Supreme Court Judgments

Supreme Court of Canada

Shawinigan Carbide Co. v. Doucet, (1909) 42 S.C.R. 281

Date: 1909-10-05

The Shawinigan Carbide Company (Defendants) Appellants;

and

Jean Doucet (Plaintiff) Respondent.

1909: May 11; 1909: October 5.

Present: Sir Charles Fitzpatrick C.J. and Girouard, Idington, Duff and Anglin JJ.

ON APPEAL FROM THE COURT OF KING'S BENCH, APPEAL SIDE, PROVINCE OF QUEBEC.

Negligence—Dangerous works—Defective appliances—Evidence— Onus of proof—Presumption—Art. 1054 C.C.—"Res ipsa loquitur."

In an action to recover damages for injuries sustained by him in consequence of an accident in the company's calcium carbide works, the plaintiff's evidence shewed that a furnace operated upon a new system had been recently installed, that he was employed with other workmen to charge the furnace, draw off the liquid carbide when it was ready through openings in the base of the furnace, clean the orifices and re-plug them with moist mortar preparatory to re-charging. While the plaintiff was in the performance of his work in re-plugging one of these orifices an explosion occurred which caused the injuries complained of. There was no evidence of contributory negligence.

Held, Duff and Anglin JJ. dissenting, that, apart from any presumption arising under article 1054 C.C., the fact of the explosion occurring under such circumstances sufficiently established actionable negligence on the part of the company.

Per Fitzpatrick C.J. and Anglin J. (Girouard and Duff JJ. contra, and Idington J. expressing no opinion upon the question), that, under article 1054 of the Civil Code of Lower Canada, masters and employers, as well as other persons, are responsible for damages caused by things under their control or care where they fail to establish that the cause of the injury was attributable to the fault of the person injured, to vis major or to pure accident, or that it occurred without fault imputable to themselves.

Judgment appealed from (Q.R. 18 K.B. 271) reversing the decision of the Court of Review (Q.R. 35 S.C. 285), affirmed, Duff J. dissenting.

[Page 282]

APPEAL from the judgment of the Court of King's Bench, appeal side[1], reversing the judgment of the Superior Court, sitting in review[2], and restoring that of Cannon J., at the trial, whereby the plaintiff's action was maintained with costs.

The material facts of the case are stated in the head-note.

G. G. Stuart K.C. and Howard, for the appellants.

S. Beaudin K.C. and Martel K.C., for the respondent.

Le Juge en Chef.—J'emprunte le récit des faits de la cause aux notes du juge Archambault à la cour d'appel.

L'appelante possède, à Shawinigan Falls, Province de Québec, une manufacture de carbure.

En 1906, l'intimé était à son emploi. Il était tenu, avec un autre compagnon de travail, nommé Laro-chelle, de surveiller une fournaise qui servait à fondre le carbure.

Cette fournaise était chauffée au moyen de l'électricité et fonctionnait jour et nuit.

L'intimé et Larochelle faisaient le travail de nuit, de sept heures du soir à sept heures du matin.

Leur ouvrage consistait à remplir la fournaise de charbon et de chaux, et à la vider chaque heure, pour en faire couler le carbure.

Cette opération se faisait par trois ou quatre orifices qui se trouvaient au bas de la fournaise, et par lesquels s'échappait le carbure en fusion.

Une fois la fournaise vidée, on nettoyait l'orifice par lequel le carbure venait de couler; puis, avant de

[Page 283]

remplir la fournaise de nouveau, on le bouchait au moyen de tampons de mortier.

Larochelle plaçait un tampon à rentrée de l'orifice, et l'intimé poussait le tampon jusqu'au fond de l'ouverture, à l'aide d'une longue tige de fer terminée par une plaque circulaire.

Le 27 juillet 1906, l'intimé et Larochelle se rendirent à leur ouvrage, comme d'habitude, à sept heures du soir.

La fournaise avait été remplie par les deux ouvriers de jour, et lorsque le moment fut arrivé de la vider, l'intimé et Larochelle firent couler le carbure; puis l'intimé nettoya l'ouverture par où le carbure venait de sortir, Larochelle plaça un tampon de mortier à l'entrée de cette même ouverture, et pendant que l'intimé l'y poussait avec sa tige de fer, il se produisit tout à coup une explosion, les deux ouvriers furent renversés par terre, et il s'échappa de l'orifice un jet de carbure liquide et enflammé qui atteignit l'intimé à la figure, et lui brûla complètement les deux yeux.

L'intimé réclame de l'appelante $10,000 de dommages-intérêts pour l'accident dont il a été victime.

La cour de première instance a maintenu l'action, et lui a accordé $4,000 de dommages.

L'appelante en appela à la cour de revision qui annula le jugement et débouta l'intimé de son action.

De la cour de revision la cause fut portée à la cour d'appel qui rétablit le jugement de la cour de première instance.

Le jugement de la cour de première instance est basé sur le fait que la fournaise qui a causé le dommage étant sous la garde de la compagnie cette compagnie était responsable des dommages et que au surplus il y avait preuve de faute.

[Page 284]

La cour de revision, au contraire, déclare que cette fournaise était sous le contrôle de Doucet et ajoute que ce dernier n'a montré aucune faute de la part de la compagnie appelante et qu'il était tenu de faire cette preuve pour avoir un recours en dommages-intérêts contre cette dernière.

A la cour d'appel, la majorité des juges semble avoir décidé en fait que la faute était établie et en droit que la preuve de faute n'était pas nécessaire vu qu'il était démontré que la fournaise était sous la garde de l'appelante, attendu que la faute est alors présumée par la loi.

Pour ma part je suis d'avis que la fournaise était sous la garde de l'appelante qui l'utilisait à son profit et qui tirait un bénéfice du risque qu'elle a créé. Celui qui perçoit les émoluments procurés par une machine susceptible de nuire au tiers doit s'attendre à réparer le préjudice que cette machine causera. "Ubi emolumentum ibi onus." D. 1900, 2, 289: D. 1904, 2, 257. Notes de M. Josserand.

J'accepte donc sur le fait de la garde la conclusion tirée de la preuve par le juge Cannon en première instance et adoptée à la cour d'appel. La fournaise appartenait à l'intimée qui l'exploitait à son profit et s'en servait en vue de réaliser des bénéfices dans son industrie.

Etant admis que la fournaise était au moment de l'accident sous la garde de l'appelante, je pense comme le juge de première instance; le fait même de l'accident et ses diverses circonstances, révélés par le témoignage, fournissent toute la preuve de négligence que pouvait et que devait nécessairement produire le demandeur Doucet. Le cas actuel relève, à mon avis, des mêmes principes que celui de McArthur

[Page 285]

v. The Dominion Cartridge Co.(1). Il est abondamment prouvé qu'il n'y eut ni faute, ni négligence de la part de l'ouvrier et aucune explication ni aucun essai d'explication de l'accident ne sont donnés par la compagnie appelante. Son surintendant, Porcheron, entendu comme témoin à décharge, dit qu'il ignore comment arriva l'accident, bien qu'il surveillât lui-même la. fournaise où se produisit l'explosion. A supposer, ce que d'ailleurs il n'est pas nécessaire de décider pour les fins du présent litige, que dans un cas relevant de l'article 1053 du code civil, il faille pour créer la responsabilité la preuve positive d'une faute, je ne puis justifier l'application de ce principe au cas d'un individu poursuivi, comme responsable de dommages causés par une chose inanimée dont il a la garde, ce qui est arrivé dans le cas présent. En un mot, en face de l'article 1053 qui d'après certains auteurs et la jurisprudence fait de la faute ou de la négligence la base de la responsabilité je place l'article 1054 al. 1 in fine qui est à mon avis le seul applicable et d'après lequel "on est responsable des choses que l'on a sous sa garde." Le sens que je donne à ce dernier texte c'est que tout propriétaire est responsable en raison même de sa qualité de propriétaire du dommage causé par sa chose lorsqu'elle est sous sa garde. Le principe de responsabilité établie par cet article est l'idée de garde. J'ajoute si faisant à cette cause une fausse application de l'article 1053 C.C. on dit: le principe essentiel est que sans faute point d'obligation, même alors, au dire de Lord Mac-naghten parlant au nom du comité judiciaire, dans la cause de Mc Arthur v. The Dominion Cartridge Co.[3], il ne serait pas raisonnable de l'appliquer en toute

[Page 286]

rigueur, vu les circonstances; en effet, comme dans

l'affaire McArthur[4], l'accident cause du dommage

fut l'œuvre d'un instant; l'œil humain n'en put découvrir l'origine ni en suivre le développement. "In lege

aquilia, et culpa levissima venit." Domat, Lois Civiles, 1ère partie, livre II., tit. VIII., sec. 4, n, I. (édit. Rémy, L, p. 480); Baudry-Lacantinerie, Obligations, vol. 3, No. 2868. La faute la plus légère suffit pour faire encourir la responsabilité édictée par l'article 1053 C.C. J'ai lu avec le plus grand intérêt le jugement très soigné et, s'il m'est permis de le dire, très complet et très savant de mon collègue Duff; et, tout en admettant une grande partie de sa thèse, j'hésite à donner à l'article du code civil de Québec (1054) qui pose, je le répète, le principe de responsabilité applicable au cas qui nous occupe, une interprétation différente de celle que les plus hautes autorités françaises donnent aujourd'hui à l'article correspondant du code Napoléon. Je suggère que mon savant collègue ne donne pas au membre de phrase qui se trouve à la fin de l'article 1054 C.C. al. 1 cité plus haut tout son effet. Que ces mots soient restés inaperçus, comme le dit Planiol, pendant près d'un siècle explique peut-être l'erreur doctrinale sur laquelle est basée la jurisprudence qu'il invoque.

Dans leur rapport, les commissaires disent (p. 16) que la série des articles 1053-1056 C.C. ne diffère pas ou ne diffère que par l'expression des articles correspondants du code Napoléon. Dans ces circonstances, nous devons attacher la plus haute valeur à l'interprétation donnée aux articles du code français par les tribunaux et par ses commentateurs les plus autorisés; et, dans toutes les questions de droit où la doctrine et

[Page 287]

la jurisprudence tombent d'accord, après des années de conflit et de discussion, je me sens presque forcé d'accepter leur conclusion commune et définitive.

Il me suffit, pour le but que je me propose ici, d'exposer les trois systèmes qui ont prévalu tour à tour, en France, et en Belgique, sur cette question. Je les trouve énoncés avec une clarté et une concision admirables dans Pas. 1904, 1, 246 (argument de l'avocat général):

Nous croyons inutile de vous remémorer l'état de la doctrine et de la jurisprudence, tant en France qu'en Belgique, sur la question de droit que nous avons à résoudre (c'est-à-dire la responsabilité du fait des choses inanimées que l'on a sous sa garde); nous nous bornerons à rappeler que trois systèmes principaux ont été suivis tour-à-tour:

"1. La responsabilité est encourue du moment où il est établi que le dommage a été causé par la chose, sans qu'il soit besoin de démontrer soit le vice de la chose, soit la faute du gardien;

"2. Cette responsabilité n'existe que si le gardien a commis une faute; mais l'article 1384, paragraphe 1er, établit quant à cette faute une présomption légale;

"3. Cette responsabilité ne peut être prononcée que si la victime de l'accident causé par la chose démontre l'existence d'une faute dans le chef du gardien."

Et le système exposé dans le second paragraphe, comme j'aurai l'occasion de le montrer, a finalement triomphé en France à la cour de cassation, chambre civile et chambre des requêtes, mais, pour emprunter le langage de la cour d'appel de Chambéry (12 juillet 1905, D. P. 1905, 2, 417) je dirais:

Sans entrer dans les controverses doctrinales soulevées sur cette question par ceux qui veulent voir dans l'art. 1384 une présomption de faute, il y a lieu de reconnaître que, sainement compris, le point de vue auquel, selon nous, s'est placée la loi, est conforme à la justice et à l'équité, puisque celui qui détient une chose et en tire avantage doit, par suite d'une légitime réciprocité, supporter les charges corrélatives à cet avantage; qu'une théorie contraire, en cas de survenance d'accident, présenterait le grave inconvénient de méconnaître les garanties de sécurité et de réparation auxquelles, dans une société bien organisée, a droit la personne humaine.

[Page 288]

Je ne puis admettre que le mot "faute" soit employé dans l'article 1054, par.1, du code civil, pour signifier exclusivement un acte illicite et dommageable du fait de l'intention ou de la négligence. Saleilles dans "Les accidents du travail' dit (p. 69):

Je rappellerai tout d'abord certaines définitions anciennes qui ne laissent apparaître que le caractère purement objectif de l'idée de faute. Je les trouve dans Doneau, le grand initiateur parmi les anciens. Il définit la faute dans des termes auxquels notre article 1382[5] semble avoir été emprunté et qui, pas plus que lui, ne laissent apparaître aucune idée de recherche subjective. Cest tout fait non prévu et exercé sans droit qui a causé dommage à autrui, culpa est omne factum inconsultum quo nascitur alii injuria; donc une qualification matérielle du fait, un fait qui n'a pas été prévu, et l'on sous-entend qu'on aurait pu prévoir, et un fait qui ne soit pas l'exercice d'un droit positif.

