference to the appellant, Jacana Education (Pty) Ltd, which has traded under the name Jacana Education throughout, despite the fact that the company was formerly styled Term Holders (Pty) Ltd.
Unlike the map, which is sold at entrance gates and shops, Jacana distributes free copies of a leaflet showing gate opening and closing times, the trading hours of shops and restaurants, and the "Rules and Regulations of the Kruger National Park" (the "Rules"). These Rules deal with a diversity of matters such as not using roller skates in camp,
5
adhering to the speed limits, remaining in one's car and not disturbing or feeding the animals. Jacana's version of the Rules is a further subject of alleged copyright infringement. At the foot of the last page of the leaflet appears the expression "(C) Jacana Education" followed by a stylised depiction of a bird, the African Jacana.
Frandsen's map is strikingly different in appearance and get-up. It comes in the form of a fold out booklet. The map is divided into seven sections, each on a separate page, preceded by a contents page showing the outline of the park and its division into seven sections. This makes the map more easy to use in a confined space than is Jacana's. The background colour is uniform and there is no attempt at all to indicate "ecozones". Again, boundaries, roads, gates, rivers, camps and
6
the like are indicated. The roads are stylised compared with those of Jacana. Kinks have been ironed out. The symbols on the key are quite different, except for the general rondavel sign for camps, the statutory "no entry" sign and the imagined blue of the lowveld rivers. Among the differences are Jacana's marking of a waterhole by a giraffe drinking, against Frandsen's more prosaic "W"; Jacana's depiction of a swing gate against Frandsen's "G" and the former's depiction of a picnic site with a cup and saucer symbol against the latter's "P". The letters W, G and P are enclosed in coloured surrounds. When the two maps are compared there are differences in the print used, the spelling of names and their positions on the maps. Frandsen's map gives the elevation of hills whilst Jacana's does not. Hills are depicted in a different way. There is not an
7
entire identity as to the hills which are depicted. Nor are all their positions identical. Some of the details of the courses of rivers differ. A distinctive feature of Frandsen's map is that the sheets are surrounded by drawings of wild animals. A striking difference in content is the depiction of neighbouring private reserves, such as Sabi Sabi, Londolozi and Timbavati. These do not appear on Jacana's map. A table of distances and a list of gate opening and closing times are included. There is also a grid, plotting camps against services and facilities, and a version of the Rules bearing similarity to that of Jacana, at least as far as content is concerned.
Unlike its map, which is divided into three parts, Jacana's grid is divided into two, northern and southern. The camps are listed on the
8
vertical axis, and the facilities on the horizontal. The descriptions of the latter which appear at the top and which read vertically are tilted diagonally towards the right to make reading easier. The colour is white on green with black print. The camps are divided into four groups: main, such as Satara and Skukuza, private, bushveld and camping. Nineteen facilities are listed, such as "petrol station", "shop", "restaurant" and "conference facilities".
Frandsen uses a single grid for the whole park, with the same basic layout. The colouring is near-black and blueish-grey upon white. The camps, needless to say, are the same. Frandsen adopts a classification the same as that of Jacana, save that it does not have the separate category of "camping". Another difference is that Jakkalsbessie is shown
9
as a bushveld and not as a private camp. As with Jacana's grid, the ordering of camps within each category is, unsurprisingly, alphabetical. Seventeen facilities are shown on the grid. Another three, which appear on Jacana's grid as available only at Skukuza, "doctor", "bank" and "post office", are listed as a separate item below Frandsen's grid. As can also cause no surprise, the facilities shown are largely common to the two grids, although their order differs. Items such as "petrol", "shop" and "caravan site" are to be found on both grids. Examples of differences in selection for inclusion are that Jacana makes provision for "educational display" and "car hire" whereas Frandsen does not. The latter, on the other hand, lists "refrigeration", "first aid centre" and "night drives", whereas the former does not. In some instances there is conflict as to
10
whether a facility common to both grids is to be found at a particular camp, for instance whether there are camping and caravan facilities at Orpen camp. The visual appearance of the contrasted grids, in so far as that matters in a literary work, differs considerably.
