Tbe following opinion was filed February 5, 1909:
Marshall, J.(dissenting). Tbe most important judicial authority lodged in this court is tbat of passing upon tbe validity of legislative enactments. Tbat great power is given to the court by tbe constitution, as definitely, if not as expressly, as power is given to the legislature to enact laws. In its special field the- court is absolutely independent. It is answerable only to tbe people as their will is seen in tbe fundamental law. Tbe power is not discretionary, now to be exercised and then not to be, according as mere expediency may seem to dictate. It is obligatory in character as to every situation legitimately invoking its activity. It must be jealously guarded and courageously vindicated upon all proper occasions, if our constitutional system of liberty is to endure.
Those who are wont to regard activity of tbe court’s power mentioned as an unwarrantable, or at least a regrettable, interference with legislative authority, evince want of comprehension of our system of government or want of appreciation of the broad scope of those constitutional limitations designed to guard at all points every individual in tbe enjoyment of every right essential to those fundamentals: “life, liberty, and the pursuit of happiness” for which “governments are instituted among men, deriving their just powers from tbe consent of tbe governed.”
Tbe importance of our constitutional restraints and tbe *233high prerogative power of applying them, is as progressive as is tbe need for regulation, to the end that such regulation may not overleap its legitimate boundaries and enter the domain of the destructive. It will be a sorry day for our country when the time comes, if it ever does, — let us hope and believe that it never will, — 'that the invincible weapon— the constitution, — vitalized by an independent and fearless judiciary, shall not efficiently bar excursions into the domain of unbridled interference with individual rights.
If that is more important to any one element in society than to another, it is the weakest, hence the most helpless. So it is of the highest importance to the public, and particularly to the most humble portion thereof, that courts should grapple, willingly and effectively, with every question presented for solution involving validity of legislation on constitutional grounds.
How wisely the fathers must have looked into the future, when — with the evident purpose of their language being regarded as a command from the body of tire people to all in authority, so long as the constitution should endure — they penned the words: “The blessings of a free government can only be maintained by a firm adherence to justice, moderation, temperance, frugality and virtue, and by frequent recurrence to fundamental principles.”
The saying that the court of last resort should willingly apply the test of constitutional limitations, is not to be taken as suggesting judicial desire or haste to declare that not law which has the form of law. In no case should the court enter upon any doubtful ground. It should accord to the coordinate department the highest consideration, not condemning its action so long as any reasonable basis can be discovered for upholding it, but if none can be discovered, not hesitating to put the stamp of judicial disapproval upon it.
The proper attitude indicated deserves, and will doubtless receive, in the end at least, the approbation of the people by *234whom all power was delegated. Tbe judicial disapproval does not'nullify law, as tbe inconsiderate would say. It merely evidences that what is clothed in the mere habiliments of law is not law at all.
Those principles, more often declai’ed in recent years than formerly, are not new. They were laid down by Chief Justice Marshall in Marbury v. Madison, 1 Cranch, 137, and other cases, where that eminent jurist, in the early days-of our constitutional system, gave thereto that vitality essential to its efficiency. He said for the court:
“This original and supreme will organizes the government,, and assigns to different departments their respective powers. . . . Those who framed written constitutions contemplated them as forming the fundamental and paramount law of the nation, and consequently the theory of every such government must be, that an act of the legislature repugnant to the constitution is void. ... It is emphatically the province and duty of the judicial department to say what the law is. . . . If then courts are to regard the constitution; and the constitution is superior to any ordinary act of the legislature ; the constitiition, and not such ordinary act, must govern the case to which they both apply.”
The idea “that courts must close their eyes on the constitution and see only the law” would give to “the legislature a practical and real omnipotence with the same breath which professes to restrict their powers within narrow limits. . . .” That it “would reduce to nothing what we have deemed the greatest improvement on political institutions — a written constitution — would of itself be sufficient, in America where-written constitutions have been viewed with so much reverence, for rejecting the construction.”
The foregoing observations are not indulged in because of any thought that my brethren, in this case, not appreciating the principles stated, have unduly bowed to legislative authority, paying too much heed to its judgment as to that reasonableness which must be regarded as the infallible test of *235legitimacy of a police regulation. They have set the proper-standard in that regard, but failed, in my bumble opinion,, respecting the question of fact as to wbat is reasonable beyond any fair doubt, basing tbeir conclusion on misapprehension of the bearing of decided cases.
If this case could rest on the mere question of abstract right, from high moral ideals as to whether a person who is-injured in the gticm-public work of operating a railroad' should have his loss charged up to the industry as a whole,, and so added to the cost of things to be consumed in the activities of life, no one would be more ready to take the side-of the injured than the writer. Why such losses from an economic standpoint, and a humane view as well, regardless-of any question of negligence, should not be treated as legitimate elements of the cost of production and distribution of' products for human consumption, would be hard to say, satisfactorily, if at all. The question here is not what ought to-be or might be under a different system than that of imposing liability for losses to employees through accidental injuries-upon the nearest employers, but what is legitimate under the-present system. My general observations are to meet the prejudice which I conceive exists in nonjudicial fields against limiting, by judicial disapproval, the extent of that system, upon constitutional grounds, showing that when duty calls for such disapproval there is no field of discretion within which tire judge can operate, though, in the individual case- and all of its class, one might wish for a system whereby the-wounds of all the injured could be healed and the road now so broad and so frequently traveled to the zone of want, could be absolutely abolished.