Au surplus, quand on lit cet article 1054 C.C., il est impossible d'étendre le sens du mot "faute" aux objets inanimés. On ne saurait supposer que les rédacteurs du code aient jamais voulu dire que toute personne est responsable du dommage causé par là faute d'une chose inanimée dont elle a la garde; l'expression ne serait pas juridique.

Je ne puis non plus interpréter les derniers mots du paragraphe en question dans le sens que celui qui a la garde ou le soin d'une chose n'est responsable des dommages qu'au cas où l'on prouve que l'accident résulte ou peut résulter d'un défaut de construction dans l'objet, ou du fait de son fonctionnement. Ces mots pris dans leur sens littéral expriment une vérité juridique que l'on retrouve dans toutes les législations "rien de ce qui appartient à quelqu'un ne peut nuire impunément à un autre."

La partie prétendue responsable peut n'avoir ni la connaissance du défaut de construction, ni le moyen de s'en rendre compte; mais, si elle en a le soin et la

[Page 289]

garde, alors, d'après les termes de l'article, elle est responsable des dommages causés par la chose dont elle a la garde. Cette interprétation qui applique la même règle de responsabilité et au propriétaire ou gardien d'une chose inanimée, et au propriétaire d'un animal, en vertu de l'article 1055, est la plus raisonnable du monde.

Si, au lieu d'une machine, un animal eût causé le dommage dont on se plaint ici, il y aurait eu présomption de responsabilité contre le propriétaire. Au nom de quel privilège pouvons-nous établir une distinction entre la chose mobilière inanimée et la bête sans raison? (Planiol, vol. 2, p. 283, no. 917, fin). De plus, pourquoi n'y aurait-il pas, dans le cas du gardien d'une chose, la même présomption que celle qui existe dans le cas de celui qui a la tutelle d'un enfant, d'un aliéné; la surveillance d'écoliers, d'apprentis? Toutes ces personnes sont visées par le même article. Il n'est pas là question de faute. Le fait de la tutelle ou de la surveillance est le seul motif qui lie leur responsabilité et le seul point sur lequel le gardien, le parent, le tuteur ou le maître soient tenus en droit de rendre raison. La responsabilité sans la faute n'est pas inconnue au code civil de Québec; par exemple, outre le cas visé par l'article 1055, l'article 1487 stipule que le vendeur d'une chose qui ne lui appartient pas est responsable des dommages à l'égard de l'acheteur, sans allusion aucune à la faute.

Quant à l'argument tiré de la théorie de l'ancien droit français, tel qu'énoncé par mon collègue le juge Duff, citant Esmein, je m'en rapporterai à Baudry-Lacantinerie, "Obligations," vol. 3, n. 2968.

M. Esmein[6] ajoute que, suivant toutes les vraisemblances, les rédacteurs du code n'ont admis que les cas de responsabilité du propriétaire,

[Page 290]

déjà connus dans notre ancien droit, c'est-à-dire la responsabilité du propriétaire d'un animal et celle du propriétaire d'un bâtiment. Et il renvoie aux ouvrages de Bourjon et de Domat. Selon nous, si nos anciens auteurs n'ont pas fait d'autre application du principe général, il ne faut point en conclure qu'ils n'ont pas reconnu l'existence de ce principe, mais simplement que les applications prévues par eux étaient les seules qui, de leur temps, eussent un intérêt pratique. Il est incontestable qu'au XVIIe et au XVIIIe siècles, les choses mobilières dangereuses étaient incomparablement moins nombreuses qu'à notre époque. Néanmoins, avant de parler de la responsabilité du fait des animaux, Domat pose la règle dans les termes les plus généraux. La façon dont il s'exprime mérite qu'on la remarque: "L'ordre qui lie les hommes en société, dit-il, ne les oblige pas seulement à ne nuire en rien par eux-mêmes à qui que ce soit, mais il oblige aussi chacun à tenir tout ce qu'il possède en un tel état que personne n'en reçoive ni mal ni dommage; ce qui renferme le devoir de contenir les animaux qu'on a en sa possession, de sorte qu'ils ne "puissent ni nuire aux personnes, ni causer dans leurs biens quelque perte ou quelque dommage." On voit que, pour Domat, la responsabilité du fait des animaux n'était qu'une application de la responsabilité du fait des choses en général. Cela résulte nécessairement de ces mots "ce qui renferme."

Je l'accorde; autrefois, en France, la doctrine absolument en vigueur voulait qu'il n'y eût responsabilité que là où la faute était clairement prouvée. D.1870, 1, 361. Mais cette doctrine, attaquée vivement et depuis longtemps, par d'éminents juristes tels que Laurent, Labbé, Lyon-Caén, Glasson, Esmein, Saleilles, Josse-rand, Marcadé, Demolombe, Baudry-Lacantinerie et Hue est définitivement abandonnée. Il est admis, aujourd'hui, par tous les tribunaux français que quand, comme dans notre cas, un accident arrive à un ouvrier, la responsabilité retombe sur le propriétaire de la machine qui a fait le dommage; sauf dans les cas d'événements dus au cas fortuit ou dans ceux de force majeure; ou encore lorsque l'accident est imputable à la faute de la personne lésée. D.1905, I., 417; D. 1908, I., 217. Dans une note relative au dernier arrêt, Josserand donne ce commentaire:

L'arrêt ci-dessus rapporté marque une nouvelle et importante étape de l'évolution: dans les motifs de sa décision, mais très nettement,

[Page 291]

la chambre des requêtes reconnaît qu' "aux termes de l'art. 1384 c. civ. on est responsable du dommage causé par le fait des choses que l'on a sous sa garde." Et si cette proposition pouvait paraître obscure, sa signification serait mise en pleine lumière par la suite de l'arrêt:" * * * Cette présomption cède * * *," ajoute la cour de cassation. C'est donc bien une présomption de responsabilité que la chambre des requêtes lit dans l'article 1384, une responsabilité par le fait des choses, donc une responsabilité libérée—partiquement tout au moins et quant à la preuv jadis imposée à la victime—de l'exigence de la faute aquilienne.

Préalablement la cour de cassation, chambre civile, s'était prononcée clans le même sens (S.V. 1897, I., 17). Le tribunal constate bien dans ce cas l'existence d'un vice de construction dans la machine; mais elle ne le fait que pour écarter l'hypothèse du cas fortuit ou de la force majeure et non pas pour placer son arrêt sous la protection de l'article 1386 C.N. (1053 C.C.) Sourdat "Responsabilité" (5 éd.), Nos. 1432 (ter) par. 3, et 1483 (ter) par. 3.

C'est seulement dans ces dernières années que les tribunaux français ont appliqué le principe de l'article 1054 C.C., concernant la responsabilité du gardien d'un chose inanimée non immobilière, à des cas comme celui qui nous occupe. A la fin de l'al. 1er de l'art. 1384, C.N., se trouve un petit membre de phrase qui, dit Planiol, vol. 2, n° 916, est resté à peu près inaperçu pendant près d'un siècle. Il y est dit qu'on est responsable du dommage causé "par le fait * * des choses que l'on a sous sa garde." Autrefois le principe de la responsabilité était censé reposer sur l'article 1053 ou sur l'article correspondant du code Napoléon. Ce fait peut, dans une certaine mesure, expliquer l'incertitude qui a régné jusqu'ici dans la jurisprudence française. Pour les fins de cette cause, j'adopte donc l'interprétation que la cour de cassation de France a, dans ses décisions récentes, donnée à l'article 1384 du code

[Page 292]

Napoléon auquel correspond l'article 1054 du code civil. J'avoue qu'en ce faisant, je m'écarte de l'ancienne jurisprudence des tribunaux de Québec. A la suite de la jurisprudence française du temps, ils soutenaient que la faute est la condition et la mesure de la responsabilité.

Au moment de la première promulgation du code civil français, en 1804, il était impossible à ses auteurs de prévoir les développements extraordinaires que l'industrie moderne allait emprunter aux applications de la vapeur et de l'électricité et le nombre infini de eus où l'ouvrier a cessé d'être le maître pour devenir lui-même l'outil de cet immense ensemble de machines qui de nos jours forme un établissement industriel moderne. Les progrès, les exigences de l'industrie, qui, d'après les juristes français, ont contribué à la transformation de la jurisprudence française, existent aussi chez nous. Ils justifieraient, dans la même mesure que là-bas, l'adoption par notre cour des principes de la nouvelle jurisprudence qui faisant une juste application des mots employés par nos codificateurs, et qui paraissent avoir été mis en oubli au Canada comme en France, reconnaît les garanties de sécurité et de réparation auxquelles dans une société bien organisée a droit la personne humaine. En nous écartant de l'ancien système d'interprétation, nous suivrions le précédent créé, dans notre cour, en ce qui touche la répartition des dommages-intérêts pour le cas où le demandeur aurait de sa part été coupable de négligence. Autrefois, à Québec, l'employé coupable par négligence de participation à l'accident ne pouvait même pas partiellement obtenir gain de cause (Canadian Pacific Railway Co. v. Cadieux[7]; Desroches v. Gauthier[8], jugements de Dorion JVC. et

[Page 293]

Ramsay J.); mais les décisions des tribunaux ont changé cela. Notre cour, suivant la jurisprudence française si équitable a, par un jugement définitif, statué que, lorsque les deux parties sont en faute, les dommages-intérêts sont partagés proportionnellement à la faute de chacune (Price v. Roy[9]).

Toute la question est si pleinement développée dans les jugements savants et approfondis de mes collègues qu'il m'est inutile d'insister davantage. Je dirai seulement à ceux qu'intéresserait une étude plus détaillée de la question qu'ils peuvent suivre l'évolution de la jurisprudence française dans les causes suivantes que je joins à celles déjà citées: Cour de Cassation, 16 juin 1896 et 30 mars 1896 (Dal. 1897, 1, 433), note Saleilles; Sir. 1897,1, 419, note Esmein; C. de Paris (6e ch.), 5 nov. 1904. (Rec. Gaz. des Tribunaux, 1er avril 1909) et plus particulièrement dans les deux œuvres admirables de Saleilles "Responsabilité du fait des choses" et "Etude sur la théorie générale des Obligations," puis dans la "Revue Critique de la Législation," 1901, p. 592.

Attendu donc que le jugement attaqué a fait à la cause une exacte application des principes qui régissent la matière je déclare l'appel non recevable, dépens contre l'appelante.

Girouard J.—Je suis étonné qu'à cette époque de notre jurisprudence l'on ait encore à se demander ce que c'est que le quasi-délit. Voilà bientôt un demi-siècle que la code civil du Bas-Canada est en force et pendant ce long espace de temps, même avant, des centaines de procès causés par des quasi-délits, et particulièrement des accidents du travail, ont été examinés et étudiés devant les tribunaux; et cependant

[Page 294]

il paraîtrait, dit-on, que jusqu'à ces derniers jours avocats et juges étaient dans une profonde ignorance de la loi et que jamais ils n'ont donné aux articles du code le vraie interprétation. D'après quelques juges et, disons-le, ils paraissent très peu nombreux, la faute qui est la base du quasi-délit serait présumée dans certains cas prévus par l'article 1054 C.C.

On affirme que les dispositions de cet article sont différentes de celles du code français. Pour ma part, je ne vois de différence que quant à la rédaction ou phraséologie. Les codificateurs nous en ont avertis eux-mêmes. Quant à la substance, ils me paraissent semblables. Les deux codes sont basés sur la faute non pas présumée mais établie. L'article 1382 du code. Napoléon dit que

tout fait quelconque de l'homme qui cause à autrui un dommage oblige celui par la faute duquel il est arrivé à le réparer.

Le Code de Québec déclare que

toute personne * * * est responsable du dommage causé par sa faute à autrui * * *.

Le code ne peut exiger moins lorsque cette personne agit par des agents ou représentants. Aussi, l'article 1054 C.C., ajoute:

Elle est responsable non-seulement du dommage qu'elle cause par sa propre faute mais encore de celui causé par la faute de ceux dont elle a le contrôle et par les choses qu'elle a sous sa garde.

Puis le législateur énumère certains cas où certaines personnes répondent des actes de ceux dont elle a le contrôle, le père, la mère, les tuteurs, les curateurs, l'instituteur et l'artisan. Enfin l'article continue:

La responsabilité ci-dessus a lieu seulement lorsque la personne qui y est a sujettie ne peut prouver qu'elle n'a pu empêcher le fait qui a. causé le dommage.

Le texte anglais traduit "la responsabilité ci-dessus" par les mots "the responsibility in the above cases"

[Page 295]

et je crois qu'il exprime mieux la pensée du législateur. Il ne s'agit ici, en effet, que de certains quasi- délits énuniérés, où le législateur fait une exception en faveur du maître qui doit l'invoquer et la prouver. Il est inutile d'ailleurs de nous arrêter sur ce point, car la présente espèce n'est pas un de ces cas.

Il s'agit ici de la responsabilité des maîtres et commettants pour leurs domestiques et ouvriers

dans l'exécution des fonctions auxquelles ces derniers sont employés.

A l'égard de ces derniers, il importe peu qu'ils puissent prouver qu'ils n'ont pu empêcher le fait qui a causé le dommage. Qu'ils puissent le prouver ou non, ils demeurent responsables, si leurs préposés. étaient dans l'exercice de leurs fonctions.

