After wending its way through the
Federal Court system in Ontario and a subsequent move to the
Quebec Superior Court, this post features the two most recent decisions in the Quebec courts pertaining to the interlocutory injunction sought by VIA against CN. These decisions, rendered in French and subsequently translated by Google Translate and me to English, are presented herein in their entirety with only light editing.
Some of the initial details have been published in previous posts and will be familiar to loyal Trackside Treasure readers, (feel free to scroll to the new, emerging material herein) but are presented again here for completeness as they form an initial basis for understanding the [new] "Analysis" portion of the judgement. Footnotes for each case are included, and some terms do not survive translation well i.e. train driver = engineer. The lack of photos originates with the dry legal documents which are completely unillustrated!
I was especially impressed by the analysis presented by Serge Gaudet, Judge of the Superior Court, both for its completeness and for its knowledgeable discussion of fairly technical train-related data and systems.
This post includes the April 23, 2025 Quebec Superior Court interlocutory injunction hearings and the related June 20, 2025 Appeal Court of Quebec appeal of the rejection thereof (top photo). I apologize in advance of your reading this post for the total lack of train photos! Now prepare a beverage, get out a pen and legal pad, and prepare to immerse yourself in this calamitous and consequential case.
* * * APRIL 23, 2025 - JUDGEMENT ON INTERLOCUTORY INJUNCTION * * *
On April 23, 2025, the Superior Court of Quebec chose not to intervene and grant an interlocutory injunction against CN. (VIA had applied to the Quebec Superior Court on an urgent basis to seek permanent and interlocutory relief, and for a safeguard order if interlocutory relief cannot be adjudicated in the very short term, to prevent continued severe and irreparable harm caused by CN's unreasonable, excessive and abusive conduct.) The Court's decision says that Transport Canada, as the authority responsible for rail safety in Canada, is best-positioned to decide this matter.
Let's begin with an 'executive summary' paragraph lifted from the analysis below by Judge Serge Gaudet:
Considering all the relevant circumstances, I agree that it would be inappropriate for the Superior Court to issue the requested interlocutory injunction, given that the Minister of Transport has already taken up the specific issue in dispute here, that she has all the necessary powers to protect VIA’s interests (and those of the public in general) and is prepared to exercise them if necessary, and that, moreover, the Department of Transport of Canada, because of its expertise in railway safety, is clearly in a better position than the Superior Court to resolve the issue.
Hearings and cross-examinations were held over the first two weeks of April, 2025, on a tight timetable befitting the interlocutory nature of the proceeding at hand. The parties completed their cross-examinations from April 3 to 7 (examinations were even held on Saturday and Sunday), while the hearing was held a few days later, on April 10 and 11, 2025.
In the course of the legal process in Quebec, VIA and Transport Canada (TC) were able to obtain key information from CN that had been previously unshared. These new insights will significantly support VIA's ongoing discussions with TC and CN. VIA claims that CN's latest data confirms that none of VIA's trains are detected at crossings any differently than other trains operating in the Corridor.
VIA's CEO shared the outcome, and VIA does not view this as a loss. In fact the main legal case - a request for permanent injunction - is still ongoing. VIA's General Counsel is reviewing the Court's decision and evaluating next steps in the legal, operational and administrative avenues.
Via Rail Canada Inc. v. Canadian National Railway Company
2025 QCCS 1269
SUPERIOR COURT
CANADA
PROVINCE OF QUEBEC
DISTRICT OF
MONTREAL
NO.:
500-17-133267-255
DATE:
April 23, 2025
______________________________________________________________________
Presiding:
THE HONOURABLE
SERGE GAUDET, J.C.S.
____________________________________________________________________
VIA RAIL CANADA INC.
Plaintiff
v.
CANADIAN NATIONAL RAILWAY COMPANY
Defendant
______________________________________________________________________
JUDGEMENT
(Interlocutory Injunction)
______________________________________________________________________
BACKGROUND
Via Rail Canada Inc. (“VIA”) is a federal Crown corporation mandated to operate a passenger rail transportation business across Canada. Except for a few portions of railways it owns, VIA does not own a rail network. Its passenger trains run primarily on the Canadian National Railway Company (“CN”)[1] tracks, particularly in the Quebec City-Windsor corridor, where VIA carries out the vast majority of its passenger transportation and which generates approximately 80% of its revenues[2].