The first question as to the constitutionality of the act of 1907 arises under the opening part of sec. 1816 and subd. 1, 2, and 9, quoted in the court’s opinion, imposing extraordinary liability upon every railroad company for injuries “to any of its employees,” regardless of the branch of service in *236wbicb they may be engaged, except “employees working in shops or offices.” I make no question but that there may be classification as to risks and radical changes in common-law liabilities as to members of a particular class, but is a classification of employees of railroads proper which entirely ignores hazards peculiar to railroads, by including in the special group all persons engaged in cutting grass upon the railroad right of way, or building fences, or building bridges, or doing work of construction and engineering, or providing supplies, such as ties and many other things that might be mentioned, even employees in any subsidiary business that might be carried on in aid of the railroad business, an army of persons in the aggregate, having no connection whatever with the operating feature of a railroad, which only is characterized by special railroad risks, and excluding shop employees who are exposed to quite as much hazard of personal injury, and office employees who are as much exposed as many of the included subclasses ? Is the mere character of the employer, by itself, a basis for classification, even then excluding a large subclass of employees laboring within, to some extent, the zone of special hazard, and including, as stated, an army of others as far removed from any special risk as employees in any ordinary business ? True, the doctrine of classification has been carried by courts so far that the distinction between special and general legislation is very hard to discover, largely nullifying the safeguards against unequal legislation, but is it true that so arbitrary a classification as we have to deal with here is legitimate, under even the very liberal rules we have adopted ?
If we, instead of tying closely to some definite rule, look to delusive expressions used in precedents here and there outside this state, not following their history back to discover what they are worth by the light of the real groundwork upon which they are based, and being governed by such groundwork rather than such expressions, which, looked at by them*237selves, would, seem to bave been used unconscious of the real premises, there will be no escape from a condition of “classification run mad” which means, in practical effect, no classification at all along definite lines, and the equality clauses of our constitution, state and national, designed to prevent class legislation, in the special sense, will be of no practical effect whatever.
True, courts have laid down as limitations of the power of classification these rules:
1. Classification cannot be arbitrary. It must be based upon substantial distinction which makes one class really different from another.
2. The classification must be germane to the purpose of the law.
3. The classification must not be based upon existing circumstances only, so as to preclude the class opening to let in or let out members.
4. The law must apply equally to members of each class.
5. The characteristics of each class must be so far different from those of others as to reasonably suggest at least the-propriety, having regard for the public good, of substantially different legislation therefor. State ex rel. Risch v. Trustees, 121 Wis. 44, 98 N. W. 954; Bingham v. Milwaukee Co. 127 Wis. 344, 106 N. W. 1071; Bloomer v. Bloomer, 128 Wis. 297, 107 N. W. 974.
In applying these rules I fully appreciate they furnish only a very general test of what is legitimate, but, nevertheless, they are quite as certain a guide as this of any other court has been able to formulate. Obviously, as has been often said, whether any particular situation falls within their boundaries is a question of fact, and is, as well as the question of necessity or propriety of any proposed regulation, primarily for legislative solution, subject to interference by judicial authority only upon its appearing beyond all reasonable doubt that the boundaries of reasonableness have been over*238stepped. But the fact that the rules stated furnish only a general test, and -that no better can be formulated, and the great importance of not extending the doctrine of classification beyond legitimate boundaries so as to practically nullify the safeguards against unequal legislation, — render it highly important that such general rules should be given a pretty definite and certain meaning, not to be regarded as elastic; leaving only questions of fact to be determined as regards whether a given situation falls within or without the general scope.
We also fully recognize that while classification must not be arbitrary, that has reference to the principles embodied in the rules, not to the boundary of any particular class. That must, of necessity, be definite and in that sense arbitrary,
Again, we fully appreciate that the rule that classification must be based upon characteristics making the particular group so far different from others as to reasonably suggest need for or propriety of special legislative treatment, has reference to the group as such, regardless of whether each and every subject within the group has such characteristics to the same degree as every other subject, or not. But the dominant feature in some perceptible degree must affect substantially all. As said in Nichols v. Walter, 37 Minn. 264, 33 N. W. 800, and often quoted with approval here:
There must be “some apparent natural reason — some reason suggested by necessity, by such a difference in the situation and circumstances of the subjects placed in different •classes as suggests the necessity or propriety of different legislation with respect to them.”
So the general rules often stated, in substantially the same language, might well be extended by these explanatory rules which, in the cases referred to and many others, have become as well understood as those reduced to set forms of expression.