Jacana claims copyright in the map as an "artistic work", and the grid and Rules as "literary works" - this in terms of s 2 (1) of the Copyright Act 98 of 1978 ("the Act"). Under the subsection, in order to be eligible for copyright, these works must be "original". Although there is a dispute as to who the "author" was, it is common cause that whoever or whatever that person was, he, she or it was a "qualified person" for purposes of s 3 (1).
Jacana further claims that Frandsen infringes its copyright in the
11
map, the grid and the Rules, relying on sections 6, 7 and 23 (1) of the Act, in that Frandsen's booklet constitutes a reproduction of Jacana's three works.
Frandsen opposes Jacana's application on two main grounds: first, that subsistence of copyright has not been proved, because the originality requirement has not been established; and secondly that, in any event, an essential ingredient of infringement, copying, has also not been proved. Prominent in this part of the defence was stress on the common sources available to the parties, earlier maps and the like, which tended to make copying less likely and proof of it more difficult. Evidence on such matters tended to merge into that relevant to the first ground of opposition, lack of originality, because the existence of prior material
12
tends also to limit the scope for originality and to require more exacting proof of its existence than is the case with truly original works. Although evidence of this kind may be relevant to what are two notionally distinct spheres, originality and copying, when one takes a broad look at the evidence of the preceding sources in this case, before one descends into the caverns of the law, one wonders how copyright infringement of a map of terra cognita is to be proved, unless reliance is to be placed upon the copying of special features ("Jock's last hunt", "The fight with the crocodile", to take a famous example), rather than of mere cartographical commonalities.
In Jacana's founding affidavit the stand was taken that because of the operation of s 26 in relation to "anonymous authors", it was
13
rebuttably presumed that (a) copyright subsisted(s 26 (3)), (b) Jacana, as publisher, owned the copyright (s 26 (3)), and (c) the works were original (s 26 (4) and (5)). Consequent upon this stand, practically no evidence of originality beyond bare assertion was tendered in the founding affidavit. Consistently with this view of the onus, some evidence was tendered in reply.
The learned Judge a quo accepted, for the sake of argument, that the s 26 presumptions did operate in Jacana's favour, but decided the case against it mainly on the basis that no infringement had been proved. However, notwithstanding the assumption as to the existence of copyright, the Judge commented on all three works, to the effect that in some respects, at least, they could not be original. There was no harm
14
or inconsistency in this, as it was contended there was, because there was
no confusion on the Judge's part, and because the presumptions, even if
they do apply, are rebuttable. So much for the proceedings below. In
considering the appeal I think that the applicability of the presumptions
should be addressed first.
The presumptions
Section 26 is contained in the infringement chapter of the Act and
is headed "Onus of proof in proceedings". The relevant subsections read:
"(3) Where in any proceedings brought by virtue of this Chapter with respect to a literary, musical or artistic work or a computer program which is anonymous or pseudonymous it is established -
(a) that the work or program was first published in the Republic and was so published within the period of fifty years ending with the beginning of the calendar year in which the proceedings were brought; and
15
(b) that a name purporting to be that of the publisher appeared on copies of the work or program as first published, then, unless the contrary is shown, copyright shall be presumed to subsist in the work or program and the person whose name so appeared shall be presumed to have been the owner of that copyright at the time of the publication: Provided that this subsection shall not apply if the actual name of the author of a pseudonymous work is commonly known.
Where in any proceedings brought by virtue of this Chapter with respect to a literary, musical or artistic work or a computer program it is proved or admitted that the author of the work or program is dead, the work or program shall be presumed to be an original work or program unless the contrary is proved.