Nulle part le code n'a exprimé la moindre intention de changer la nature de la responsabilité. Toujours et dans tous les cas elle résulte de la faute qui doit être établie par le demandeur, soit par une preuve directe ou par des présomptions.

C'étaient les dispositions du droit romain et du vieux droit commun de la France en force dans la province de Québec et que Ton trouve résumé dans Pothier, "Traité des Obligations," au titre des Délits et Quasi-Délits, Nos. 116 à 122. Pothier nous dit que

le quasi-délit est le fait par lequel une personne, sans malignité, mais par une imprudence qui n'est pas excusable, cause quelque tort à une autre.

Il ajoute que

on rend aussi les maîtres responsables du tort causé par les délits et quasi-délits de leurs serviteurs ou ouvriers qu'ils emploient à quelque service. Ils le sont de même dans le cas auquel il n'aurait pas été en leur pouvoir d'empêcher le délit ou quasi-délit, lorsque les délits ou quasi-délits sont commis par les dits serviteurs ou ouvriers dans l'exercice des fonctions auxquelles ils sont employés par leurs maîtres, quoiqu'en l'absence de leurs maîtres; ce qui a été établi pour rendre les maîtres plus attentifs à ne se servir que de bons domestiques.

[Page 296]

Ces principes ont reçu leur application dans la présente cause devant au moins deux cours, la cour supérieure et la cour d'appel. Elles ont considéré que les officiers préposés par rappelante au fonctionnement de ses fournaises n'étaient pas de bons ouvriers; qu'ils n'avaient pas pris les précautions nécessaires pour éviter l'accident dont le demandeur a été la victime. Leur jugement n'est aucunement appuyé sur la faute présumée:

Considérant que le jugement de la cour de première instance est bien fondé;

voilà le seul motif donné par la cour d'appel. Il n'y a que deux juges qui parlent de la faute présumée et encore ils n'invoquent pas cette raison comme étant la seule qui les engage à supporter le jugement de la cour; ils trouvent aussi que le demandeur a prouvé la faute de la part de la défenderesse. Ce jugement, comme celui de la cour de révision et de la cour supérieure, sanctionne les principes que nous venons d'énoncer. Voici le texte du jugement de la cour supérieure qui est confirmé en appel purement et simplement:

Considérant que d'après la preuve, les circonstances dans lesquelles l'accident est arrivé au demandeur établissent que les ingrédients avec lesquels la défenderesse fabriquait le carbure contenaient des matières explosibles qui, en faisant explosion dans la fournaise devant laquelle le demandeur travaillait rejetaient avec violence par l'orifice que le demandeur devait boucher avec du mortier qu'il poussait avec un tisonnier, le carbure à l'état liquide et enflammé.

Considérant que la défenderesse était responsable de cette fournaise qu'elle avait sous sa garde.

Considérant qu'en droit d'après une jurisprudence constante, les patrons ont le devoir de veiller à la conservation de leurs ouvriers et de les protéger contre les périls qui peuvent être la conséquence du travail auquel ils les emploient; que sous peine de faute, ils doivent prévoir les causes possibles d'accidents et prendre et faire prendre par leurs agents toutes les mesures de précautions pouvant les prévenir ou les éviter;

Considérant que, dans l'espèce, pour protéger le demandeur contre un accident comme celui dont il a été la victime, la défenderesse

[Page 297]

devait pourvoir le demandeur et les autres ouvriers qui faisaient le même ouvrage de lunettes et de masques;

Considérant que le demandeur n'était aucunement protégé contre l'explosion qui a eu lieu, sans qu'il y aît aucune faute de sa part, et qui a eu pour résultat de lui brûler la figure et de lui faire perdre complètement la vue.

M. le juge Cannon, qui a rendu le jugement, nous avoue qu'il a été guidé par la jurisprudence jusqu'alors suivie dans la province de Québec.

J'ai suivi (dit-il), surout une décision dans la cause de "Asbestos and Asbestic Company v. Durand"[10], décision de la cour suprême, dont le jugement de première instance, rendu par M. le juge Lemieux, a été confirmé par la cour d'appel, et confirmé plus tard par la cour suprême, rapportée au trentième volume des rapports de la cour suprême. Aussi une cause "La Corporation de la Cité de Montréal v. Gosney"[11], jugement rendu par M. le juge Lavergne. Ce jugement a été. confirmé par la cour d'appel. Voir aussi un arrêt dans Sirey pour l'annô 1897, premier volume. Cet arrêt paraissait si bien résumer notre loi, telle qu'interprétée par nos tribunaux et notre jurisprudence, que je me suis servi des mêmes termes dans le considérant où je détermine la responsabilité de la compagnie défenderesse.

En révision, Cimon, Pelletier and Lemieux JJ. ont renversé ce jugement, non pas parce que les principes qu'il applique étaient mal fondés, mais parce que la preuve faite démontrait que

le demandeur avait le contrôle et la garde des choses qui ont produit les dommages réclamés,

et que par conséquent s'il y avait faute, c'était la sienne propre et non celle de la défenderesse.

Il me semble, observe M. le juge Cimon dans une opinion très élaborée, que le dommage n'a pas été causé par une chose sous la garde de la défenderesse; mais que ce dommage est le résultat de l'opération faite par le demandeur, opération absolument sous son contrôle: il était responsable envers la défenderesse de cette opération, c'était à lui de surveiller, de soigner, de bien faire cette opération. Cette opération et les choses dont il se servait pour la faire étaient alors sous sa garde particulière.

[Page 298]

Puis le savant juge conclut en insistant sur l'application des principes énoncés plus haut:

Sous ces circonstances, la défenderesse ne peut être tenue responsable, à moins que le demandeur montre une faute spéciale de la défenderesse, faute qui aurait causé l'accident, et il n'y a aucune preuve de telle faute.

Et il ne faut pas oublier que la jurisprudence de la cour suprême, jurisprudence qui paraît ferme et absolument arrêtée, exige toujours de la part du demandeur, en pareil cas, la preuve d'une faute de la défenderesse, faute qui aurait produit le dommage.

En appel, (Taschereau J.C., Lavergne, Cross, Archambault et Carroll JJ.) le jugement a renversé celui de la cour de revision, non pas parce qu'il y avait faute présumée en loi, ce qui paraît être l'opinion individuelle de M. le juge Archambault et de M. le juge Carroll, mais parce qu'il y avait preuve de faute, ainsi que le juge Gannon l'avait jugé.

Remarquons bien que dans la cause de McArthur v. Dominion Cartridge Co.[12], le conseil privé n'a pas déclaré que la doctrine de la cour suprême[13], était trop absolue ou erronée, mais qu'elle avait fait une fausse appréciation de la preuve qui fournissait des présomptions de fait suffisantes pour justifier le verdict du jury, confirmé par deux cours en faveur du demandeur. Le conseil privé ne dit pas un mot de la prétendue faute présumée en l'article 1054 C.C.; c'était cependant le moment de le faire si elle était fondée.

Voilà ce que la jurisprudence de la province de Québec nous enseigne à l'unanimité, au moins jusqu'à ces derniers jours. L'on nous dit qu'elle ne répond plus à la situation du monde industriel. C'est possible. Mais qui doit donner le remède? Est-ce le juge ou le législateur? Notre code civil commande aux

[Page 299]

juges de suivre dans l'interprétation de ses articles les lois en force lors de sa promulgation[14]; et pour ma part je ne me laisserai certainement pas guider par les théories de quelques auteurs contemporains qui paraissent dominés par des raisons qui nécessiteraient un changement de la loi. Nous sommes ici non pas pour faire des lois, encore moins les changer.

On cite la décision de cette cour dans Price v. Roy[15]. Dans cette cause nous n'avons pas voulu changer l'ancien droit français ou les lois qui existaient lorsque le code a été promulgué; nous avons tout simplement rétabli une ancienne règle de droit qui, pendant quelques années, avait été ignorée par des juges trop imbus des principes de la loi commune anglaise qui ne donne aucune action dans le cas de négligence commune ou contributoire, principe tout à fait différent de l'ancien droit français et du droit romain. A tout événement, et ceci n'est pas sans importance, dans Price v. Roy15, cette cour n'a pas tenté de renverser sa propre jurisprudence, comme nous sommes invités à le faire dans la présente espèce, bien qu'elle soit consacrée par au moins une demi-douzaine de décisions.

Nous voilà donc en face de deux jugements rendus par deux cours sur une question de fait, savoir, la faute ou la négligence de la compagnie appelante, tel que développée par le juge Cannon dans le texte de son jugement. Dans ces circonstances, je ne crois pas que j'aurais raison de juger qu'il n'y avait pas faute et de renverser ces deux jugements. Voir Lodge Holes Colliery Co. v. Wednesbury Corporation[16], page 326.

Je suis donc d'avis de renvoyer l'appel avec dépens.

[Page 300]

Idington J.—Whether the doing so or not rested upon him the respondent having duly shewn to the satisfaction of the learned trial judge that he had followed such instructions as his employers gave him and was otherwise blameless, the questions raised herein are reduced to a narrow compass.

Passing for the present other alternatives about to be referred to, it seems to me that we must infer upon the general evidence given of the relevant surrounding circumstances, coupled with such a finding that the explosion was the result of a defect in the thing or things in appellant's charge, or the result of faulty methods (by which I mean such as due care had not been taken to avoid fault in respect of), in its use of them or some of them. In this I include, of course, all done by those for whom appellants stand responsible.

We have no serious effort on the part of appellants to determine and inform the court of the true cause of the explosion. I, therefore, conclude this suggested inference stands good.

If one lawfully passing upon the highway were injured by a building falling for which fall there was no apparent cause except inherent defects, I do not think it would be necessary in an action for damages to determine exactly whether the fall thus obviously resulting from age or ill-construction came about by reason of defect in the nature of the mortar which had been used in its walls or from too great weight of superstructure having been placed thereon by the builders selected by the owner, or otherwise exactly what was the cause thereof.

Assuming, for argument's sake, the much discussed part of article 1054 C.C. blotted out, how does the proof of liability arising from the use of a machine,

[Page 301]

perhaps a new or untried machine, causing actionable damage differ in regard to the nature of proof from what is needed in the case of the falling house? The Code deals with each of them in separate articles which, I take it, are, though of different origins, subsidiary to and illustrative of the comprehensive rule that precedes both in article 1053 C.C.

Surely the good sense and reason embodied in the maxim "res ipsa loquitur" may, in either case, be allowed operation and suffice to solve the questions arising when, once general evidence has been given, reducing the question to one of fair inference.

With great respect, it seems to me idle to suggest that this oven or furnace or both and all connected therewith were, as between the parties hereto, in charge of the respondent, in the sense claimed herein and so as thus to relieve the appellants.

If, on the other hand, it be said the inference above suggested is unwarranted because the work of the respondent was known to be of a dangerous character and explosions were to be anticipated as something not unlikely to happen, I submit that, in such case, there clearly arises the alternative inference that the nature of the respondent's work was known to the appellants, or by due care of the appellants ought to have been known to them, to involve just such risks of explosion, or as the evidence suggests, were of an experimental character in the handling of what were highly dangerous forces likely or liable to produce explosions of a dangerous character, and, therefore, in any such alternative or alternatives, of a like nature, the appellants had clearly neglected the plain duty of giving the necessary instructions and warning and (or) of furnishing such adequate protection as obviously would be necessary in such a case.

[Page 302]

In any way we can look at the case, unless the learned trial judge's finding in regard to absence of fault in the respondent be set aside, which is not suggested, the conclusion seems irresistible either that the learned trial judge and the court below are absolutely right or, at all events, not so clearly wrong that we are entitled to reverse it.

The appeal should be dismissed with costs.

Duff J. (dissenting).—This appeal arises out of an action brought by the respondent against the appellants to recover damages for the injuries suffered by him in consequence of an accident in the works of the appellants, who are manufacturers of calcium carbide at Shawinigan Falls.

The trial judge gave judgment for the respondent; this judgment was reversed by the Court of Review[17], unanimously, but restored by the Court of King's Bench[18], (Cross J. dissenting).

The Court of King's Bench proceeded upon a view of the effect of art. 1054 C.C., which it will be necessary to examine, as well as upon the view that the injury of which the respondent complains was due to the fault of the appellants in failing in one manner or another to exercise the degree of care which the law requires in the protection of their employees.

Both of these views are in controversy on this appeal, and I proceed to discuss the questions thus raised in the order mentioned.

The view of art. 1054 C.C., upon which the court below acted, as indicated in the leading judgment (that of Archambault J.), is that the first paragraph

[Page 303]

of that article embodies a self-sufficient rule of general application—subject only to an exception to be found expressed in the 6th paragraph. Stated in the terms of these two paragraphs, the rule is as follows:

1054. He is responsible not only for the damage caused by his own fault but also for that caused by the fault of persons under his control and by things which he has under his care; * * * The responsibility attaches in the above cases only when the person subject to it fails to establish that he was unable to prevent the act which caused the damage.