As of 2022, VIA began using Venture trains manufactured by Siemens. These replace the Legacy trains previously used by VIA, which are at the end of their useful life. The transition is not yet complete, so VIA currently uses both Venture and Legacy trains in its operations.
On October 11, 2024, relying on a clause in the contract that binds the parties[3] and allows it to set safety rules on its network, CN implemented restrictions that have the practical effect of requiring Venture train drivers to slow down and visually check that the signaling devices (lights, bells or gates) at more than 300 level crossings in the Quebec City-Windsor corridor are working properly for a minimum period of time before entering them. Normally, such a check is not necessary because the level crossings are equipped with systems that automatically trigger these devices in a timely manner.
CN imposed these restrictions on Venture trains because it considers that they raise a problem of “shunting”. As will be seen below, shunting is the technical means by which the arrival of a train near a level crossing triggers the automated signaling devices. VIA believes that these restrictions, which create significant delays to scheduled schedules and significantly disrupt its operations, are unjustified because they are disproportionate and unreasonable.
According to VIA, the relevant data do not demonstrate that Venture trains have a shunting problem, and numerous tests would establish the opposite. It considers that CN is acting arbitrarily and in bad faith[4], and that these restrictions are in fact motivated by another dispute between the parties that is taking place before the Canadian Transportation Agency regarding the renegotiation of the conditions of the Transport Service Agreement.
Furthermore, citing the opinion of a specialist in experimental psychology, VIA believes that requiring Venture train drivers to perform numerous additional manoeuvres at almost every level crossing on their route risks causing excessive cognitive overload, which is likely to lead to catastrophic accidents. VIA therefore believes that, ultimately, CN's restrictions increase rather than reduce the risk of rail accidents and that rail safety requires that they be immediately suspended.
In November, 2024 VIA filed an appeal for judicial review before the Federal Court seeking the quashing of CN’s directive of October 11, 2024, and the suspension of its effects during the proceedings (stay). This application for a stay was to be heard by the Federal Court on February 25, 2025, after the parties had filed numerous sworn statements, expert opinions and other documents, often technical, totaling several thousand pages.
However, CN filed a motion requesting the preliminary dismissal of the appeal on the grounds that the Federal Court did not have jurisdiction to hear it since CN, in issuing its directive, did not act as a “federal board” within the meaning of the Federal Courts Act[5]. The Federal Court ruled in favor of CN and declared itself without jurisdiction over the dispute[6].
VIA then filed proceedings before the [Quebec] Superior Court in early March, 2025. While reserving its right to damages (the quantum of which remains to be determined), VIA is essentially requesting a permanent injunction prohibiting CN from applying the restrictions in question. It is also requesting that an interlocutory injunction to the same effect be issued during the proceedings. It is this latter request that is the subject of this debate.
CN vigorously contests VIA's request for an interlocutory injunction, denying any bad faith, arguing that the restrictions it put in place in October, 2024 are necessary for public safety and that such safety imperatives must prevail over VIA's commercial interests. It is of the opinion that the requested interlocutory injunction must be refused because the conditions required to obtain such an order are not met.
CN also argues that the Superior Court should not interfere with the administrative process initiated in December, 2024 by the Minister of Transport of Canada under the Railway Safety Act[7], which is specifically aimed at determining whether the restrictions put in place by CN on October 11, 2024, are appropriate in the circumstances to ensure railway safety.
I emphasize that, given the urgency invoked by VIA, an extremely tight timetable was put in place by the Superior Court so that the debate on the interlocutory injunction[8] could be conducted quickly, despite the numerous sworn statements and expert opinions filed as evidence and the resulting cross-examinations.
Thus, the parties completed their cross-examinations from April 3 to 7 (examinations were even held on Saturday and Sunday), while the hearing was held a few days later, on April 10 and 11, 2025. I also emphasize that VIA’s introductory application is some 115 pages long and that the file is extremely voluminous (surely approaching tens of thousands of pages), in particular because VIA chose to file as exhibits before the Superior Court all the documentation produced before the Federal Court with regard to the applications for judicial review and stay[9].