Subsection (4) shall also apply where a work or program has been published and -
the publication was anonymous or under a name alleged by the plaintiff or the State to be a pseudonym; and
it is not shown that the work or program has ever been published under the true name of the author or under a name by which he was commonly known or
16
that it is possible for a person without previous knowledge of the facts to ascertain the identity of the author by reasonable inquiry."
Common to the invocation of subsections (3) and (5) is Jacana's
assertion that the author of the map and the Rules (they being contained
in two different publications) is anonymous. Indeed, it is a fact essential
for Jacana to prove in order to bring the presumptions into operation.
Yet the path to anonymity, after some strange twists, seems to me to
peter out. Surprisingly, the deponent to the founding affidavit (Mrs
Thomas) reveals herself and two others, Lana du Croq and Greg Stewart,
to be authors in the normal sense. But she refrains from attempting to
describe what they originated. Of course, she was not in the position of
"stout Cortez with eagle eyes" as he "star'd at the Pacific . . . Silent,
17
upon a peak in Darien." The Kruger Park was well explored and well mapped. To establish that a new map which is original had been created would have required exacting demonstration. The resort to the presumptions spared this arduous labour. If the course adopted is a permissible one it must open a new chapter in copyright litigation. Anonymity, real, or even better, pretended, must become the preferred mode. If this case were not capable of decision on the footing that the authorship of both publications is not anonymous, it might have been necessary to enquire whether a plaintiff, who well knows who the author is, may resort to the presumptions arising from anonymous authorship.
However that may be, Mrs Thomas in her founding affidavit justifies the absence of evidence of originality by claiming that the author
18
was anonymous. The immediate enquiry is then whether that is in fact so.
The map (which includes the grid) states that it was "created" by Jacana Education and the Kruger National Park. The two complementary verbs used in the definition of "author" in s 1 are "makes or creates." Therefore the use of the word "creates" is an indication of a claim to authorship. Is the authorship indicated anonymous or not? A publication is anonymous where the name of an author is wanting. That is not the case here. Two names are given, Jacana Education and Kruger National Park. The existence of anonymity appears to be a simple fact, unrelated to whether there has been a misnomer, that is where the wrong person has been named as author. On the face of it, accordingly, the author of
19
the map is not anonymous. Jacana seeks to overcome this difficulty by contending that only a natural person can be the author of a literary or artistic work. I do not propose entering into the elaborate debate to which this submission may lead. Even if Jacana be correct about artificial persons, I fail to see how a person is not named when he is in fact named, even though he is not in law entitled to be an author. In any event, the name Jacana Education is not on the face of it that of a corporation, but rather that of a firm. So even the elementary factual basis for this elaborate legal argument is lacking.
As far as the leaflet containing the Rules is concerned, it simply says "(C) Jacana Education." Supposing that the Rules do constitute an original compilation, this name could be that of the publisher, that of a
20
third party who has acquired the copyright, or that of the author. Once the third alternative is a clear possibility, to my mind Jacana fails to establish, also in the case of the leaflet, that the author was anonymous.
In reply there was an attempt to introduce pseudonymity (also catered for by s 26 (3) and (5)). The point was not raised in the founding affidavit as a basis for the application of the presumptions. For that reason it cannot be raised now. Pseudonymity was not pressed in oral argument in this Court.
My conclusion is that the presumptions contained in s 26 (3), (4) and (5) do not apply to either publication. As there was no attempt to prove originality by means of evidence in the founding affidavit, such a conclusion is fatal to the appeal. Mr Franklin rightly conceded as much.