This language, however, does not perhaps at first sight convey the full significance which in the opinion of the court below it really bears. The words "things which he has under his care" are, in this view, taken to embrace every inanimate thing not an immoveable used or worked for one's behoof, whether by one's self or by others; so that even though the complainant should have within his own hand the actual custody of the thing causing the harm in respect of which redress is sought, and even though he alone should consequently be in a position to give an account of the train of events actually leading to the injury, yet the onus is upon the defendant to bring himself within the exception; and this principle applies of course not only where the relation of master and servant exists, but wherever one person suffers an injury which is "caused" by "a thing" that another "has under his care" in the sense explained. Such is the effect of the extracts from Mons. Josserand, at p. 275, and of the learned judge's own observations in the following passage:

Dans le présente cause, la fournaise appartenait à l'intimée elle était exploitée par cette dernière qui s'en servait en vue de réaliser des bénéfices dans son industrie. C'est elle, par conséquent, qui en avait la garde, et c'est elle qui est responsable des dommages qu'elle a causés.

[Page 304]

Nos annales judiciaires, et celles des autres pays, fourmillent de décisions rendue en matière d'accidents du travail dans lesquels un ouvrier, mécanicien ou autre, a été tué ou blessé par l'explosion ou par le fonctionnement d'une machine ou d'un mécanisme, et jamais on n'a songé à dire que la machine ou la chose se trouvait sous la garde de l'ouvrier et non sous celle de la compagnie ou de la personne qui l'exploitait ou qui s'en servait pour son avantage et pour son profit.

I shall first give my reasons for thinking that this construction of article 1054 C.C., is not in harmony with the law of delicts and quasi-delicts as it has actually been expounded and applied by the Court of King's Bench and by this court in passing upon appeals from the Province of Quebec.

In examining the decisions of the provincial courts it is necessary to keep clearly in view the distinction between decisions based upon a statutory presumption, or presumption arising from some specific legal rule such as that which the court below finds declared in art. 1054 C.C., and decisions based upon inferences or presumptions which can be justified only as inferences of fact legitimately arising out of the facts established by the evidence. Whether in a given case a presumption of the last mentioned character is or is not well founded is not a question of law at all; but merely a question of sound reasoning to be tried by the same tests, whether the tribunal sit to administer the civil law, or to administer the common law; and judicial decisions, in so far as they involve inferences of fact only, although often useful as affording illustrations of judicial methods are not in the strict sense, legal precedents at all. Some misapprehension as to the real bearing of the decisions of the Court of King's Bench has, I think, arisen from not attending sufficiently to this distinction.

Of the decisions of the Court of King's Bench, I

[Page 305]

will take three—each separated from the others by a considerable interval of time—and all dealing with the precise point in question here, the question of the burden of proof where an employee claims reparation for loss suffered in consequence of an injury inflicted by an explosion or otherwise through the immediate agency of something used for the benefit of the employer in the business in which the employee was engaged.

St. Lawrence Sugar Refining Company v. Campbell[19] in 1885. The grounds of this decision are stated by Dorion C.J., page 295, in these words:

There is no responsibility unless there is fault, and fault must be proved. Here, two men were engaged in some work for the company. There is no proof that there was any danger in what they were doing. But an explosion occurred by which the plaintiff was injured, and the other workman killed. There is not a single witness who states how the accident occurred, nor is there anything to shew that the company was in fault, or that the other workman was in fault. It cannot be presumed that the accident occurred through their fault. The respondent's action, therefore, must fail in the absence of any evidence adduced by him to support it.

In a subsequent case, Dominion Oil Cloth Company v. Coallier[20], Dorion C.J., at page 269, states the ratio of the decision just cited in these words:

The respondent was injured by an explosion, and there was no evidence of the cause of the accident, or of fault on the part of the employers, the judgment was also reversed and the action dismissed.

In Mercier v. Morin[21], in 1892, it appeared that the plaintiff had been injured by the collapse of a "chaussée" on which he was working. The Court of Queen's Bench held that the employers, in the absence of evidence that the accident was due to some fault on their part, were not liable. Bossé J., delivering the

[Page 306]

judgment of the majority of the court (Baby, Bossé, Blanchet and Wûrtèle JJ.), said:

Dans ces circonstances, la règle me parait facile à appliquer. Elle résulte des articles 1053 et suiv., l'impéritie, la négligence et l'inhabilité de toute personne capable de distinguer le bien du mal, si ce fait, cette impéritie, cette négligence ou cette inhabilité ont causé des dommages, créant l'obligation de réparer des dommages. Tel est notre texte, et, dans l'application de ce principe, il a été constan-ment et avec raison jugé que le maître est tenu de fournir, aux ouvriers qu'il emploie un local sûr et machines, engine et outils construits et aménagés suivant les règles de l'art et de la prudence. Ainsi, un cable défectueux, une bouilloire mal aménagée, des lisses de chemin de fer mal placées, et généralement un vice de construction ou d'installation quelconque, cause du dommage, que le maître connaissait ou aurait dû connaître, entraînant responsabilité. Nos tribunaux en ont fait à maintes fois l'application.

Mais lorsqu'il n'est pas prouvé qu'il y a eu vice de construction ou d'installation; lorsqu'il y a absence de preuve de faute ou de négligence, notre jurisprudence a toujours déclaré qu'il n'y a pas lieu au recours en dommages par l'employé contre le maître.

Pas de preuve de faute ou de négligence, pas de lien de droit et pas d'action.

En ce sens, vide Dominion Oil Cloth Co. v. Coallier[22]; St. Lawrence Sugar Refining Co. v. Campbell[23]; Compagnie de Navigation du Richelieu et Ontario v. 8t. Jean[24]; Lavoie v. Drapeau[25].

In Canadian Pacific Railway Co. v. Dionne (in 1908)[26], the precise point before us was formally decided by the court of appeal (Taschereau C.J., Bossé, Trenholm, Lavergne, Cross JJ.). The plaintiff's husband, an employee of the Canadian Pacific Railway Company, was run down by a shunting locomotive at a point near a place where he had been engaged in shovelling snow for the company. At the trial, before Langelier C.J., the company was held liable, the ground of liability being thus expressed:

Considering that on the 18th of February last the said Elzéar Doinne had been for several years in the employ of the defendant

[Page 307]

to work in its yard near its station at Quebec, and that on the day in question he was working at shovelling snow in the said yard under the order and direction of one Ouellette employed by the defendant as foreman of this work; considering that while the said Elzéar Dionne was in the said defendant's yard he was killed by a locomotive owned by the defendant and under its care.

On appeal this judgment was reversed and the principle upon which the Court of King's Bench proceeded was stated in the formal judgment as follows:

Considering that the respondent failed to prove the material allegations of her declaration, and in particular, failed to prove that the death of her husband, Elzéar Dionne, which occurred at Quebec on the 18th February, 1907, was due to fault, negligence or imprudence on the part of the appellant or of its servants:

Considering that, before defendant from whom damages are demanded by reason of quasi-offence on his part or on the part of his servants, can be held responsible, it is incumbent upon the plaintiff to establish affirmatively not only the existence of the damage claimed, but also the fault, negligence or imprudence on the part of the defendant, and that the fact of the injury alleged having been caused by a thing under the control of the defendant, has not in law of itself the effect of placing upon the defendant the burden of proving that the injury was caused without fault on the part of the defendant or his servants.

The decisions in the Province of Quebec relied upon in the judgment of the learned judge who delivered the judgment of the majority in the court below do not, with great respect, appear to me to impugn the principle stated in the cases to which I have just referred. The observations of Andrews J. in Dupont v. Quebec Steamship Co.[27], at page 194 (a decision of the Court of Review), are, it is true, capable of construction supporting the learned judge's views, but it is a well-known principle that such observations must be construed with reference to the facts of the case to which they relate—secundum subjectam materiam; and the facts of that case appear to have

[Page 308]

afforded ample ground for drawing the inference that the tackle which failed had been negligently set up, or that there had been some lack of inspection, and that, I think, is in substance the process by which the Court of Review reached its conclusion.

The decisions referred to by Andrews J. in the course of his opinion (Ross v. Langlois[28], and Corner v. Byrd[29]) lend support to this reading of it. In Boss v. Langlois28 the decision was put expressly upon the ground that the defendants had used defective tackle, without taking the care, before using it, to ascertain its condition. In the judgment of Jetté J. (see page 284 of the report), there is the finding that the hook which gave way was of

bad quality, in a bad state, and insufficient to support without danger, the burden put upon it; and, moreover, that the condition of the hook had not been ascertained before using it for the purpose to which it was applied.

That is a virtual finding of negligence; and with that finding the court of appeal agreed. In Corner v. Byrd29, the ground of the decision is stated by Dorion C. J., at page 271, in these words:

It was one of those accidents for which the appellant is liable, because it could have been prevented by care on his part.

To come to the decisions of this court: In Paquet v. Dufour[30], Girouard J. states the law conformably to the second considérant quoted from the judgment of the Court of King's Bench in Dionne v. Canadian Pacific Railway Co.[31], and to the views expressed in the judgment of Dorion C.J. and Bossé J. in the two preceding cases.

Before closing, I wish (says the learned judge), to point out a considérant of the trial judge to which I cannot subscribe:

[Page 309]

"Considérant que le dite explosion ayant été causée par de la dynamite dont le défendeur était le propriétaire et dont il avait la garde, il doit être tenu responsable des dommages qui en sont résultés pour le demandeur, à moins qu'il n'ait prouvé qu'il lui a été impossible de l'éviter."

We have so often decided in our court that proof of fault, whether by direct evidence or by presumptions, rests upon the plaintiff, that it is not necessary to quote authorities.

This passage, I think, with great respect, summarizes with accuracy the effect of a long series of decisions in this court on appeals from the Province of Quebec. I will mention some of them. In Cowans v. Marshall[32], it was shewn that the plaintiff while employed in the defendants' iron-works was injured in consequence of an explosion caused by molten lead coming in contact with oakum in a wet condition. This raised precisely the point in hand, the accident being exactly one of that class to which, if the view under discussion be a sound one, the suggested presumption would apply. It was held, however, that the onus was on the plaintiff to prove negligence, and that onus not having been satisfied the complainant must fail. In The Canada Paint Co. v. Trainor[33], it was shewn that the plaintiff had in some unexplained manner come in contact with the machinery of a printing press at which she was employed in the defendants' establishment, and the defendants were charged with failure to make proper provision for the protection of their employees. The Court of Queen's Bench held that the plaintiff had made out her case by establishing affirmatively such want of care. Here again it is clear that upon the proposed construction of art. 1054 CO., the defendants would be charged with the onus of relieving themselves from the presumption of negligence by shewing that they could not have prevented

[Page 310]

the accident. This court held, however, (directly negativing the view now advanced) that although the plaintiff had unquestionably been injured by the machine, she had failed to satisfy the burden upon her to shew that the injury was due to some fault of the defendants.

In Dominion Cartridge Co. v. Cairns[34], it appeared that the plaintiff was injured by an explosion originating in the pressing machine of the defendants, who were manufacturers of cartridges. This court reversed the judgment of the Court of Queen's Bench, and held (a ruling quite irreconcilable with the theory under review) that the onus was on the plaintiff to trace the explosion to some fault of the defendants.

It is unnecessary to multiply examples; the course of decision has been clear and uniform; but there remains the question of the effect of the decision of the Privy Council, in McArthur v. Dominion Cartridge Go.[35], which it appears has been assumed in some of the decisions of the Quebec courts to be inconsistent with the views on this point to which this court has consistently given effect.

This court had held in that case that no negligence had been proved—that there was no direct evidence, and nothing upon which a presumption of fact could be founded. It had been further held that assuming negligence, the rule applied by the French courts in such cases requiring certain proof of a causal relation between a fault on the part of the defendants and the injury complained of was sufficient to defeat the plaintiff's claim. In the judgment of the Privy Council delivered by Lord Macnaghten it is not suggested that the onus was not upon the plaintiff to shew that

[Page 311]

the injury was due to some fault of the defendants; on the contrary, the judgment proceeds upon the assumption that it was. It was held, however, that there was sufficient evidence to justify a verdict by the jury that the defendants had been negligent and that this negligence was the cause of the explosion to which the injury was attributable.

At the same time the Privy Council negatived the existence as a part of Quebec law of any rule (of general application) of the character indicated above (that the law exacts "proof of a fault which certainly caused the injury"); holding it to be sufficient to adduce evidence from which the tribunal may fairly infer both the existence of the fault and the connection between that fault and the injury complained of.

There is nothing, therefore, in the decision in that case to cast any doubt upon the principle upon which the Court of King's Bench and this court have hitherto acted.

I should not be prepared to give my adherence to the view that it is open to this court to accept (as a part of the law of Quebec) a principle the exact reverse of that upon which this mass of high judicial authority is based; but assuming the question raised to be still open for examination it would require very cogent reasons to lead me to hold that all these decisions are erroneous and must be overruled. A careful examination of these reasons adduced in support of this view satisfies me that the great weight of argument lies in the opposite scale.

The first step in considering the construction accepted in the court below is to ascertain what is the proper method of interpretation. I do not think it is a proper method to apply one's self to an examination

[Page 312]

of the language of art. 1054 C.C., in such light only as may be had from an exclusive consideration of the words of the article itself, and the immediate context, disregarding the history of the law on the subject of which it treats. In Robinson v. Canadian Pacific Rway. Co.[36], at page 487, Lord Watson, speaking for the Judicial Committee, applied to the Civil Code generally the rule enunciated by Lord Herschell in Bank of England v. Vagliano[37], at page 144, the effect, of which is that when the Code contains provisions of doubtful import, resort is to be had to the preexisting law. The previously uniform current of decision of the two Canadian courts of appeal having jurisdiction over the Province of Quebec in the sense opposite to the decision below affords at least some ground for thinking that the construction upon which that decision is based does not yield the single exclusive meaning attributable to the language employed, and prima facie, that in itself would seem in accordance with Lord Watson's canon to be a good reason for not ignoring the pre-existing law.