Finally, it will already be understood that the debate raises questions of a very technical nature and that the parties, on both sides, have filed expert reports in support of their respective claims. That said, I would point out that the Court was not able to benefit from viva voce explanations on the content of these reports – which are not necessarily easy to read – as none of these experts was called to testify. After explaining what shunting is and the crucial role it plays in railway safety, I will outline the relevant factual context, then briefly describe the content of the expert reports and then move on to analyze the application for an interlocutory injunction in light of the applicable legal principles.
SHUNT AND OTHER TECHNICAL CONSIDERATIONS
To fully understand the debate, it is important to begin by explaining the concept of "shunting," as well as other technical aspects associated with it. To prevent collisions between a train and a vehicle (or other road user), when a train approaches a level crossing, automated systems activate powerful light (flashing red lights) and sound (bell) signals to warn road users of the imminent arrival of a train. In principle, according to current regulations, these light and sound signals must be activated approximately twenty seconds before the train crosses the level crossing.[10] Such automated systems can also activate barriers that lower to prevent road users from crossing the railway when a train approaches the level crossing, which must be horizontal for a certain time before the train arrives, depending on the maximum speed allowed at that location[11].
If, for some reason, the light and sound signals or barriers do not activate quickly enough, this is called a "short warning time", which obviously creates a significant risk of accidents, as road users are not warned sufficiently in advance (according to the regulations) of the imminent arrival of a train. These automated signaling systems are activated by the interception of an electric current flowing in the rails near the level crossing, and it is this diversion of electric current that is called a "shunt" in English or a "shunting" in French. The expression comes from the English "to shunt" which refers in particular to the fact of carrying out the diversion of a flow, electric or otherwise[12].
In railway matters, shunting works in the following way. A portion of the rails located near the level crossing is crossed by a weak electric current. When a train arrives on this portion of the railway track, the wheels and axles of the convoy (which are made of a conductive material) intercept the electric current circulating in the rails, which causes a "short circuit" (i.e. the length of the electric circuit is reduced). It is this short circuit (or "shunt") which is detected by the system and which then activates the signals or barriers located at the level crossing in order to warn road users of the imminent arrival of a train.
to witThere are two types of automated system of this kind.
The first is based on a motion detector. In this configuration, the system detects the presence of the approaching train without being able to determine its speed. Warning times must therefore be established based on the fastest trains traveling on the track to avoid "short warning times." This has a disadvantage, because slower trains may take longer to arrive at the crossing, with the consequence that people who are warned of the train's presence by activated light or sound signals but who do not see a train coming may decide to ignore them and cross the track despite these signals, which creates a risk of collision.
A more sophisticated system (implemented, as I understand it, by CN at most level crossings in the Quebec City-Windsor corridor) is based on technology capable of determining the speed of the approaching train upstream of the crossing. It is then possible to adjust the warning time according to the speed of the approaching train. These systems are called "motion predictors" because they are designed to anticipate the time of the train's arrival at the level crossing. However, this assumes that the train will travel at a constant speed after triggering the short circuit. If the train driver accelerates after this point, there could be a "short warning time" resulting not from a miscalculation by the automated system but rather from a driver maneuver that thwarts these calculations, just as there could be a "long warning time" if the driver significantly reduced speed after the system detected the presence of the train.
Whether one or the other of these automated systems is used, it is understood that the ability of a train to perform adequate shunting is essential for the safety of both people who may be on the train (employees, passengers) and those near a level crossing (driver, cyclist, pedestrian). It is also understood that this shunting capacity is directly dependent on the conductivity of the wheels and axles of this train so that they are able to intercept the weak electric current flowing in the rails. According to the regulations in force, the resistance of the wheels and axles of a train to an electric current must not exceed 0.06 ohm[13]. Even if such a maximum electrical resistance standard is met, it still happens that the shunting does not take place as expected, which can lead to inadequate detection of the train by the system and therefore to a warning time that is shorter than required.
In the language of the railway industry, these inadequate detections of the short circuit are called "loss of shunt" or "erratic shunting". These events are rare, but their consequences can of course be catastrophic. They are therefore subject to constant monitoring by railway companies and regulatory bodies in matters of railway safety. A loss of shunt or erratic shunt occurs when the electrical current does not flow as well as expected through the wheels and axles of a train, affecting the system's ability to detect the presence of the train. Loss of shunts are essentially unpredictable events with multiple causes. In some cases, the electrical current does not flow as efficiently as desired due to rust on the rail or the presence of some contaminant (grain residue, leaves, grease, etc.) that provides increased resistance to the electrical current. In other cases, it is due to the fact that the contact surface between the train wheel and the rail is reduced due to the characteristics of the train. This occurs particularly in the case of trains that are shorter, faster and lighter, and also when the wheels of a train are relatively new or have been recently serviced[14].