21
However, lest Jacana should feel that it has lost on a technicality or as a result of bad advice, 1 will say something brief about what would have been the merits of the case if the presumptions had operated and the onus of disproving originality had rested on Frandsen, with the consequence that Jacana would have been entitled to deal with originality in reply. The map
Even with reference to the reply there is a question whether Jacana has succeeded in meeting Frandsen's rebutting evidence upon which its denial of originality is based. I do not intend exploring that question, because I consider that Jacana has, in any event, failed to establish infringement, where it bears the onus. As Corbett JA pointed out in Galago Puvlisher(Pty) Ltd and Another v Erasmus 1989 (1) SA 276
22
(A) at 280 B-D, in order to prove copyright infringement by reproduction
the plaintiff must establish two distinct things, namely:
"... (i) that there is sufficient objective similarity between the alleged infringing work and the original work, or a substantial part thereof, for the former to be properly described, not necessarily as identical with, but as a reproduction or copy of the latter; and (ii) that the original work was the source from which the alleged infringing work was derived, ie that there is a causal connection between the original work and the alleged infringing work, the question to be asked being: has the defendant copied the plaintiffs work, or, is it an independent work of his own?"
It is upon the first element that I would focus. The existence of
prior subject-matter may render proof of objective similarity more
difficult for a plaintiff. Burger J said in Bosal Afrika (Pfy) Ltd v Grapnel
(Pty) Ltd and Another 1985 (4) SA 882 (C) at 889 C-D:
"Mr Puckrin, on behalf of defendants, is correct when
23
he argues that the 'the objective similarity' must be judged in the light of the state of the art as at the date of the making of the alleged original work. Thus, although the alleged infringement and the original work may bear a close resemblance, this resemblance may be explained by the fact that they both incorporate common prior art,"
See also the remarks of Nicholas J in Laubscher v Vos and Others 3 JOC (W) at 6, where the learned judge observed that in the case of truly original artistic works a mere comparison usually provides a ready answer, whereas the answer may not be so readily reached if the copyright and the alleged infringing works may have a common source.
The importance of the plaintiffs identifying those parts of his work which he alleges have been copied is made clear by Millett J in a case where a "Dog-N-Boots" design was alleged to be a reproduction of a "Puss-N-Boots" design. The case was Spectravest Inc v Aperknit Ltd
24
[1988] FSR 161. The relevant passage appears at 170:
"Accordingly, where the reproduction of a substantial part of the plaintiffs' work is alleged, a sensible approach is first to identify the part of the plaintiffs' work which is alleged to have been reproduced and to decide whether it constitutes a substantial part of the plaintiffs' work. The test is qualitative and not, or not merely, quantitative. If it does not, that is an end of the case. If it does, the next question is whether that part has been reproduced by the defendant. Reproduction does not mean exact replication. A man may use another's work as an inspiration to make a new work of his own, treating the same theme in his own manner; but he is not entitled to steal its essential features and substance and retain them with minor and inconsequential alterations. The question is whether there is such a degree of similarity between the salient features of the two works that the one can be said to be a reproduction of the other. In considering whether a substantial part of the plaintiffs' work has been reproduced by the defendant, attention must primarily be directed to the part which is said to have been reproduced, and not to those parts which have not."
Moving from the general to the particular, a decision of the High
25
Court of Australia concerning a map is instructive. It is reported as Sands & McDougall (Pty) Ltd v Robinson [1917] 23 CLR 49. The
plaintiff was the author of an original work because:
"He had unquestionably prepared it by taking the common stock of information in Australia and, by applying to it personal, that is, independent, intellectual effort in the exercise of judgment and discrimination, had produced a map that was new in the sense that, in respect of its size and outlines, its contents and arrangement and its general appearance, it presented both in its totality and in specific parts distinct differences from other existing maps" (per Isaacs J at 52).
The defendants (the appellants) had infringed the plaintiffs
copyright for these reasons:
"It is true that the appellants changed the colours of the political divisions, corrected the Balkan boundaries, introduced some places that had then acquired recent prominence, and cut out some places that were interfered with by some further features of arrangement of their own map. Their map was not a mere copy
26
in the ordinary sense of the term, but it was clearly a reproduction of a substantial part of the respondent's map in a material form, which necessarily violated the respondent's copyright if his work be protected by the Act. Notwithstanding all the differences referred to, there still remained in respect of size, of draftmanship, of style, of printing type, and geographical selection and general appearance, a manifest wholesale adoption of the individual work which the respondent had bestowed upon his map, and which had given to it its distinct characteristics and individuality" (at 52-3).