A far stronger reason against excluding the preexisting law from consideration is afforded by the terms of the enactments under the authority of which the Code came into force as law which evince very plainly the intention to declare, in arts. 1053, 1054 and 1055, the law as it then stood. There was first an Act of the Province of Canada (20 Vict. ch. 43), authorizing the appointment of commissioners and directing that they should embody in the Code to be framed by them, to be called the Civil Code of Lower Canada, such provisions as they should hold to be then actually in force, giving the authorities on which their views

[Page 313]

should be based, but stating separately any proposed amendments. Then (the Commissioners having in due course framed their report and laid it before Parliament), there was another Act (29 Vict. ch. 41) declaring a certain roll attested in the manner described in the Act to be the original of the Civil Code reported by the Commissioners as containing the existing law without amendments; directing the Commissioners to incorporate in this roll certain specified amendments eliminating and altering the provisions of it only so far as should be necessary to give effect to these amendments; and providing that the Code so altered should, on proclamation by the Governor, have the force of law.

It hardly seems necessary to comment on the effect of this legislation. It very manifestly exhibits the intention of the legislature that the provisions found in the roll referred to were not, excepting in so far as they should be affected by the amendments specified, to effect any substantial alteration in the law then actually in force in Lower Canada. Among the provisions contained in this roll (and untouched by the amendments sanctioned) are arts. 1053, 1054, and 1055 C.C.; and in construing them we have therefore this clear and important guide to the intention of the legislature.

It is proper to observe that in Robinson v. Canadian Pacific Railway Co.[38], the Privy Council had to consider art. 1056 CO., which does not appear in the report of the Commissioners and to which, therefore, these considerations do not in their full force apply. Furthermore, in the construction of such a statute as the "Bills of Exchange Act," which the House of

[Page 314]

Lords had to consider in Bank of England v. Vagliano[39], one has no such key by which to ascertain whether the legislature in enacting a particular provision intended thereby to change or leave unchanged the existing rules of law upon the topic dealt with; and, consequently, one can only proceed upon the construction of the words themselves. In such a case the pre-existing law is an extraneous thing and only as an extraneous circumstance can be brought into play for the purposes of interpretation; while on the contrary in the case before us the legislature has itself, in effect, declared its purpose of embodying the existing law in the provision to be construed.

I am not disposed to question that if these articles were framed in language reasonably capable of only one meaning, it is not easy to see how, conformably to the postulate of the constitution touching the supremacy of Parliament, a judicial tribunal could refuse to accept as the law existing when the Code was framed that which in unmistakable terms a competent legislature had declared that law to be; but neither do I doubt that if the state of the law these articles were intended to declare be not itself seriously open to question, and the articles can reasonably be read in such a way as to bring them into conformity with it, we cannot be required to give them a reading (though the words themselves should be capable of that reading also) out of harmony with the ascertained intention of the legislature.

It is probably unnecessary to consider at length the state of the common law of Quebec upon this point at the time of the adoption of the Code. It will not be disputed that the passages already quoted from Dorion

[Page 315]

C.J. and Girouard J. and Bossé J. state it correctly. In France there has been a great deal of discussion upon the construction of the corresponding article (1384) of the Code Napoléon; but all schools of interpretation seem to be agreed in this that under the common law non-contractual liability was based upon delict or quasi-delict, and except in the specific cases provided for in art. 1055 C.C. (1385 C.N.), the burden of proof was upon the complainant.

Mons. Esmein, the eminent authority on the history of French law, in a note, Sirey, 1897, 1, 17-19, states the traditional view thus:

A. Le patron propriétaire se pourrait être responsable envers ses employés de l'accident survenu qu'autant que celui-ci résulterait de sa faute personnelle ou d'une faute de ses préposés. C'est l'obligation délictuelle ou quasi-délictuelle fondée sur les art. 1382 et 1383 C. civ. La faute alors doit être démontrée, et le fardeau de la preuve incombe à celui qui réclame les dommages-intérêts, * * *.

Notre ancien droit français ne connaissait que deux cas dans lesquels le propriétaire était tenu, par une obligation quasi-délictuelle, de réparer le préjudice causé par la chose. C'étaient "l'action contre celui dont le bâtiment, par sa chute totale ou par portion, a blessé quelqu'un," et "l'action contre le maître d'un animal qui a causé quelque dommage." (Bourjon, Le Droit commun de la France et la Coutume de Paris, liv. 6, tit. 2, ch. 6 et 7; cf. Domat, Les Lois civiles, tit. 8, sec. 2, et. 3.)

Pothier ne parle ni de l'un ni de l'autre cas dans son Traité des Obligations. Sans doute, on entendait assez largement la chute d'un bâtiment; on donnait l'action lorsque la blessure avait été causée par la chute d'un entablement, ou d'une seule tuile, mais on n'allait pas plus loin; la responsabilité du propriétaire n'était point étendue à d'autres hypothèses.

Indeed, for the greater part of the 19th century, such was the effect which the French tribunals gave to the provisions of the Code Napoléon dealing with this topic. Mons. Josserand, at pp. 5 and 6 of his monograph "Responsabilitité du fait des choses inanimées," says:

Au contraire, la dernière source de responsabilité, le fait des choses inanimées, a été fort négligée jusqu'à ces dernières années.

[Page 316]

On reconnaissait bien que nous sommes responsables parfois du dommage causé par les choses dont nous avons la propriété ou la garde; l'article 1384, 1°, le déclare formellement. Mais sans creuser la nature de cette responsabilité, sans s'arrêter même au texte fondamental qui en consacre le principe, on se contentait de lui appliquer le droit commun en maitère de délits; on faisait rentrer la responsabilité du fait des choses dans la théorie plus large de la faute Aquilienne, l'article 1384, 1°, n'étant ainsi qu'une application de l'article 1382, le dommage causé par notre chose n'étant pas autrement traité que le dommage directement causé par notre personne même. C'était le triomphe complet de la faute délictuelle, le règle souverain de l'article 1382 qui commandait ainsi au chapitre II. tout entier. La théorie avait donc le mérite de l'unité et les conséquences en étaient claires; le victime d'un accident occasionné par une chose inanimée ne pouvait obtenir une indemnité que si le propriétaire avait commis une faute, et seulement à la condition d'établir cette faute; actori incumbit probatio.

Pendant trois quarts de siècle on se contenta de cette conception.

The view of article 1384 C.N., advocated by M. Josser and and Mons. Saleilles—the doctrine of "le risque professionnel"—proceeds upon the hypothesis that the common law was altered by the positive enactments of the Code Napoléon. Josserand, "Responsa-bilité du fait des choses inanimées," pp. 90-95.

Such having been the state of the common law the question would appear to be concluded; unless indeed it can be maintained, in face of the decisions already cited, that the words of art. 1054 C.C., will not reasonably bear a construction consistent with the principles which ex hypothesi it was intended to embody.

That question I proceed to consider, first taking the English version of the article alone. The first paragraph of the article declares that every person

is responsible not only for the damage caused by his own fault, but also for that caused by the fault of persons under his control and by things which he has under his care.

Then follow four paragraphs declaring responsibility in the cases of parents, curators, tutors and artisans (for acts of apprentices) respectively:

[Page 317]

2. The father, or, after his decease, the mother, is responsible for the damage caused by their minor children;

3. Tutors are responsible in like manner for their pupils;

4. Curators or others having the legal custody of insane persons, for the damage done by the latter;

5. Schoolmasters and artisans, for the damage caused by their pupils or apprentices while under their care.

Assuming the first paragraph to state, as held in the court below, a rule of law of general application under which one is liable in every case for damages caused by the fault of persons under one's control— then nobody will be disposed to question that each of these four succeeding paragraphs states a special application of that rule. In other words, paragraphs 2 to 5 declaring the responsibility of parents for their children, and so on, embody particular cases within the general rule.

Coming now to paragraph 6; we have these words

The responsibility attaches in the above cases only when the person subject to it fails to establish he was unable to prevent the act which has caused the damage.

Some difficulties are to be noted here as arising from the construction proposed. First, if this paragraph (the 6th) embodies an exception which attends the rule stated in the first paragraph throughout its whole extent—and which applies, therefore, to cases falling under the rule other than the cases mentioned in paragraphs 2 to 5—why was the exception placed at the end of the 5th paragraph, and not stated at once as an exception to the rule? Secondly, why is the exception made applicable in express terms to the "above cases" only, and not stated as an exception to the rule itself? Both position and phraseology seem natural and appropriate, if the exception was to affect the four preceding cases alone; neither is natural or appropriate upon the alternative hypothesis.

[Page 318]

Again, how shall we read the seventh paragraph? Is that not a "case" under the general rule? If so—and it seems the more natural reading to treat it as such—are we to read the sixth paragraph as applying to it?

But the last words of the sixth clause seem to create a more serious and, I venture to think, an insurmountable difficulty. Under that clause the person sought to be charged may escape responsibility, in the oases to which it applies, by shewing that he was unable to prevent—not the damage—but "the act which has caused the damage."

The effect of these words in the view of the court below is that, primà facie, one's responsibility arises from the circumstance alone that damage occurs through the immediate instrumentality of something which is being used or worked for one's profit; and upon the construction proposed the phrase "act which caused the damage" must be read as descriptive of the fact that damage has thus occurred.

That is attributing to the words "act which caused the damage" a sense quite out of keeping with the true meaning of them; and is to assume on the part of the Commissioners who framed the article and the legislature which adopted it a very imperfect appreciation of the meaning of common English words. It is not necessary to dispute that if no other application for these words could be found one might accept this view rather than reject the words as wholly senseless. But here there is no such excuse. The words in their ordinary sense find their natural application when read as referring to the acts of the persons mentioned in the four paragraphs immediately preceding. Giving them this, their natural and obvious, application a

[Page 319]

perfectly consistent and intelligible construction is given to all the paragraphs from 2 to 6.

The exception should not then be attributed to the first paragraph, unless some reason for doing so is to be found outside the language in which the exception is framed. I think there is no such reason. It may, of course, be suggested that the first paragraph, if taken alone (freed, that is to say, from the operation of the exception declared in the 6th paragraph) would establish a legal rule still more widely at variance with the common law. The doctrine of le risque professionnel mentioned which has the support of many able French commentators (and has possibly at last received the sanction of the Cour de Cassation Dal., 1908, 1, 217), regards the corresponding paragraph of the Code Napoléon as a positive enactment imposing an absolute liability for all harm caused through the instrumentality of things owned by one and worked. for one's profit irrespective of proof or presumption of fault—a liability which one can only escape by establishing that the true cause of the harm was force majeure or the fault of the victim.

This, it may be argued, is the true effect of reading the first paragraph as freed from the exception declared in the 6th; and one must meet the question whether the words of the 1st paragraph are so intractable as to require this construction of them. No difficulty respecting this point would seem, however, to arise unless one lose sight of the office the 1st paragraph is designed to fill. It is sufficiently plain, I think, from what has already been said upon the history of the legislation, that the proper mode of approaching these articles is to regard them as an exposition of one topic in a coordinated system of law

[Page 320]

already in force—and not at all as a string of detached legal enactments. From this point of view the paragraph in question presents itself not as embodying a self-sufficient and self-operating legal rule, but as one step simply in the progress of the exposition. Thus art. 1053 C.C., explains that fault may consist in positive act or in omission and so on. Arts. 1054 and 1055 C.C., seem designated to deal with the instruments of damage in respect of which one may be answerable. Article 1054 is primarily concerned with one's responsibility for the acts of persons; but the first paragraph is most naturally read, I think, as introductory to the whole subject dealt with in the two articles; and it is seemingly utilized to state the broad general considerations which are the raison d'être of the positive legal rules stated in the subsequent paragraphs. One's responsibility may arise (so in effect the expositors proceed in that paragraph) not only from one's personal fault, but from the fault of those persons whom the law regards, for the purpose, as being under one's control, and out of harm caused by things in respect of which the law imposes upon one a duty to take care that they are not the instruments of harm to others. The personal relationships and the classes of things thus referred to are then specifically stated, the first in art. 1054 C.C., and the second in art. 1055 CO. This view is concisely put by Mons. Esmein, S. V. 1897,1,19.

Par le membre de phrase en question, le législateur, qui, dans l'art. 1384, ne dispose en réalité que sur la responsabilité du dommage causé par le fait des personnes, a simplement indiqué, par préoccupation de symétrie, qu'une responsabilité semblable résultait aussi du fait de certaines choses; et le dispositif sur ce second point se trouve dans l'art. 1385. Le membre de phrase discuté de l'art. 1384 ne contient point la règle qu'on prétend en tirer; il n'a pas disposé sur ce point, mais simplement annoncé la disposition de l'article suivant.