6. Rules 1 to 5 do not constitute a definite test of legitimacy of classification except as to general characteristics, *239leaving the question as to whether any given situation falls within their boundaries or not, matter of fact.
7. Whether, upon the facts of any particular situation, it falls within the boundary of the rules for classification, is primarily for the legislature, subject to judicial disapproval for unreasonableness beyond any fair doubt.
8. The rule that classification must not be arbitrary refers to the group as such, not to differences in degree in which the individuals of the group are affected by the special circumstances calling for the classification.
There is this further principle to be observed in dealing with this subject which, for the purpose of having a reasonably complete code, so to speak, for testing any legislative enactment, challenged as invading the equality clauses of national or state constitutions, may well be stated.
9. In classification for the purpose of legislative treatment under the police power, it is a judicial function to determine whether the particular subject is within the police power, also whether the act has a real relation to the subject it ostensibly deals with, and whether the manner of treatment is unquestionably unreasonable. State v. Redmon, 134 Wis. 89, 114 N. W. 137; Bonnett v. Vallier, 136 Wis. 194, 116 N. W. 885; Lawton v. Steele, 152 U. S. 133, 14 Sup. Ct. 499.
Having placed before us pretty fully, it is thought, the principles by which the law in' question must be squared as regards whether it transcends constitutional limitations, let us proceed to make the measurements. I choose to test the •enactment in the first instance, at least, by principles, not by mere precedents, which latter method, as before indicated, in ■case of failure to fully analyze the examples, is liable to lead one astray. That liability, I am constrained to believe, is what, in the main, in this instance, has led my brethren to the conclusion embodied in the judgment from which I dissent.
The purpose of the law was to promote public safety both *240as regards patrons of railway companies and the servants of' such companies, and that satisfies the ninth rule as to one feature, to wit: that as to whether it is within the police power, unless the manner of treatment is clearly unreasonable.
The law applies equally to all subjects within the class, and the class is not fixed so as to prevent opening to let in or let out members, satisfying the third rule.
Now how about substantial distinctions between the particular class and others under the first rule, and such differences in situations and circumstances of subjects within such class, satisfying the first and fifth rules? In considering this we necessarily blend the domain of discussion with the-eighth rule, that, while the boundaries of the class, as a whole, must be arbitrary, and the differences in situation between the subjects within the class and those without must be significant, the classification must not be arbitrary as regards, extending it to classes of individual subjects not in any degree affected by the particular situation or circumstances-forming the basis upon which the classification is grounded, though the degree of affection as to individual subjects may widely differ. These three rules, the first, the fifth, and the-eighth, and also the second, form one general field which may-well be treated as such, since the domain of each rule so blends with that of the others as to render segregation for more particular analysis liable to confuse.
The purpose as stated is public safety, the term “public-safety” being used, as indicated, with reference to the particular body of employees and those they indirectly serve; the patrons of the employers. Is it germane to such purpose to extend the classification, not incidentally, but substantially, to put it not too strongly, manifestly, to a very large-degree beyond the field of special hazard in view in the purpose to conserve public safety, giving special benefit to an army of employees not affected at all by such special hazard *241or baying to do at all with such hazard as regards patrons of the service and laying upon employers burdens accordingly ? Face the rule that the difference in situation calling for the special treatment must pervade the class, not with equal degree as to each member thereof, but to some extent as to substantially all, keeping in mind and giving due weight to the impracticability of fencing the class about so accurately as to exclude every one not affected in any degree by the special hazard and include every one so affected. How can the whole be substantially pervaded by the special circumstances if a very large proportion of the employees are not affected at all ? To affirm that it can, is to indulge, in my judgment, in a plain contradiction. So plain does it appear, as an original proposition, that I cannot perceive how one could venture to assert the contrary and ground a judicial decision upon it.
The mistake of my brethren at this point, as I view the matter, is in testing the law by the second rule with reference to the mere classification of the employees, instead of the business they are engaged in. If a mining company conducts a store, a farm, and other industries, subsidiary to the primary business, in which subsidiary employments the hazards of personal injury are no greater than like employment by private individuals or corporations, yet the number of the employees therein constitutes as large a proportion of the whole, perhaps, as the number in the specially hazardous branch, would a classification as regards extraordinary liability of the employer for personal injuries to employees, including all the subsidiary employments, be germane to a law having for its purpose public safety as regards extraordinary hazard of mining? The negative seems so plain as to be beyond possible question.
Does not the illustration exactly fit the case in hand and condemn the classification attempted ? The special situation and circumstances of railroad employment having to specially do with public safety, is confined to the operating department *242in the moving of cars and trains or work around them, or upon the track, or in some way so as to come, to some perceptible degree, within «the zone of special risk, because of being more than ordinarily within the reach of those physical situations to which railroad perils, as ordinarily specialized, are incident.