How do the two maps in this case compare? Needless to say there
were numerous maps of the Park in circulation long before the parties
entered the field. As one must have expected, those maps, like the maps
of the parties show the same boundaries, roads, rivers, camps, gates and
so forth. Nor does one learn with any surprise that both parties used
earlier maps, with the result that much is similar or the same. But the
common features contained in the earlier maps should be put aside in
27
considering whether there is an objective similarity between the maps in issue, such a similarity as may lead one to say that the one appears to be a reproduction of the other. Mrs Thomas made this important statement in her reply "[Jacana] at no time stated that [Frandsen's] map and [Jacana's] map look alike. They do not." The Judge below added his observation "Visually one would never say that the one map has been copied from the other." I agree with him entirely. Had I not read the papers in this case it would never have occurred to me that Frandsen's map is a reproduction of Jacana's. Having heard argument that impression remains. It was pressed upon us that there had been a reproduction of a selection of features included by Jacana in its map. I have mentioned many of the similarities and dissimilarities between the
28
two maps earlier in this judgment. Apart from overall impact, a study of details dependent on choice confirms the impression that the one map is not a reproduction of the other. This conclusion makes it unnecessary to consider the second element mentioned in Galago, causal connection, about which there was much evidence. The distinction between the two separate elements may be illustrated in this way. Suppose a tyro of great ineptitude attempts to paint a copy of a masterpiece, but the ensuing work is not recognisable as a copy. There would be no reproduction, however great the application of the tyro.
For these reasons I am of the opinion that Jacana has failed to prove infringement of the map. The grid
29
The legal principles and problems mentioned in relation to the map are relevant also to whether the grid has been reproduced, so that there has been an infringement. Again I shall assume, without deciding, that Jacana displayed originality in making its compilation.
Jacana has rightly disavowed reliance on a grid format as such, on the division of camps into classes and on the idea of showing which facilities are available at which camps. Its complaint is that it has expended skill and effort in compiling the information set out in its grid and that Frandsen has reproduced it. In deciding whether there has been a reproduction regard must be had not only to similarities but also to dissimilarities. Many of the similarities are almost inevitable in two competent lists, for instance "shop" or "caravan park." So they are of no
30
real account. When one looks at the dissimilarities, they are quite numerous and they are not unimportant. I have set out some of them above. Jacana lists some facilities that Frandsen does not. And vice versa. In some instances there is a conflict as to the facts.
On balance I am not persuaded that there has been a reproduction of Jacana's grid. For this reason I would hold that infringement in this respect also has not been proved by the onus bearer, Jacana. The Rules
Here again Jacana relies on a compilation. For reasons given earlier it is not entitled to rely on the presumptions. Assuming for the sake of argument that it may rely on them, Frandsen has demonstrated, in trying to rebut them, that various of the rules had been compiled
31
previously in a variety of publications. However, it has not produced a complete version of the Rules (assuming such a thing existed) antedating Jacana's leaflet. Nor has Jacana produced such a document in reply, despite its claim that its originality lay in "rewriting" the Rules. More generally it has not shown what prior forms of the Rules were available to it or what its input was in "rewriting" them. This is important because even an Ovid might have difficulty in metamorphosing such unyielding material as: do not feed the animals, stay in your car and vacate your accommodation by 9.00 h. And if Jacana's originality consisted in collecting rules from various sources and arranging them, it has failed to show exactly what it did. In short, once Frandsen disturbed the operation of the presumptions (still assuming that they operated) Jacana
32
failed to demonstrate what exactly it did, something peculiarly within its own knowledge. So even with the presumptions operating Jacana should fail for absence of originality.
The appeal is dismissed with costs.
W P SCHUTZ JUDGE OF APPEAL
CONCUR
HARMS JA SCOTT JA
ZULMAN JA PLEWMAN JA