[Page 321]

And by Fromageot, "de la Faute," at p. 99:

Le premier alinéa de l'article 1384 du Code civil n'établit pas en effet, une disposition absolue et particulière se suffisant à elle-même; il ne fait que poser un intitulé des principes qui sont précisés dans les alinéas suivants pour les personnes dont on doit répondre, dans l'article 1385 pour les choses animées et dans l'article 1386 pour les choses inanimées.

And of M. Planiol, vol. 3, at page 303:

917. Origine et sens de cette disposition. Il est facile de montrer pourquoi cette idée a été énconcé à cette place et sous cette forme. Dans le projet primitif, les dispositions des articles 1384, 1385 et 1386 figuraient déjà à peu près telles qu'elles existent actuellement, mais les deux articles 1384 et 1385 n'en faisaient qu'un seul, qui débutait par une phrase générale, annonçant les dispositions particulières qui en forment les autres alinéas. La formule de l'alinéa 1er n'a donc d'autre but que d'énoncer le principe dont les dispositions suivantes sont les applications. Or, dans le projet de l'an VIII., il est visible que le législateur a pensé à la fois aux personnes et aux choses dont quelqu'un a la surveillance et la direction et qui peuvent être pour les tiers une source de dangers. La responsabilité est donc fondée, à raison des choses comme à raison des personnes, sur une idée de garde, c'est-à-dire sur une faute personnelle de celui qui s'est mal acquitté de sa tâche de surveillant. On ne peut faire sortir de ce texte l'idée d'une responsabilité indépendante de toute faute.

Cette fin de l'alinéa 1er de l'article 1384 était restée inapperçue pendant près d'un siècle; mais vers 1895 on a cherché à en tirer parti pour élargir la disposition particulière de Varticle 1386, qui paraissait trop étroite à certains tribunaux. J'estime que c'est à tort. Voyez cidessous nos. 927 et suiv.

So we are not driven to a construction out of harmony with the declared intention of the legislature. It is proper to add that we need not concern ourselves with the rival interpretations of the Code Napoléon which during the past twenty years have been so much debated in France. The key to the legislative intention furnished us by the legislature itself is wanting to the interpreters of that instrument; and we are not required to surmise what the result might have been, had we been called upon to construe art. 1054 C.C., so to speak in vacuo—disregarding both the state of the law when the Code came into force and the course of judicial decision since.

[Page 322]

On the other hand the views respecting the effect of those articles of the Code Napoléon upon which the provisions of the Quebec Civil Code were founded which had prevailed in France down to the time of the promulgation of the latter Code have, I think, a material bearing upon the construction of those provisions.

The view of the first paragraph of art. 1384 Code Napoléon above indicated, is that which (as Mon. Josser and says in the passage quoted)[40] prevailed in France for three-quarters of a century. It was the view which the French tribunals still applied when, in 1861, the Commissioners reported their draft of the provisions upon the title of "Obligations" and, in 1866, when the Code was promulgated. That in the verbal changes made by them in adopting the provisions of the French Code the Commissioners had no suspicion that they were affecting any substantial alteration of the law upon the point in question here is made quite clear by the passage in their report (first report, p. 16) referring to those provisions.

Here is the passage:

The articles of chapter III. "of offences and quasi-offences" correspond with the articles of the French Code, except that the wording has been changed to obviate certain objections raised to the latter; and in No. 74.(79) an addition has been made to the enumeration of cases to which the article applies. These are the paragraphs relating to tutors and curators of insane persons.

Except upon the inadmissible hypothesis that the distinguished Commissioners were deliberately practising a deception upon the legislature this is not the language of trained lawyers, who, having been instructed to present to the legislature a Code embodying the law then in force, were intentionally introducing a profound modification into the rules of law upon

[Page 323]

the subject of torts both as they were found in the existing law and as they were then understood to be declared by the instrument the Commissioners were in substance professing to adopt. It is hardly to be supposed that the legislature with this report before it would be keener than the codifiers to detect the change.

Indeed, it is no exaggeration to say that the assumption involved in the view of the court below is this: that this self-delusion of the Commissioners and the consequent delusion of the legislature respecting the true construction of art. 1054 C.C., were participated in for almost a generation in every quarter of the Province of Quebec; and that the Court of King's Bench, down to the year 1908, and this court, down to this moment, have continued to give decisions in actions of tort arising in that province in ignorance or disregard of the fact, that in the deeply important point of the onus of proof the law was radically altered in 1866; that, in a word, the Commissioners who framed the law, the legislature that passed it and the courts of appeal that, for forty years, have applied it have all shared in this far-reaching misconception of its meaning and effect. That is an assumption which (with great respect) I decline to act upon without some reason more convincing than any yet advanced.

In the view expressed above an examination of the 'French text becomes unnecessary. Article 2615 C.C., expressly provides that, where the French and English versions differ, that construction is to be adopted which most nearly accords with the existing law upon the topic dealt with; and, assuming the language of the French version to lend some support to that view of the article, which would appear if adopted to effect

[Page 324]

little less than a revolution in the law on this subject as in practice administered in this court, the provision just quoted even on that assumption requires us to resort to the law of Quebec as it stood in 1866.

I now come to the question whether in this view of the law the plaintiff has shewn that the injury of which he complains was due to the fault of the appellants. The relevant facts not in dispute are these; the carbide of calcium is, for the purposes of commerce, produced by the fusion in an electric furnace of carbon and unslaked lime. The appellants use furnaces of two kinds. One in which the constituents of the carbide are reduced to a liquid state, and the liquid carbide itself is run off through one or more openings at stated intervals; another in which the crucible itself containing the fused product is, from time to time as it becomes filled, removed, and another substituted. The first mentioned of these types of furnaces, according to the evidence of the respondent, was known at the appellants' works as la fournaise experimentale; the last mentioned as the "Willson" furnace. The ports through which in the first mentioned type the liquid carbide is run, are, except when open for that purpose, sealed with mortar. When the liquid is to be run off the furnace is tapped, that is to say, the plug of mortar is pierced, and the orifice cleaned out. When the run has been completed the opening is sealed again. It was the respondent's duty, with the assistance of another workman, to tap one of these furnaces and to clear the opening; and it was while engaged in this duty that he met with the injury giving rise to this litigation. The practice was that the liquid having been run off and the orifice cleaned, the mortar for plugging it again was placed on the bottom of the opening by the respondent's helpers. Then the respondent

[Page 325]

proceeded to press it home, employing for that purpose a long poker having a circular iron disc: at its end. On the day of the accident he had just pressed the mortar to the bottom of the orifice, when a loud explosion occurred, accompanied by a discharge through the opening of flaming liquid carbide which burned the respondent severely, destroying his sight.

Thus far, I say the facts are undisputed; and the disputes, so far as concerns the primary facts, are confined to two points. First, it is said that no proper instructions were given to Doucet. Upon this point, there is no finding by the learned trial judge; but there is a finding in favour of the appellants by the Court of Review, and, having regard to the admissions of the respondent, and the positive evidence given by the assistant superintendent of the appellants, it would, I think, be quite impossible to justify the reversal of that finding. Assuming, moreover, some lack of instruction, there is not upon the evidence, I think, the slightest ground for suggesting that the explosion can be attributed to any want of knowledge on the part of the respondent.

The other point in controversy is this: The assistant superintendent says that the respondent was "the boss of the job," that he was responsible for feeding, sealing and tapping the furnace, during his shift. The respondent denies that he was ever appointed foreman; but he constantly speaks of his associate as "mon homme" and his own account leaves little room for dispute that he assumed charge of the furnace and directed the operations mentioned. Here, again, there is no finding by the trial judge, and it seems clear no adequate reason for disturbing the finding of the Court of Review in the appellants' favour.

[Page 326]

The question to be determined is whether these facts afford a sufficient foundation for the inference that the injury suffered by the respondent was due to the fault of the appellants.

The duty of the appellants under the law of Quebec in respect of the care to be exercised for the protection of their workmen (speaking now apart from any question arising out of the construction of article 1054 C.C.), has in this court been more than once explained. The employer is not to insure the safety of his workmen. He is bound to take all reasonable steps to the end that they shall not, because of any defect or insufficiency in his plant or appliances or because of anything unnecessarily dangerous in his system or methods, be exposed to any avoidable risk or harm; and further, where the employment is hazardous (so that in spite of all practicable precautions accidents likely to cause harm are to be expected) he is bound to take the necessary steps to minimize, so far as may reasonably be possible, the evil effects of such accidents when they do occur. It is only a corollary of this to say that in the last mentioned case—where the employment is dangerous—all protective measures are required (reasonably consistent with the prosecution of the duties of the employment) which the knowledge and the practical experience of a prudent and competent employer would suggest.

Such are the measures which in such circumstances are demanded by the requirement of reasonable care on the part of the employer. That is the effect of the passage from Dal. Jur. Gen., 1870, 3, 63, adopted by the Chief Justice in Royal Paper Mills v. Cameron[41], at pages 368 and 369, which I think summarizes the

[Page 327]

law on this point as it has been enunciated and applied in the decisions of this court.

I understand the judges below to put the responsibility of the employer in respect of the safety of his employees rather higher than this. Archambault J., for example, quotes with approval a dictum to the effect, that the employer is bound to know when a machine or process is dangerous. I do not think that can be supported.

The employer is bound to have the knowledge of a competent person; but he is not to be held to know that which competent persons do not know. To hold otherwise would result in making him liable in a very large number of cases which now fall under the head of "inevitable accident."

The question then is, first, whether the discharge of carbide from which the respondent suffered was due to any failure of the duty; and, secondly: Does the evidence shew that the appellants ought at any rate to have anticipated as a possible accident the discharge of carbide from the furnace in such a way as to endanger the safety of their employees? And—if so— was the injury to the respondent due to the omission of any precaution which prudence and competent skill and experience would have dictated as likely to reduce the chances of such an accident, or to minimize the injurious effects of it, should it occur?

The learned trial judge has in general terms found that the mishap was due to the negligence of the appellants. The specific grounds upon which this finding appears to be based are, that, first, the continuous heating of the furnace, and, secondly, the use of mortar for filling the openings mentioned, constituted two unnecessary sources of danger.

[Page 328]

With great respect, the, first of these grounds be- longs, I think, to the realm of pure conjecture. There is literally no evidence to support or even suggest it. The second, also must be rejected for want of evidence to shew that the appellants should have known (or at least foreseen the possibility) that the use of mortar would be a source of danger.

The learned judge based his view apparently upon some rather doubtful evidence to the effect that after the accident the appellants discarded lime as an element in their sealing mixture. Such evidence is, I think, quite irrelevant to the question to be tried, which is whether, before the accident, in the then existing state of experience, the failure of such knowledge or foresight is to be imputed to them for a breach of duty. Conduct pursued in the light of experience derived from the accident itself can hardly be taken as a sufficient basis for a charge of want of care in not taking the same course before the accident occurred. The views which I think have generally prevailed upon this point are summarized by Mr. Justice Gray in delivering the opinion of the Supreme Court of the United States in Columbia and P.S. Railroad Co. v. Hawthorne[42], at pages 207-208, in the following passage:

The only States, so far as we are informed, in which subsequent changes are held to be evidence of prior negligence are Pennsylvania and Kansas, the decisions in which are supported by no satisfactory reason. McKee v. Bidwell[43], and cases cited; St. Louis & San Francisco Railway. Co. v. Weaver[44].

The true rule and the reasons for it were expressed in Morse y. Minneapolis & St. Louis RIy Co.[45], above cited, in which Mr. Justice Mitchell, delivering the unanimous opinion of the Supreme Court of Minnesota, after referring to earlier opinions of the same court

[Page 329]

the other way, said: "But on mature reflection, we have concluded that evidence of this kind ought not to be admitted under any circumstances, and that the rule heretofore adopted by this court is on principle wrong; not for the reason given by some courts, that the acts of the employees in making such repairs are not admissible against their principals, but upon the broader ground that such acts afford no legitimate basis for construing such an act as an admission of previous neglect of duty. A person may have exercised all the care which the law required, and yet, in the light of his new experience, after an unexpected accident has occurred, and as a measure of extreme caution, he may adopt additional safeguards. The more careful a person is, the more regard he has for the lives of others, the more likely he would be to do so, and it would seem unjust that he could not do so without being liable to have such. acts construed as an admission of prior negligence. We think such a rule puts an unfair interpretation upon human conduct, and virtually holds out an inducement for continued negligence."

The same rule appears to be well settled in England. In a case in which it was affirmed by the Court of Exchequer, Baron Bramwell said: "People do not furnish evidence against themselves simply by adopting a new plan in order to prevent the recurrence of the accident. I think that a proposition to the contrary would be barbarous. It would be, as I have often had occasion to tell juries, to hold that, because the world gets wiser as it gets older, therefore it was foolish before." Hart v. Lancashire and Yorkshire Railway Co.[46].

The considerations developed in this passage, it will be observed are considerations of general application, the force of which would not appear to be affected by anything peculiar to the system of law prevailing in Quebec.

The respondent, it is true, attributes the accident to an excess of water in the mortar; but it is quite clear at the same time that this is simply a guess; Archambault J. expresses and proceeds upon the view that the contact of such an excess of water with the flaming carbide would produce acetylene gas, which is explosive. This view, again, is not based upon evidence; and with great respect, I do not think it

[Page 330]

competent for any court to arrive at an adjudication imputing fault and consequent liability upon the basis of scientific opinions having no foundation in the evidence upon the record, touching questions respecting the nature and results of chemical processes which, to say the least, are not immediately obvious.