These latter factors are more common in passenger trains. Indeed, freight trains are generally much longer, heavier and slower and their wheels are not renewed as often, which promotes a better contact surface between the wheel and the rail and therefore improves the flow of electric current between them[15]. Since shunting is done through the wheels and axles, the length of a train and therefore the number of pairs of wheels and axles it has has an impact on shunting.
A locomotive or railcar has four pairs of wheels and therefore four axles. A train with one locomotive and three railcars will therefore have 16 axles, while a train with two locomotives and six railcars will have 32. Freight trains, which are sometimes very long, can of course have a large number of axles, while passenger trains are generally much shorter. The VIA Venture trains of interest here have 24 axles, as they consist of one locomotive, four railcars, and a cab car, which also has four axles. [16]
FACTUAL CONTEXT
It would be easy to get lost (without gaining much) in the intricacies of such a complex and voluminous case. Furthermore, since this is an interlocutory decision, I will limit myself to a broad outline of the most relevant factual elements. The Transport [Train] Service Agreement (TSA) and the dispute before the Canadian Transportation Agency. The parties entered into a TSA in 2009[17]. This agreement sets out their respective rights and obligations regarding VIA's use of CN's tracks. This agreement has been renewed multiple times and remained in effect, as amended over time, until June 30, 2023. The TSA establishes CN's right to set reasonable rules to govern VIA's passenger service and VIA's obligation to comply with them:
3.5 Rules Governing Operations
CN shall make reasonable rules, orders, and regulations to govern the Passenger Service. The operation of the Passenger Service shall at all times be carried out in conformity with CN's rules, orders, and regulations and with the Agency's orders and requirements.
Despite negotiations to renew it, the parties were unable to agree on the terms of a new TSA. On June 2, 2023, relying on the right conferred by section 152.1 of the Transportation Act[18], VIA applied to the Canadian Transportation Agency to decide the dispute and to determine the respective rights and obligations of the parties with respect to VIA’s use of CN railways for the operation of its public passenger transportation service. On June 27, 2023, the parties signed a Status Quo Agreement pursuant to which the terms of the TSA (as amended) remain in force during the proceedings before the Canadian Transportation Agency.
The documents filed with the Agency show that the conflict between the parties is serious. VIA claims that CN is using its dominant position to favor its own commercial operations at the expense of much-needed improvements in passenger service, while CN believes that its obligations to VIA place it at a competitive disadvantage and that the only real way for VIA to achieve its performance and frequency increase targets would be to build its own rail network.[19]
THE 2021-2022 SHUNT TESTS
In 2018, VIA signed a contract with Siemens to manufacture 32 Venture trains to replace Legacy trains that were reaching the end of their useful life. As already mentioned, these trains consist of a locomotive, four coaches, and a cab car, for a total of 24 axles. Siemens determined the electrical resistance of the wheels and axles of these trains to be 0.001264 ohms, well below the maximum standard of 0.06 ohms required by regulations.[20] In the summer of 2021, VIA contacted CN to organize dynamic shunting tests for Venture trains on its network. CN then indicated that it would instead carry out simulations, notably using data collected on similar Venture trains already used in the United States[21].
On October 15, 2021, following these simulations, CN authorized the Venture trains in their 24-axle configuration, while reserving the right to implement any restrictions should their monitoring reveal shunting problems.[22] CN will authorize the use of the new 24-axle train sets on its routes. A ten-day notification to CN is required before placing these trains in revenue service on any service route. CN will deploy test equipment at strategic locations along its route to monitor these movements. CN reserves the right to implement any restrictions it deems necessary should shunting problems occur.