The difficulty, under the act before us, is not that the large number of employees outside the special zone of hazard are not affected in the same degree by the special circumstances as those within, but is that they are not affected thereby at all, thus rendering the inclusion of them within the group receiving special protection for the public benefit, manifestly arbitrary, and the ostensible purpose to promote the public welfare a subterfuge for the actual purpose of extending a special privilege to a large number of persons without any legitimate common basis therefor.
At this point we may well note that the ostensible purpose of a law does not govern at all in testing it by constitutional limitations. The court may, yea it is its duty, to look behind the mere veil of any law which may appear legitimate, and if, in its substance, it is bad, to characterize it by its substance, not by its pretense. Mugler v. Kansas, 123 U. S. 623, 661, 8 Sup. Ct. 273; Matter of Jacobs, 98 N. Y. 98, 110; State v. Redmon, 134 Wis. 89, 114 N. W. 131.
If anything more was needed to demonstrate; on principle, that the classification is purely arbitrary and so not germane to the purpose to conserve public safety, it is furnished in the exclusion of shop and office employees. Shop employees who, by common knowledge, form a very large subclass, are in some degree within the zone of special hazard. The repair of locomotives, testing them for suitableness for the service, and the performance of other shop duties that might be mentioned, require service to quite a degree within the zone of special hazard.
Upon what ground was this large subclass of employees ex-*243eluded from tlie peculiar "benefits of the enactment, and other large classes, entirely removed from the zone of special hazard, included? I have searched in vain for any possible legitimate ground therefor and so been’forced to the conclusion that the exclusion was purely arbitrary and probably unmindful of the constitutional restraints -involved.
Of course, the foregoing condemnation could be easily avoided so as to save the law if it were subject to construction and could be held as intended to apply only to employees within the field of special hazard, but that has been rejected, necessarily, by the judgment of the court in holding that plaintiff, a mere fence builder, is in the class affected favox*-ably by the law.
My brethren treat quite cursorily the feature of the law last discussed and say “shop and office employees” were excluded because not within “the unusual dangers and hazards of the business.” How can we say that, when the special danger in a large degree by common knowledge does extend to such employees ? Again, how can one say that and in the same breath justify the inclusion of many other subclasses of employees, including fence builders, manifestly not exposed to the special dangers at all ? I confess I do not understand the logic.
Again, my brethren, it seems, endeavor to escape from the dilemma the shop employee feature of the law presents, notwithstanding the doubtful justification I have referred to, by asserting that if the subclass were improperly excluded that would not invalidate the classification; resting on that mere ’ assertion and a reference without comment to State v. Evans, 130 Wis. 381, 110 N. W. 241.
I confess inability to discover anything in that, case warranting the reference. The court there merely laid down the principle that, in case of the establishment of a boundary "Krug of classification with general characteristics of those on the side excluded from the special rights granted or duties im*244posed, differing from those included, warranting the segregation, the fact of some excluded being as worthy, or as much in duty bound to bear the special burden, as those on the other side, does not militate against the legitimacy of such line. Here the case is far different.. The general character of the excluded subclass, not mere exceptional instances, calls for the special legislative treatment, quite as clearly as members of the included class, taken as a whole. The decision of the case, as I read it, condemns rather than supports the court’s conclusion. It clearly is to the effect that the special class need not include all affected to any degree by the peculiar characteristics, but it must do so as far as practicable, and a plain unnecessary inclusion of a subclass not so affected and exclusion of a subclass so affected is fatal. The language of the opinion is:
“It is suggested that this exemption is improper because these employees may be subjected to hazards or perils equally dangerous to those to which other employees are subjected. . . . That would not invalidate the classification.”
It will be noted that the court’s logic is not confined to an excluded class, having now and then a member on an equality with members of the included class, but to an excluded entire subclass of a general group, composed of men on the same plane, generally, as the common mass. It seems that, on more mature consideration, my brethren would not wish to adhere to their logic. Would it be legitimate to legislate specially for all cities of the third class, except one or more specially named, or all cities having a population of 10,000 people except one or several specially excepted? Certainly not, on the most familiar principles of constitutional classification. Otherwise room for that special legislation which is illegitimate would be so broad as to nullify completely the fundamental law as to equality.
I have thus shown, it seems, that the law in question plainly offends against governing principles. Adjudications elsewhere which run counter thereto should not be adopted. *245The principles, not the adjudications inconsistent therewith, should prevail. The correct rule, as I understand it, is to make use of precedents to illustrate principles, not to avoid them. When the principles and the precedents conflict it is the latter, not the former, which should give way. I say this conceding for the purpose thereof that the authorities upon which the court rely are out of harmony with the principles, but such want of harmony will disappear as we proceed, and disappear in favor of my dissent.
There is no precedent in our adjudications out of harmony with my conclusion. If it were otherwise, I would not hesitate to contend with the doctrine of stare decisis in order to uphold the principles. Which would have to give way would be governed very much by circumstances unnecessary to discuss.