Assuming, for example (a fact well enough known though not touched upon in the record), that the elements of water and of calcium carbide (when the two substances are brought together at ordinary temperatures) do separate and recombine to form, among other things, acetylene gas, can one affirm that the introduction of water into an electric furnace in full operation could bring about the decomposition of a mass of molten carbide there? Would the water not be converted into steam before it could by any possibility come into contact with the carbide? Will steam and 'molten carbide decompose one another? Such questions must obviously be considered in any rational investigation of the point mentioned, and the record presents literally no materials enabling one to answer them intelligently.

But there is another view which must be examined, and that is that there is sufficient ground in the circumstances of this case for affirming that the explosion in question was due to some want of care on the part of the appellants, although the specific point in which they failed in their duty is not made to appear. There are many cases in which the fact alone of an accident occurring is held to be a sufficient foundation for such an inference. Speaking broadly, in England and in the United States, this inference is held to be permissible when the injury has been caused by something wholly within the control of the defendant or of

[Page 331]

persons for whose actions he was responsible, and the occurrence to which the injury was due was not of such a character as would ordinarily take place in the absence of negligence. Given these conditions, the inference in the absence of explanation is a plain one, but the question whether the inference is, or is not permissible, is in truth, not a question of law at all. Apart from specific rule it is merely a question of right thinking. Although, therefore, the conditions under which the maxim "res ipsa loquiter" is normally applied in jurisdictions where the common law prevails, are not present here (in that vis-a-vis, the respondent, the furnace cannot be said to have been wholly within the control of the appellants); still, if the facts in evidence in the present case could be said to justify the conclusion that assuming the absence of negligence on the part of the respondent and his helper (and I think we cannot, on the evidence attribute the accident to any such negligence), such an explosion would not occur except as the result of some negligence in respect either of the construction of the furnace, or of the methods of operation (although we might be entirely in the dark as to the specific thing wrongfully done or omitted) a primà facie case of negligence would clearly be made out. To say that such conditions were present is only another way of saying there was evidence of negligence.

But these conditions are not present. The explosion stands, as a single, unexplained, isolated fact. The evidence relating both to the furnace and to the nature of the process is strikingly meagre. The construction of the furnace itself is not really described; and apart from the facts that the carbide is produced by the fusion of lime and coke by means of an electric furnace, and that the liquid carbide is run off in the manner

[Page 332]

already described, we know nothing whatever of the process. Is it then safe on such a state of the evidence to affirm as the basis of a judicial decision that this single occurrence was anything other than a pure accident? That it would not have taken place without negligence? Or that the possibility of it ought to have been foreseen? After ransacking the evidence for materials to support a reasoned conclusion on these points I find myself still in the region of surmise.

Some minor contentions remain. It is said that the description of the furnace as "la fournaise experimentale" that the use of it was a mere experiment; and that the evidence of the defendants' superintendent contains what was in effect an admission that there was a known risk of such an escape of gas as might cause injury. The experiment indicated by the name "la fournaise experimentale" may have been, and it most likely was a trial of the comparative cost of operating the two types of furnaces. The evidence at all events does not point to anything else; and there is literally nothing to indicate that at the time of the accident the furnace was in respect of the safety of operating it in an experimental stage. At that time the "Will-son" furnace appears to have been almost, if not altogether, displaced. The descriptive phrase applied to the other type had evidently survived the experiment. There is nothing whatever to shew that one type of furnace had not been as fully tested as the other; and to assume that either of them is in more general use than the other, or that furnaces in which the molten product is drawn off by tapping have not been tried and generally adopted is again to pass entirely beyond the region of legitimate deduction from the facts before us.

[Page 333]

Then as to the admission of the superintendent. This is the passage in which it is said to be found:

A.—Well, the proper way was to fill the furnace, and to make the carbide, certain details about these furnaces are not necessary to be given to the men, they are simple, but they were given just the same; I told him to empty or blow the furnace when it was necessary to do so at certain intervals.

Q.—And did you ever notice that he had not done it according to the instructions received? A.—Mr. Doucet was a good workman, but it is possible that on certain occasions the work has not been done well, it is possible it was so about the feeding of the furnace, but I cannot tell.

Q.—What did you say to Doucet, besides? A.—I told him how to do the work right as it had to be done.

Q.—What had to be done for the feeding of the furnace? A.— Well, if the gas would come out too fast, they must look out and put something in the furnace.

Q.—Is the operation a dangerous one? A.—No, I did not say it was dangerous.

Q.—Why did you give these instructions to Doucet then? A.— Because I was in charge of the furnace.

Q.—And what did you say besides about the furnace? A.—I told him how to feed the furnace and how to do all the operations he had to perform.

This passage has been, I think, misunderstood. The witness is not on the point of danger at all—he is discussing the operation of feeding the furnace; and he says that when the gas escapes freely and in large volume, the workmen whose duty it is to do so, must put in a fresh supply of the materials out of which the carbide is produced. The reason for this is not explained; but one naturally supposes that a too rapid escape of gas (which probably finds its outlet through these materials and the opening by which the furnace is fed), indicates that the supply of material is low and that in consequence some waste of power is taking place. That would seem to be the simple explanation; but the passage as it stands, without elucidation, cannot certainly with justice be regarded as

[Page 334]

containing an admission or even a suggestion that the process should have been known to be (or was in fact) a dangerous one.

The court below has held that a proper regard for the safety of its employees would have suggested to the appellant the use of masks for the protection of their faces. But, why, if there was no reason to anticipate an accident of such a character as frequent or make useful that kind of protection? The fact that after the accident masks were supplied is relied upon; and I have already given my reasons for thinking this circumstance to be immaterial. Moreover, what ground is there for suggesting that any mask having a mesh large enough to leave the employees free to perform their necessary duty would have afforded them any protection against the spurt of liquid from which the respondent unfortunately suffered. This, again, is a question we may guess about, and again by no means intelligently answer from the materials before us.

For these reasons, I think, the Court of Review right in holding that the respondent has failed to prove his case—has failed, that is to say, to establish it by direct evidence or by proof of facts which afford sufficient ground for a reasoned inference that the injury from which he unfortunately suffers was due to the appellants' fault.

Anglin J.—The plaintiff lost his sight by an explosion, which occurred while he was performing duties assigned to him by his employers, the defendants, in attending an electric carbide furnace. This furnace had been recently installed and was operated upon a system materially different from that which had obtained

[Page 335]

in the defendants' establishment in connection with furnaces of another class. It was attended by two sets of men—a day-shift and a night-shift. It was the duty of these men to fill the furnace with coke and lime; to draw off the liquid carbide, when ready, through openings at the base of the furnace; and then to cleanse the orifices and to close them with plugs of mortar preparatory to recharging. This work was done once every hour.

The plaintiff and his assistant, who constituted the night-shift, came on duty about 7 o'clock in the evening, and the plaintiff when injured had been at work about twenty minutes. The men of the day-shift had charged the furnace and left a quantity of mortar ready for use. After opening the orifices and allowing the liquid carbide to run off, the plaintiff cleansed them, and the explosion occurred while he was closing one of the orifices with a plug of mortar—which one, or whether the others were all open or all closed, or some open and some closed, does not appear.

Although not directly established by the evidence, there can be little doubt that contact of carbide in the furnace or in one of its orifices with water, or water vapour from the mortar used to close them, caused the explosion.

The Court of Review did not dissent from, and the Court of King's Bench expressly affirmed, the finding of the learned trial judge that the explosion cannot be ascribed to any fault or neglect of the plaintiff. This conclusion appears to be well supported by the evidence. The plaintiff had discretion neither as to the work which he was required to do nor as to the manner in which he should carry it out, and he was, when injured, discharging his duties in the manner

[Page 336]

in which he had been instructed to perform them. The carbide furnace had been left charged by the men of the day-shift. The plaintiff and his assistant, shortly after coming on duty, drew off the liquid carbide and, having cleansed the orifices as instructed, they quite properly proceeded to plug them with mortar prepared for that purpose by the men of the day-shift. These circumstances preclude any imputation of fault or neglect against the plaintiff.

Moreover, I would note en passant, that, to me, they seem to establish that, as found by the learned trial judge, and the majority of the judges in the Court of King's Bench, the things which caused the explosion, though the plaintiff was actually engaged with them, were not under his control or care, but were under the control and care of the defendants. They all belonged to them and were in use for their immediate purposes and profit. They had the direction and control of the manufacturing operations. The plaintiff was an unskilled workman—a servant acting in conformity to orders. I entertain no doubt that, upon the true construction of art. 1054 C.C., these things were in the control of the defendants.

I think it quite possible that evidence could have been adduced that it was highly and unnecessarily dangerous to plug the orifices of a carbide furnace with mortar made with water, and that this in itself constituted a defective system for which the defendants should be held responsible; that their chemist should have anticipated the occurrence of an explosion and that they could easily have taken precautions which would have saved the plaintiff from the severe injuries which he sustained. But the record does not contain such evidence, and without it I would hesitate

[Page 337]

to rest a judgment upon these grounds. If the plaintiff must succeed upon proven, as distinguished from presumed, fault on the part of the defendants, he is charged with the burden of furnishing the proof.

In the absence of fault or neglect on the part of the plaintiff, the very fact of the explosion establishes that there must have been some defect in the system, the furnace, the means employed for cleaning out the orifices, the materials used to plug them, or the composition of these materials, because, with conditions entirely proper, such an explosion should not have occurred. What the precise defect was the evidence does not disclose. Whether it was patent or latent, discoverable or non-discoverable, and if discoverable by what degree of care or by what tests, are questions to which the evidence does not afford answers. But that the explosion was due to some defect—to this extent res ipsa loquitur.

Yet, unless patent, or discoverable by reasonable care and diligence, the existence of the defect would not per se constitute negligence of the defendants.

The master is not an insurer of the safety of his workmen, or servants, and I am unable to understand the principle upon which some of the Quebec decisions proceed, in which, apparently without invoking the provisions of art. 1054 C.C., the employer has been held liable for injuries due to latent defects. Mignault, Droit Civil, vol. 5, p. 372. In several of the cases cited by this learned author the facts in evidence appear to have justified a finding of negligence on the part of the defendants. But, if by "latent defects'' we are to understand defects not discoverable by the exercise of reasonable care or skill, I am, with respect, unable to agree that for injuries due to such defects,

[Page 338]

apart from the provisions of art. 1054 C.C., the employer should be held responsible. Mr. Walton in an interesting article[47], points out that, in the French Cour de Cassation, the liability of the master for injuries to his servants, due to occult faults, was formerly restricted to cases in which the plaintiff could establish a definite vice de construction and was then based upon art. 1384 C.N.

While it seems not improbable that the plaintiff might have procured evidence to shew that the explosion was due to some defect in the defendants' system, or in their plant or its accessories, the existence of which, because discoverable by reasonable care and susceptible of remedy, should be deemed negligence on their part, he has failed to adduce this evidence and the mere occurrence of the explosion does not, I think, warrant the inference that it was caused by such negligence. Without evidence that the defendants should have anticipated danger to their workmen, I am not prepared to hold that they were guilty of fault or negligence because they omitted to warn the plaintiff of the danger to which he was exposed or to furnish him with a mask or spectacles or other means to guard against its consequences.

Upon this view of the evidence I am unable to perceive the applicability of the decision of the Judicial Committee in McArthur v. Dominion Cartridge Co.[48], cited by Carroll J. In that case there was such evidence of defects in machinery and process likely to cause a disastrous explosion, that Lord Macnaghten felt impelled to say:

The wonder really is, not that the explosion happened as and when it did, but that things went on so long without an explosion.

[Page 339]

That these proven defects actually caused the explosion the jury were, in the peculiar circumstances, allowed to infer without direct proof.

I am, with great respect, of the opinion that the judgment in appeal cannot be supported on the ground that, apart from any presumption of fault, the evidence sufficiently establishes that the explosion which injured the plaintiff was caused by actual fault or negligence on the part of the defendants.

It is therefore necessary to deal with the contention of the plaintiff that, under art. 1054 C.C., upon the facts in evidence, there is a presumption of fault on the part of the defendants.

Arts. 1053 and 1054 of the Civil Code are as follows:

1053. Every person capable of discerning right from wrong is responsible for the damage caused by his fault to another, whether by positive act, imprudence, neglect or want of skill.

1054. He is responsible not only for the damage caused by his own fault, but also for that caused by the fault of persons under his control and by things which he has under his care;

The father, or, after his decease, the mother, is responsible for the damage caused by their minor children;

Tutors are responsible in like manner for their pupils;

Curators or others having the legal custody of insane persons, for the damage done by the latter;

Schoolmasters and artisans, for the damage caused by their pupils or apprentices while under their care;

The responsibility attaches in the above cases only when the person subject to it fails to establish that he was unable to prevent the act which has caused the damage.

Masters and employers are responsible for the damage caused by their servants and workmen in the performance of the work for which they are employed.