VIA subsequently retained experts (HATCH) to conduct tests on the shunting capacity of the Venture trains. These tests showed that the electrical resistance of the wheels and axles was below the standards of the American Railway Engineering and Maintenance-of-way Association (AREMA). In addition, dynamic shunting tests were conducted by the same firm in 2021 and 2022 on the VIA rail network (located between Montreal and Ottawa) as well as on the Metrolinx network (in the Toronto area), and these were also conclusive. [23]
DEPLOYMENT OF VENTURE TRAINS
Starting November 8, 2022, VIA will begin using Venture trains in its operations, first on the Quebec City-Montreal-Ottawa route, and then, starting in October 2023, on the Ottawa-Toronto and Montreal-Toronto routes. Venture trains are then used with increasing regularity. On May 1, 2023, in response to a request from VIA, CN confirmed that its periodic monitoring (every 60 and 120 days) revealed that the shunting of Venture trains is acceptable: "Graphs show acceptable shunting with the 24 axle configuration." [24]
THE OCCURRENCE OF SHUNTING PROBLEMS IN THE DRUMMONDVILLE SECTOR
On March 22, 2024, CN informed VIA that it had detected cases of "potentially inconsistent shunting" with passenger trains at certain level crossings on the Drummondville Subdivision. The CN representative then mentioned that VIA might have to increase the number of axles on its trains, but that CN did not yet have enough information to reach a conclusion in this regard. [25] On the same day, CN issued a directive stating that passenger train drivers must now apply the requirements of Rule 103.1 (f) of the Canadian Rail Operating Rules (CROR)[26] with respect to seven level crossings in the Drummondville sector. A few days later, CN limited the scope of this restriction to passenger trains with fewer than 32 axles. On March 28, CN also subjected six other level crossings in this sector to this same restriction, still for trains with fewer than 32 axles.
Thus, as of March 28, 2024, passenger trains with fewer than 32 axles at 13 grade crossings in the Drummondville sector must comply with the requirements of Rule 103.1(f) CROR. This rule reads as follows:
When special instructions indicate that rusty rails or other conditions may be present, the occupancy of grade crossings equipped with automatic warning devices must be manually protected unless it has been confirmed that the warning devices have been operating for the minimum required time. This means that if a level crossing equipped with automated signalling devices is subject to such a requirement, it must be protected by a physical person warning road users of the imminent arrival of a train (which is called “manual protection” in railway jargon), failing which the train driver must ensure that the automatic signalling devices have “operated for the minimum required time”[27] before the train crosses the level crossing.
On April 1, 2024, CN sent VIA the documentation justifying the implementation of these restrictions. On the one hand, it communicated data relating to “short warning times” in the Drummondville sector. On the other hand, it sent a PowerPoint document showing examples of erratic shunting on VIA trains in the Drummondville sector[28]. This latter document also contains certain results of the research work of the National Loss of Shunt Technical Committee on the causes of “loss of shunt” and on ways to remedy them[29]. This committee was formed in 2014 in the United States with representatives from several railway companies, including CN. The PowerPoint presentation states that the committee is about to report its findings in a white paper to be published in 2024. [30]
IMPLEMENTATION OF THE DIRECTIVE AT THE HEART OF THE DISPUTE
(CROSSING SUPPLEMENT)
On October 11, 2024, without prior notice to VIA, CN issued a new directive. This directive specifically targets VIA's Venture trains and now affects more than 300 level crossings, virtually all of the crossings in the Quebec City-Windsor corridor equipped with predictors. The directive—which covers a long list of grade crossings classified according to the various subdivisions of the corridor—reads as follows:
Unless operating with 32 axles or a shunt enhancer, the following miles of crossings listed under each subdivision must be manually protected, unless the warning devices are known to have operated for at least 20 seconds in accordance with CROR 103.1(f).[31]
Since VIA Venture trains have 24 axles and, moreover, the installation of a shunt-enhancer is currently prohibited by federal regulations,[32] the practical effect of CN's decision is to require VIA Venture train drivers to comply with CROR 103.1(f) for each of the approximately 300 grade crossings in question.
It is hardly feasible to post a physical person at each of the level crossings in question at all VIA train crossings, so this means in practice that Venture train drivers must, for each of the level crossings in question, reduce the train speed sufficiently to be able to visually verify that the warning devices have operated for the minimum required period, and then accelerate again.