The only case referred to on the subject discussed, decided by this court, is Ditberner v. C., M. & St. P. R. Co. 47 Wis. 138, 2 N. W. 69. It is sufficient to say that the question of constitutional classification was not there raised, discussed, or passed upon.
Reference to the decision on that subject is quite misleading, in my opinion. The legislation considered was upheld, solely as a legitimate amendment to corporate charters under the reserve power in the constitution. How that conclusion could have been reached, in view of the fact that there was no intent by the legislature to make any such change, as evidenced beyond controversy by the fact that it referred to foreign as well as domestic corporations, and the fact that no legislative authority exists to deal with the organic law of the former, is not perceived.
The court does not in the instant case attempt to support the legislation as an amendment of corporate charters. Therefore, we need not pursue the subject of the Ditberner Case at length. It is sufficient to show that the decision is quite beside the matter it is now cited to support.
It is by no means certain the court, as at present consti*246tuted, would follow the Ditberner Case on the precise point there decided. It is interesting to note that it has been frequently cited in other jurisdictions with the same misapprehension as to the, real point decided, as appears in the present instance, looking at language only. I think I am justified in saying, in passing, that the court does not at this time-intend to indorse the doctrine that legislation of the sort under consideration can be regarded as an amendment of eor-porate charters, and does not intend to condemn it, but to leave the subject open for consideration at some future time-when the precise point may require consideration.
It is unfortunate, in my view, that the court now says that on the former occasion it rejected the doctrine of Deppe v. C., R. I. & P. R. Co. 36 Iowa, 52, respecting the proper basis-for classification of railway employees for legislation of the-sort in question, since such basis was the subject decided in the latter but was one entirely foreign to the former. The-doctrine of the Iowa case, as regards anything heretofore decided by this court as to constitutional law, stands entirely untouched.
My brethren refer to Mo. Pac. R. Co. v. Mackey, 127 U. S. 205, 8 Sup. Ct. 1161, in support of their decision. A history of the subject there treated will show that the case should be really regarded as authority to the contrary of such decision. It involved the law of Kansas making every domestic railroad company liable to any person injured by negligence of its engineers or other employees. It was quite general in its terms as regards the risks contemplated. It had received construction by the supreme court of Kansas, where it was sustained only by giving it the construction it had theretofore received by the supreme court of Iowa, from which state it was borrowed after such construction. It was restrained, though broad in its literal sense, to the particular persons exposed to-the special hazard of railroad operations, or to the particular risks peculiarly incident to railroading. Mo. Pac. R. Co. *247Haley, 25 Kan. 35. It was beld that the act permitted such construction and required it, as the parent law was so construed prior to its adoption- in Kansas. The federal court sustained the law on the ground of the hazardous nature of the business, adopting the construction given thereto by the court from which the case was transferred.
Referring to the Iowa decision we find, unmistakably, that the initial legislation was only saved from condemnation as unconstitutional by restraining it, by quite extreme rules for judicial construction, to acts of persons within the narrow compass of engagement in the operating department characterized by the special hazard. It was said it could not be saved if it were held to extend to persons engaged in constructing the road. That “if it were so construed as to apply to all persons in the employ of railroad corporations without regard to the nature of their service it would be a clear case of class legislation. . . . Hence would be unconstitutional and clearly so.”
My brethren also rely on Minneapolis & St. L. R. Co. v. Herrick, 127 U. S. 210, 8 Sup. Ct. 1176; Chicago, K. & W. R. Co. v. Pontius, 157 U. S. 209, 15 Sup. Ct. 585; Peirce v. Van Dusen, 78 Fed. 693, 47 U. S. App. 339; and Tullis v. L. E. & W. R. Co. 175 U. S. 348, 20 Sup. Ct. 136. The first followed the case already reviewed, so we read it contrary to the use thereof in the court’s opinion. The same is true of the second. It went, unmistakably, upon the construction of the Kansas statute, given by the supreme court of that state,- following that of the state from which the legislation was borrowed. Plaintiff prevailed because he was engaged in the peculiarly hazardous-occupation of operating a railroad, to which occupation the law was supposed to be restricted. Whether he was within or without the zone of special hazard was the mooted question. The case, as T read it, does not deal with the subject under discussion at all in any other respect. It seems to have been inadvertently cited.
*248The fourth case is said to deal with the validity of an act abolishing the defense of negligence of a fellow-servant as to railway corporations and that it extended to the same class of railroad employees as does the law in question and that it was held proper to treat railway employees as a special class, reaffirming the case I have reviewed involving the Kansas law. I cannot understand the case that way at all. The Kansas cases were referred to as warranting special regulations as to specially hazardous occupations, but by way of argument. The law was of a most general nature, applying to all except municipal corporations, as regards particularly specified hazards of a peculiarly dangerous character and other specified risks, out of the ordinary. The court adopted the construction given by the Indiana court in Pittsburgh, C., C. & St. L. R. Co. v. Montgomery, 152 Ind. 1, 49 N. E. 582, to the effect that it dealt only with the special hazards peculiar to the business, not with employees regardless of whether they were within the zone of special hazard or not, excluding some within such zone and others not and including others wholly outside thereof, as in the law in question.