There is no decision of this court upon the construction and effect of article 1054 C.C. In Paquet v. Dufour[49], Mr. Justice Girouard, at page 334, de-

[Page 340]

clined to subscribe to the view that where an explosion had been caused by dynamite under the care of the defendant he should be held responsible for the resulting damages unless he should prove that it was impossible for him to have prevented the occurrence. No other member of the court took this position. Although in several cases, not unlike that now under consideration, which have come to this court on appeal from the Province of Quebec, there have been broad statements that in order to succeed the plaintiff must affirmatively prove actionable negligence on the part of the defendant, the claim put forward has invariably been founded upon an allegation of negligence which the plaintiff has failed to establish by the evidence which he adduced, and the court acting upon the view that the judgment should be "secundum allegata et probata," has refused him relief. In none of these cases, so far as I have been able to discover, has the construction or effect of article 1054 C.C., been discussed, and I have found no case except Paquet v. Dufour[50], in which there has been any expression of opinion by a judge of this court upon the construction of this provision of the Code.

In several cases in the Quebec courts there are broad statements to the effect that proof of fault on the part of the defendant is essential in order to establish his liability» Mercier v. Morin[51], is an instance. In most of these cases, however, no express reference is made to article 1054 CC. There are other decisions of the Quebec courts, such as Dupont v. Quebec Steamship Co.[52], cited by Mr. Justice Archambault, in which it has been held that the effect

[Page 341]

of article 1054 CO., is that, upon proof by the plaintiff that his injuries were caused by things under the care of the defendant, a presumption arises that such injuries are ascribable to the fault of the defendant, and the burden of proof is shifted, with the result that the defendant will be held responsible in damages, unless he shews that he could not (presumably by the exercise of reasonable care and skill) have prevented the occurrence. This was the opinion of the learned trial judge in the present case and of the judges who composed the majority of the Court of King's Bench, and it was not dissented from in the Court of Review.

It is manifest that it would be futile to attempt to reconcile this view of the law with the considérant of the judgment of the Court of King's Bench in Canadian Pacific Railway Co. v. Dionne[53], quoted by my brother Duff. But we may not reverse the judgment now in appeal merely because it is not in accord with an earlier opinion of the Court of King's Bench. We must be satisfied that the judgment in the present case is erroneous.

The earlier French authorities held the master responsible where a workman had sustained injuries of which the cause was unknown. But at a later period, possibly because they feared that the defendant's responsibility for damages caused by inanimate things in his care would otherwise under the Code Napoléon be absolute and unqualified, the French courts and authors of repute took the view that it could not have been intended that he should be subjected to such a serious burden without proof of fault or negligence. The contrary view, however, was steadfastly maintained

[Page 342]

by such able commentators as MM. Marcadé and Josserand. More recently the French Cour de Cassation, which apparently evaded the question as long as possible (Dalloz, 1907, 1, 177), seems to have felt itself constrained by the explicit language of art. 1384 C.N., to accede to the view supported by MM. Marcadé and Josserand, and, in a very late decision, reported in Dalloz[54], that court appears to have committed itself to the proposition that a plaintiff who proves that he has been injured by an inanimate thing, e.g., machinery, in the care of the defendant, is entitled to recover damages from the latter without adducing evidence of fault on his part; the presumption thus established, says the court, may be met by proof that the injuries of the victim are due entirely to his own fault.

From a still more recent case decided in the fourth Chamber of the Court of Appeal of Paris, I take the following head-note:

Si l'article 1384, alinéa 1er, du code civil, en disposant que Ton répond du dommage causé par le fait des choses que Ton a sous sa garde, crée une responsabilité basée sur une présomption de faute, cette présomption peut être détruite par la preuve d'un cas de force majeure ou d'un cas fortuit ou de l'inexistence de toute faute pouvant être imputée à celui qui la garde de la chose, soit à raison de l'état matériel de cette chose, soit à raison des conditions de son emploi. Recueil Phily, 1909, p. 926, No. 5039.

While the opinions of the French courts (Cassation and Appeal) are not binding upon this court, they are entitled, particularly when they deal with provisions of the French Code similar to those of the Quebec Civil Code, and of French and not of English origin, to the very greatest respect.

It was strongly pressed upon us by counsel for

[Page 343]

the respondent that the sixth paragraph of article 1054 CO., is applicable not only to the second, third, fourth and fifth paragraphs, but also to the first paragraph of the article. The judgment of the Court of King's Bench proceeds upon this view. For the appellant it is contended that the application of the sixth paragraph does not extend to the first paragraph, but is restricted to paragraphs 2 to 5 inclusive.

For convenient reference I shall set out here the sixth clause of article 1054 of the Civil Code and also the corresponding clause of the Code Napoléon.

The former reads as follows:

The responsibility attaches in the above cases only when the person subject to it fails to establish that he was unable to prevent the act which has caused the damage.

And, in the French version:

La responsabilité ci-dessus a lieu seulement lorsque la personne qui y est assujettie ne peut prouver qu'elle n'a pu empêcher le fait qui a causé le dommage.

The provision of the Napoleonic Code, article 1384, is as follows:

La responsabilité ci-dessus a lieu, à moins que les père et mère, instituteurs et artisans, ne prouvent qu'ils n'ont pu empêcher le fait qui donne lieu à cette responsabilité.

The restricted application of this clause in the Napoleonic Code admits of no doubt. But the language of the provisions of the Quebec Civil Code does not at all so distinctly define its applicability. That the responsibility under the first paragraph of article 1054 C.C., in the case of damage caused by things which the defendant has under his care, if fault or negligence is to be presumed against him, may otherwise be exceedingly onerous because absolute and unqualified, affords a very strong argument in support of the view

[Page 344]

that the exculpatory clause should be held to apply to it, and the departure from phraseology of the Code Napoléon in substituting for its precise and unmistakable language the somewhat indefinite phrase "la responsabilité ci-dessus" which is, translated, "the responsibility in the above cases," is quite consistent with and in fact rather indicative of a design to extend the application of this clause to all the cases covered by the article except those of masters and employers. Otherwise, why should not the codifiers of the Quebec Civil Code have adopted the language of the Code Napoléon merely inserting "curator and custodian of the insane" in addition to "father and mother, schoolmasters and artisans?"

Pressed by these arguments, I was for spine time of the opinion that the exculpatory clause of art. 1054 C.C., should be held applicable to all cases within the first paragraph of the article except those in which the damage had been caused by the fault of the defendant himself. But upon further consideration I am convinced that this clause does not apply at all to the first paragraph. That it cannot apply to cases of damage caused by the fault of the defendant himself is obvious. That it was not meant to apply to the responsibility of masters and employers for damage caused by their servants and workmen is also clear. Yet if the clause should be held to apply to the cases covered by the first paragraphs of damage "caused by the fault of persons under his (the defendant's) control," would it not apply to the liability of masters and employers for damage caused by their servants and workmen, inasmuch as servants and workmen are persons under the control of their masters and employers? Then when it is sought to apply this clause to cases of damage caused by things under the defendant’s care, the

[Page 345]

language of the English version, viz., that he was "unable to prevent the act which has caused the damage," presents a very serious difficulty because it is certainly inaccurate to speak of the "act" of an inanimate thing. It is true that upon the word "fait" of the French version the difficulty may not be so great. But both versions of the Code are of equal authority, and in this instance the original was the English version rather than the French. (McCord's Civil Code, Preface, p. ix.) While I cannot but think the departure at this point from the language of the Napoleonic Code unfortunate, yet notwithstanding the uncertainty to which the words "la responsabilité ci-dessus" are calculated to give rise, I think it reasonably clear that the cases mentioned in paragraphs 2, 3, 4 and 5 of article 1054 C.C., are given as specific instances of "damage caused by persons under his (the defendant's) control," and that to these cases—"the above cases"—the sixth paragraph applies, whereas to the other specified cases of damage caused by persons under the defendant's control, viz., those mentioned in the seventh paragraph, which deals with the responsibility of masters and employers, for damage caused by their servants and workmen, as well as to any other cases within the description "damage caused by the fault of persons under his (the defendant's) control" not particularized in the article, the sixth paragraph has no application. It follows that this paragraph affords no assistance in construing the first paragraph of the article, which must now be carefully examined.

The codifiers of the Civil Code had before them the Code Napoléon. In article 1054 C.C., they followed the form of article 1384 C.N. The differences in language cannot be regarded as merely accidental. They

[Page 346]

were no doubt carefully considered and their significance may not be ignored. Thus we find in the first paragraph the word "faute" in the Quebec Civil Code in lieu of the word "fait" of the Code Napoléon. We find that in the case of "things" the responsibility under the Code Napoléon is for damage "causé par le fait * * * des choses que l'on a sous sa garde" while under the Civil Code it is for damage "causé * * * par les choses qu'elle a sous sa garde" Care was taken when substituting the word "faute" for "fait" that it should not extend to the case of damage caused by things. These differences between the terms of the corresponding paragraphs of the two Codes certainly indicate that the Quebec codifiers intended that evidence of fault of the defendant or of defects in the things themselves should not be a necessary element in the plaintiff's case where damage has been caused to him by things under the care of the defendant.

But, according to all authorities, the cardinal rule for interpreting the Code is, that where the language is clear and free from ambiguity, effect must be given to it and it may not be explained away or controlled by referring to other sources. Here the terms of the first paragraph of article 1054 C.C., are simple and unequivocal. He (every person capable, etc.) is responsible for damage caused "by things which he has under his care." These words immediately follow an article in which proof of fault as the basis of liability is the dominant idea. That idea is continued in this paragraph as to the acts or omissions of the defendant himself or of persons under his control, and its application to these cases is perhaps emphasized by the use of the word "faute" where the Code Napoléon employs the word "fait" But in the case of damage caused by things under the care of the defendant the idea of

[Page 347]

proof of fault is entirely, and I see no reason for thinking that it was not deliberately, eliminated.

No part of article 1054 C.C., is found in brackets. This indicates that the Commissioners did not intend in this article to alter or to amend what they understood to be the existing law of Lower Canada. Under that law, an injured plaintiff was required to adduce positive evidence of fault on the part of the defendant, if he would have the court hold the latter responsible. Because they regard article 1054 C.C., as merely delaratory of the former law, writers and judges have expressed the view that when the plaintiff proves that he has been injured by things under the control of the defendant he must, in addition, adduce positive evidence of fault or negligence on the part of the latter. Mignault, Droit Civil, vol. V., pages 683-4. But the language of the article is explicit, and, in view of the evidence of deliberation in its use afforded by the comparison with the Code Napoléon, I perceive no sufficient reason for reading into it a qualification which has apparently been designedly omitted. In Trust and Loan Company of Canada v. Gauthier[55], art. 1301 of the Civil Code is dealt with. This article was reported by the codifiers as an embodiment of the existing statute law of Lower Canada as judicially interpreted. DeLorimier Bibliothèque du C.C., vol. X., page 303. Their Lordships say, at page 101:

Article 1301 clearly goes further than the law which prevailed in Lower Canada before the Code was framed; but their Lordships cannot accede to the argument that the language used and deliberately adopted in the Code must be narrowed and held to have no greater effect than the previous law for which it has been substituted.

I would be better satisfied if I could find that article 1054 C.C., admitted of a construction upon

[Page 348]

which the exculpatory provision of its sixth paragraph might be extended to the case of damage caused by things in the care of the defendant. I fully realize that without this qualification the responsibility of persons in the position of these defendants may be very onerous, because I find it difficult, owing to the presence of an exculpatory clause restricted in its application to cases within the paragraphs 2, 3, 4 and 5, to accede to the view expressed by the Paris Court of Appeal that the first paragraph should be read as subject to a very similar qualification. Expressio unius, etc Yet it seems to me that to hold that the defendant is responsible for damage caused by things under his care only when it is proved by the plaintiff that he has been negligent or at fault is to give no effect to the concluding words of the first paragraph of the article, because fault imputable to the defendant must be either that of himself or of persons under his control, and damage caused by such fault is expressly declared actionable by the earlier clauses of the paragraph. Notwithstanding that judges for whose opinions I entertain the very greatest respect have taken a different view of the proper construction of the first paragraph of article 1054 C.C., so far as it relates to cases of damage caused by things under the care of the defendant,. I am of opinion that the terms of this paragraph are so clear and unambiguous that it is impossible to refuse to give effect to them merely because the responsibility to which they subject defendants may be unusually onerous.

But, inasmuch as the defendants have offered no evidence to shew that the explosion was attributable to fault of the plaintiff, or to vis major, or that it was a case of pure accident, or that it occurred without

[Page 349]

fault imputable to themselves, it is not in this case necessary to determine whether or not such proof would suffice to relieve a defendant from responsibility. My conclusion is that, at all events, in the absence of such proof, a defendant must, without other evidence of fault or negligence, be held responsible for damages shewn to have been caused by things under his care.

Cases in which injury has been caused by things in the care of the defendant, but in use under statutory authority, form an exception to this general rule.

The legislature is supreme, and if it has enacted that a thing is lawful, such a thing cannot be a fault or an actionable wrong.

Canadian Pacific Railway Co. v. Roy[56]. In such cases the defendant cannot be presumed to be at fault. Actual fault or negligence on his part must be established by evidence.

The appeal should, in my opinion, be dismissed with costs.

Appeal dismissed with costs.

Solicitors for the appellants: McLennan, Howard & Aylmer.

Solicitors for the respondent: Martel & Duplessis.