On October 18, 2024, after VIA asked what data was based on this new restriction affecting practically the entire Quebec City-Windsor corridor without receiving a clear answer, the president of VIA sent a letter to CN. He mentioned that the company's operations were severely affected by this restriction affecting more than 300 level crossings, that nothing required VIA trains to have 32 axles and that 24-axle Venture trains had been authorized by CN on its network in 2021. He added that the application of this restriction significantly increases the cognitive load of Venture train drivers, thus reducing rail safety rather than improving it[33].
On November 5, 2024, CN's Chief Network Operating Officer Mr. Patrick Whitehead sent a response[34]. He states that the October 11 decision is essentially based on CN's work as a member of the National Loss of Shunt Committee, in addition to the short warning time events on the Drummondville Subdivision. He also mentions that, based on data collected over the past ten years by the committee, shunt losses are attributable to the following factors: "consistently trued wheel profile, minimum axle count, lighter equipment, and head hardened rail." He goes on to write that shunt losses have been observed on Venture trains used by Amtrak, which are very similar to VIA's Venture trains. He specifies that CN is prepared to collect data to assess the risk of catastrophic accidents, but that "it is not prepared to take a risk of loss of life, however statistically low it may be." Mr. Whitehead indicates, however, that he has formed a team to conduct additional tests on the shunting capacity of Venture trains.
PUBLICATION OF THE NATIONAL LOSS OF SHUNT COMMITTEE REPORT
(WHITE PAPER)
On November 15, 2024 the National Loss of Shunt Committee published its report entitled "Summary Report of Root Cause and Remediation for Loss of Shunt on Passenger Trains"[35]. I note that a senior CN official, Mr. Michael Burgett (Signals & Communication Division Manager), who interacted with VIA representatives on rail safety matters regarding Venture trains, was the National Loss of Shunt Committee's Project Manager at the time of the White Paper's release and was involved in its drafting. This report is extremely technical in nature and not easy to read, to say the least. Here, we can simply quote the first lines of its executive summary:
Short passenger trains (those with fewer than 32 axles) have a history of poor shunting, which can lead to the loss of vital wayside train detection and crossing activation failures. These issues present clear risks to both railroads and the public, and have contributed to fatal accidents in the past. The National Loss of Shunt Committee, formed in 2014, has extensively studied the contributing factors to these shunting problems over the past decade, seeking solutions.
Through significant research, the root causes of loss of shunt issues for passenger trains have been identified as: equipment with consistently “true” wheel profile, fewer axles compared to freight trains, lighter equipment weight, and hard-headed rail.
Since modifying any of these factors would be “highly impractical for railroad operations,” the committee instead recommends the installation of “shunt enhancers” on passenger trains with fewer than 32 axles. As I understand it, a "shunt enhancer" is a device installed under the locomotive at the front of the train, just above the rails, and which serves to induce an additional electrical current of sufficient strength ("wetting current") in the rails, thereby improving the shunting. [36] As already mentioned, such a device is not permitted in Canada, as regulations require a minimum distance between the rails and the underside of a locomotive.
REQUEST FOR INFORMATION FROM TRANSPORT CANADA
As we will see in more detail below, the Railway Safety Act grants significant powers to the Minister of Transport of Canada to ensure the safety of railway operations in the country. In particular, the Minister may, for the purpose of verifying compliance with the provisions of the railway law and regulations, request by order that a railway company provide information. [37]
On December 10, 2024, Transport Canada forwarded such a request for information to CN. [38] The Minister requires in particular that CN communicate to her for a large number of specified level crossings "all available records demonstrating warning activation times, including the time gates (if equipped) were in the horizontal position, for each crossing activation for the period starting in May 2024 to the date of this Order", as well as, for the same period "any other pertinent data that has been compiled and analyzed by the supplier of equipment of each grade crossing warning system".
In the letter accompanying the order, Transport Canada states that this information is being requested in response to the directive issued by CN on October 11, 2024, and that it will be used to determine whether there is a safety issue with the shunting of Venture trains, which could, if necessary, lead to orders against either party to mitigate any risks that may exist.[39]
It is Transport Canada’s understanding that CN’s rationale was that VIA’s trainsets are too light and too short to properly and consistently shunt—or activate—the sensors in the track that set off the grade crossing warning systems with sufficient advance time to meet regulatory requirements under the Grade Crossings Regulations (GCR), and that this may increase the risk of a collision between a train and road traffic.