The Indiana court, except as hereafter stated, grounded its decision on adjudications of the supreme court of Kansas, Iowa, and Minnesota and the approval thereof by the federal supreme court under the rule that the construction of a state statute by the highest court of the state will be followed by the federal court as to cases arising in such state, each and all of which decisions are of the character of those heretofore referred to. Atty Gen. v. Railroad Cos. 35 Wis. 425, and the Ditberner Case were referred to, neither of which dealt with the subject of constitutional classification at all, as we have heretofore seen.
Thus it appears plainly the learned Indiana court did not discover the real basis for the decision in either of the cases cited, though the fact remains that it sustained the law and the federal court followed its decision upon the theory that *249it dealt only with a class including all within the zone of peculiar hazards and not including others. The point being made in the Indiana court that the law included many corporations whose business was not characterized by any hazard other than those incident to the same business conducted commonly by natural persons, the court declined to deal with the question, holding that the law was good in any event as to railroad companies. Whether that was sound 'or not may admit of some doubt.
At this point, following the order of the court’s treatment, we reach a further reference in the opinion to the subject of shop and office employees, and their exclusion is justified upon the authority of numerous cases cited, including Minneapolis & St. L. R. Co. v. Herrick, supra; Chicago, K. & W. R. Co. v. Pontius, supra; and Pittsburgh, C., C. & St. L. R. Co. v. Montgomery, supra. I need not pursue that branch' of the case. Suffice it to say the opinion does not point out their application. How they justify exclusion of a subclass which is largely within the field of special hazard dealt with and inclusion of other subclasses which are not, my reading ■fails to discover. So far as they touch the subject they all go bach by a path, unmistakably marked, to the Iowa decision -sustaining such legislation within the zone of special hazard as the only legitimate basis of classification and the only way of saving it from invalidity. That is so, as we have seen, -as to all cases we have referred to.
The same is true with the court’s additional citation, Georgia R. & B. Co. v. Miller, 90 Ga. 571, 16 S. E. 939. There was no question of exclusion and inclusion disregarding hazards, as in this case.
A further additional citation, Atchison, T. & S. F. R. Co. v. Matthews, 174 U. S. 96, 19 Sup. Ct. 609, deals solely with the validity of a law allowing attorney’s fees in addition to damages in case of recovery under a statute designed to penalize corporations for violating police regulations as regards *250setting fires. The federal court simply followed the decision; of the state court, but in. any event the case has too remote a bearing here, if any at all, for me to appreciate it.
I have now reviewed all of the supports relied upon for the conclusion from which I dissent, on the subject of constitutional classification. The entire group of citations are referable to Deppe v. C., R. I. & P. R. Co. 36 Iowa, 52,—which they expressly, directly or indirectly, follow, — conceded by my brethren to be contrary to their conclusion. They failed, as it seems, to discover that the decision of the supreme court, of Kansas, adopted by the supreme court of the United States, merely upheld the Kansas law with the construction given thereto by the Iowa court.
My brethren disregard without difficulty Lavallee v. St. P., M. & M. R. Co. 40 Minn. 249, 41 N. W. 974; Johnson v. St. P. & D. R. Co. 43 Minn. 222, 45 N. W. 156; Pearson v. C., M. & St. P. R. Co. 47 Minn. 9, 49 N. W. 302; Weisel v. E. R. Co. 79 Minn. 245, 82 N. W. 576; and O’Niel v. G. N. R. Co. 80 Minn. 27, 82 N. W. 1086. But the Minnesota law was quite as sweeping in its provisions as ours, omitting the feature of the latter as to shop and office employees. It was construed the same as the Iowa law and the Kansas law and sustained upon the same ground as they were sustained. It seems my brethren do not give proper dignity to those cases by merely suggesting that the basis for classification was held to be the extraordinary hazard to employees. I think in that there is confusion between purpose of the law and the element of being germane to the purpose. The purpose of all such laws is public safety. The basis of classification has regard to the element of adaptation. The decisions which are supposed to support the conclusion this court has reached, as well as the others, all make the special hazard the basis for the classification. They all deal with a single class of legislation and are in harmony.
True, the Minnesota court said, “The manifest purpose *251\was to give tbe benefit to employees engaged in tbe hazardous business of operating railroads.” Can there be any possible question but that the purpose of the act in question is to give special benefits to the same class and to others ? The fact that the special benefits, in a sense, penalize the employers, and so operate to the benefit of the people generally as regards safety, does not militate against the fact' that the primary benefit in the mind of the legislature was to employees. I see no escape from the conclusion that all these cases support the conclusion I have arrived at on principle alone and are-in perfect harmony, leaving no room to reject some and adopt others.