Transport Canada takes this matter seriously and, through this order, is seeking information to ensure compliance with grade crossing requirements such as those found in the GCR, and ascertain whether there is an active safety issue. This information will assist TC in determining next steps, which could include requiring companies to implement corrective measures to mitigate any risks.
It goes without saying that Transport Canada's request for information involves the disclosure of massive amounts of raw data, and while CN has already submitted most of the relevant data to Transport Canada, its analysis is not yet complete, as discussed below.
SHUNTING TESTS ON THE CN NETWORK
In mid-December 2024, as announced in Mr. Whitehead's letter of November 5, CN would conduct shunting tests on Venture trains on certain portions of its network. These tests revealed only one "short warning time," and CN suspects that this was in fact a case where the train accelerated after triggering the system. However, CN specifies that these tests were conducted in wet weather, thus under ideal conditions for electrical conductivity. The CN report also mentions that the shunting profile of Venture trains with 28 axles shows signs of a “potential floating shunt situation”, and even more so for trains with 24 axles, even if no “floating shunt” was detected during the tests in question[40]:
During the testing we saw three distinct shunting profiles. The freight train has an ideal shunting profile, where the train is detected by the crossing warning system once it gets on the approach, and it shunts in a linear manner. The 28-axle VIA shunting profile shows that the crossing warning system also detects the train once it hits the approach, but the shunt profile is moving away from being linear with signs of erratic shunting. Lastly, the 24-axle VIA shunting profiles show that the crossing warning system has a hard time picking up the train once it hits the approach; the profile is even less linear than the 28-axle profile, and it is far more erratic as well. Looking at both the 28- and 24-axle profiles, there are signs of a potential floating shunt situation. During our testing, no floating shunt was detected due to the weather conditions, but given the data, the potential exists.
NEW DATA PROVIDE BY CN IN FEBRUARY, 2025
On February 13, 2025 as part of the information gathering required by Transport Canada, CN provided VIA with new data. This data reported 206 short warning time events between August, 2024 and January, 2025 at six crossings in the Drummondville sector. According to CN, 90% of these events were associated with Venture trains.[41] As will be seen below, VIA's experts dispute these conclusions, believing that CN's data are unreliable.
THE EXPERTISE
The parties have filed numerous expert opinions and supplementary expert opinions in support of their respective claims—and this will come as no surprise given the nature of the debate—[42] These expert opinions can be grouped around two fundamental issues:
- In light of the available data, is CN's directive of October 11, 2024 a reasonable safety measure, as CN claims, or is it a disproportionate, unreasonable, and unjustified measure, as VIA claims?
- Does the application of the directive create risks of rail accidents by unduly increasing the cognitive load of Venture train drivers?
Again, without going into too much detail, it is useful to briefly summarize the content and conclusions of these expert opinions.
REASONABLENESS OR OTHERWISE OF THE OCTOBER 11, 2024 DIRECTIVE
EXPERTS RETAINED BY VIA (MESSRS. FARAROOY, KIRMAN AND IRELAND)
The experts retained by VIA on the question of the reasonableness of the October 11, 2024 directive are Messrs. Saeed Fararooy, Harold T. Kirman, and John Ireland, all of whom work for DB E.C.O. North America Inc. ("DB"), a subsidiary of Deutsche Bahn AG, Germany's state-owned railway company. These individuals have extensive railway experience, specifically in railway safety (Mr. Fararooy), railway operations (Mr. Kirman), and data analysis (Mr. Ireland).
I note that an initial report by Messrs. Fararooy and Kirman's report was filed in December 2024 in the Federal Court case, and a new report, this time with the participation of Mr. Ireland, was filed on March 3, 2025, in the current case. Since this latest report essentially repeats the content of the first report with some additional information, we will focus on it. These experts are of the opinion that CN did not follow railway industry practice or act reasonably when it implemented the October 11, 2024, directive.
This conclusion is based in particular on the following:
- CN did not have sufficient data at that time to suddenly impose "a blanket restriction across 304 crossings in the Corridor," since there were only 148 short warning time events, without it being clear that these were actually related to VIA trains[43];
- The directive applies only to Venture trains, while Legacy trains, which often have only 16 axles, are not affected;
- It does not apply if there is a shunt-enhancer, but such a device is contrary to Canadian regulations;
- CN ignored the tests conducted by Hatch and Siemens regarding the shunting capacity of Venture trains;
- CN did not conduct a risk assessment before implementing the directive, contrary to applicable regulations and industry practice.