The foregoing covers all ground traveled over by the court to its conclusion. I might safely, I have shown, rest my contrary conclusion on the precedents the court relies on if precedents alone were to govern. It .was no fault of the learned counsel for appellant that the court did not take notice that the starting point of all the judicial authority referred to by respondent’s counsel and the court is Deppe v. C., R. I. & P. R. Co., supra. These further cases cited by appellant’s counsel from other states, also based on the logic of the Iowa ease, were not referred to in the court’s opinion. Bradford C. Co. v. Heflin, 88 Miss. 314, 42 South. 174; Ballard v. Miss. C. O. Co. 81 Miss. 507, 34 South. 533; Beleal v. N. P. R. Co. 15 N. Dak. 318, 108 N. W. 33. The gist of all is. in the early Iowa case.
The upshot of the foregoing is that if the act in question-could be sustained at all, it would have to rest on a preliminary construction of the statute in harmony with the numerous decisions referred to. I see no way to reach the preliminary construction because of the plain, unmistakable purpose-in the act to include all employees regardless of whether operating strictly within the zone of railroad risks or not, with the exception noted, and that exception necessarily negatives any theory that any other employees are excluded under the *252canon of construction expressio unius est exclusio alierius. As the injured person here was a mere fence builder and so not injured by any peril, properly denominable as a railroad risk, the complaint fails to state a cause of action under the universal rule for construction of such statutes as ours, where so framed as to be open to construction. ‘Therefore, from any viewpoint the pleading can be reasonably measured it fails, utterly, in my judgment, to state a cause of action.
The decision upholding the third subdivision of the law upon the theory that it is a mere declaration of the unwritten law, I cannot subscribe to. True, we should not declare a law unconstitutional if a construction of it can be reasonably given which will avoid that result, but just as true a law not open to construction should never be varied in its plain, ordinary sense by an effort to construe it. It is only where uncertainty of sense begins that the office of judicial construction can properly become active. Further, just as true it is a canon of construction that it is to be presumed in the enactment of a law that the legislature intended some change in existing law, written or unwritten, or to make a new regulation of some sort. Any construction is to be avoided which would cqnvict the lawmaking power of merely spreading.a collection of words on the record in the form of a law, but in fact meaning nothing, if any rational view can be taken of the enactment leading to a different result. Again, in construing a law, after discovering ambiguity justifying it, the court should look, among other things, to' its reason and spirit and read it according to the intent, if that can be discovered, expressed within the broadest reasonable scope of the words used; the fair, ordinary meaning of such language, however, to be taken, unless it appears clearly that some other was intended. All these principles are so familiar that citation of authority in support of them is unnecessary.
In face of the principles stated and the common knowledge that at the time of the enactment there was more or less sentiment against courts exercising the judicial power to take *253cases from juries on tbe question of contributory negligence, upon sucb a question appearing by tbe evidence to be one of law only, tbe purpose of tbe legislature is quite plain. Tbe state of public mind suggested, wbetber reasonable or not, we need not take time to discuss. Considerable unrest under tbe administration of tbe law in that regard existed going to tbe extent of suggesting that courts were prone to usurp tbe functions of tbe jury. It was not appreciated, it is thought, that the question of wbetber evidence shows contributory negligence beyond any reasonable view to tbe contrary, was never a jury question, and that no trial court could bold to tbe contrary without a breach of judicial duty. In that light and in tbe light of the plain letter of tbe enactment in question and tbe rules we have referred to, I could not, if I would, escape tbe conclusion that tbe legislature intended to, and supposed that it could, take from courts a power they bad been accustomed to exercise in tbe ordinary performance of their duties under tbe plain mandate of tbe fundamental law.
Can there be any mistaking tbe meaning of the words, especially in the light of what has been said:
“Tbe court shall submit to tbe jury tbe following questions: First, wbetber tbe company . . . was guilty of negligence directly contributing to tbe injury; second, if that question is answered in tbe affirmative, wbetber tbe person injured was guilty of any negligence which directly contributed to tbe injury; third, if that question is answered in tbe affirmative, wbetber tbe negligence of tbe party so injured was slighter or greater as a contributing, cause to tbe injury than that of tbe company . . . ; and sucb other questions as may be necessary.”
Discretion was left, it will be noted, as to “sucb other questions” but none as to tbe special questions covering tbe vital points of negligence. So looking at tbe legislative effort, all would agree that it is a usurpation, though an innocent one of course.
I would be exceedingly slow in reaching a conclusion that *254the legislature intended to pass an unconstitutional law. In all my experience as a lawyer and judge, covering a period of nearly forty years, I have never known of an instance of suck •an effectuated intent and do not expect to meet with any in tbe future. It was a mistake, in my judgment, that is all. Nevertheless, it evinces that the entire enactment was allowed to take the form of law without being subjected to careful legislative scrutiny in the light of constitutional tests. To some extent, I should say, the rule applies to such a case, that in case of a court’s conclusion of fact based on a wrong rule of law, the ordinary presumption of its correctness does not obtain. So when the legislature makes a law, so called, not taking note that its power is limited, its judgment on the basis entitling it to controlling significance within all reasonable bounds, is not incorporated into the work.
The view the court took of the act rendered it unnecessary, as was supposed, to consider subd. 8, providing that in case of an action in this state to enforce liability for an injury occurring in a sister state, the defendant shall not be permitted to plead or prove the decision of the latter state as a defense. The view I have taken rather calls for such consideration, but it may be done briefly. The provision was copied substantially verbatim from the law of Indiana. Doubtless it was borrowed, overlooking the fact that January 17, 1902, and prior to the adoption here, the supreme court of Indiana held that it was clearly an unconstitutional interference with property rights. The logic of the court’s opinion on the subject is unanswerable. When a person is injured in a sister state through breach of duty by another as regards his personal safety, such person becomes vested, at once, with such rights of action to remedy the wrong as the law of such sister state affords and none other, and, in case of his death, the rights of personal representatives are likewise restricted. The action being transitory, the defense is likewise transitory. The two necessarily, upon familiar principles, go to-*255.getber. A law of a state other than that of the injury, allowing prosecution of an action for the wrong upon a different basis than the right and preventing the use of a vested defense, is, confiscation. That doctrine has support in our own decisions, in the decisions of the federal supreme court, and in the general trend of authority. Second Ward Sav. Bank v. Schranck, 97 Wis. 250, 73 N. W. 31; Peninsular L. & C. Works v. Union O. & P. Co. 100 Wis. 488, 76 N. W. 359; Eau Claire Nat. Bank v. Macauley, 101 Wis. 304, 72 N. W. 176; Pritchard v. Norton, 106 U. S. 124, 1 Sup. Ct. 102; Hovey v. Elliott, 167 U. S. 409, 17 Sup. Ct. 841.
The parts of the law thus invalid, as I think, leave nothing which tire legislature would have enacted independently. Therefore, upon familiar principles the enactment is wholly ■void.
It is interesting to note that in any event the law may be very short-lived. The federal supreme court in the Employers’ Liability Cases, 207 U. S. 463, 28 Sup. Ct. 141, held that a law which includes employees engaged in interstate commerce is as to that feature a regulation of such commerce. The federal law was condemned because it dealt with all employees of a railroad doing interstate business whether engaged at the time of the injury in such business or in intrastate business, and that the two features were inseparable. Hence the whole was held unconstitutional. Hollowing the same line of reasoning, this court has very recently held that an act limiting the hours of labor of railway telegraphers of a railroad company doing both interstate and intrastate business, is necessarily a regulation of interstate commerce, as the two kinds of business done by the same employees are inseparable ; hence that it is unconstitutional, Congress having already occupied the field as to interstate commerce and by a law materially different from the state law.
Since April 22, 1908, the same situation we dealt with re.-garding telegraphers has existed as to injuries to railway em*256ployees. On that day Congress passed an. act, supposed to-be free from tbe infirmity which led to condemnation of the one in the Employers' Liability Cases and thereby as to personal injuries to railway employees received in the course of their service in interstate transportation, occupied the field under the commerce clause of the constitution. The law (see 35 U. S. Stats. at Large, 1907-8, p. 65) conflicts in a-radical degree with our state provision. The two may not be able to stand together under the logic of the decision in the Telegraphers' Case (State v. C., M. & St. P. R. Co. 136 Wis. 407). Whether such legislation will not wholly displace our.state act is in sight, but not for decision at this time.
We now close this rather lengthy, but I trust not too-lengthy, opinion. The vast importance of the question at issue seemed to justify the extensive treatment I have devoted thereto. With the amount of labor we have, work of this independent sort would not be entered upon other than-under the stimulus of a supposed command of duty. The-learned counsel for appellant and counsel for respondent as-well, presented the case with distinguished ability and helpfulness. In my view, as indicated, the position of appellant is grounded on principle and on a long line of well-reasoned judicial authority with which there is, substantially, no-conflict. So believing, it seemed a duty to the court, to the profession, and to the administration of the law generally, to-express fully my views. I will say here, as on another occasion for the court, we do not doubt but that the very best of" intentions were the mainspring of the enactment in question, but good intentions can never save a legislative effort, if the paramount law condemns it. The constitution was made to-guard the people against the dangers of good intentions as well as bad intentions and mistakes. The former may excuse-a void enactment, but never justify it.
It is never a pleasant duty to perform, to condemn a law as void. The dignity of the legislative office is high. Co*257ordinate branches of the government owe great deference thereto while maintaining the dignity of their own field of activity. It is upon co-operation of the three grand divisions of our governmental system upon that high plane, each doing its duty firmly and submitting cheerfully to the authority of the other within the constitutional scope thereof, — that we must rely for that strong, efficient government which the fathers endeavored to establish by a constitution, which deserves our most distinguished consideration as an ideal declaration of principles essential to the purpose declared in its opening lines.
Perhaps it is unnecessary to close with the statement that it is my opinion the demurrer should have been sustained.