They also believe that CN "did not provide sufficient analysis and evidence to justify the crossing supplement." According to these experts, the data used by CN is insufficient, incomplete, or erroneous. The February, 2025 data also contains "significant data and methodology errors and gaps," which they detail over several pages. In short, these experts are of the opinion that the data on which the CN relied to issue the directive of October 11, 2024 lacks any probative force and therefore cannot justify such a draconian measure.
These experts also mention that the conclusion of the National Loss of Shunt Committee report that passenger trains with 32 axles or fewer should be equipped with "shunt-enhancers" is at odds with the report's own findings, which indicate that it is trains with fewer than 24 axles that are likely to have shunting problems. They also point to research and measures adopted by railway companies in the United States and Europe that are not in harmony with the findings and conclusions of the National Loss of Shunt Committee, none of which concern trains with 32 axles or fewer, although some concern trains with fewer than 30 or 28 axles. They are also of the opinion that the risks of collision at a level crossing equipped with automatic signaling devices involving a Venture train are extremely low and that, even if the consequences of such a collision are "critical", the risk remains, in their opinion, "tolerable". They indicate that in many cases, the "short warning times" were too short by only a few seconds, which does not significantly increase the risks.
Finally, not only do they believe that the directive of October 11, 2024 was not necessary from a rail safety perspective, but that it actually creates more risks of accidents by causing cognitive overload for Venture train drivers.
A supplementary report by Mr. Ireland was filed at the last minute by VIA at the start of the hearing on the morning of April 10, 2025[44]. According to Mr. Ireland, new data disclosed by CN on April 5, 2025, during cross-examinations—which he claims are by far the most comprehensive he has seen to date—show that the "short warning times" detected on the CN network are not specifically associated with VIA trains but also with CN trains and that, consequently, this data "strongly suggests [that] CN's infrastructure is the cause [of the short warning times], not VIA Venture trains." This new report is accompanied by said data in the form of technical computer files spanning several hundred pages. [45]
EXPERT MANDATED BY CN (MR. LI-LIAN LUI)
CN retained the services of Mr. Li-Lian Lui, an Ontario engineer who owns a firm (Partum) specializing in rail safety. Mr. Lui has acted as an expert and expert witness on the subject of rail safety on numerous occasions in Ontario and Quebec. Again, Mr. Lui filed a report in the Federal Court case (January 30, 2025), and two other supplementary reports (March 23 and April 1, 2025) in the context of this application for an interlocutory injunction.
In his April 1 report, Mr. Lui indicated that the consequences of inadequate shunting are not limited to the possible occurrence of "short warning times" at level crossings, but that this can also affect the operation of automated signaling intended to manage operations and traffic on the railway ("wayside signaling systems"), which can create a risk of collisions between two trains. He also considers that a short warning time of 3 seconds is a significant risk, contesting the opinion of VIA's experts on this subject who seemed to diminish its importance in their report.
Mr. Lui also conducted his own risk analysis based on Metrolinx's methodology and concluded that the probability and possible consequences of inadequate shunting of Venture trains created an "elevated" risk, such a risk requiring mitigation measures. Mr. Lui disagrees with DB's assessment, which considered the risks "tolerable," and points out that this assessment was made using methods that involved fewer factors to consider, which, in his opinion, diminishes its value[46].
Mr. Lui echoes the findings of the National Loss of Shunt Committee on the causes of "loss of shunt" (trued wheel profile, minimum axle counts (24 axles and under), lighter equipment consists, and head-hardened rail) and notes that the first three factors are related to train configuration rather than track infrastructure. He concludes that "the research conducted by the LOS Committee supports a conclusion that the poor shunting observed with VIA's Venture trains is due to the train equipment, and not to deficiencies with CN's infrastructure."
In this regard, Mr. Lui puts into perspective the value of the shunting tests conducted by HATCH in 2021 and 2022. He explains that, in light of the unpredictable nature of loss of shunts and the fact that the HATCH tests were conducted over only a few days, they are less reliable than the data collected by the LOS Committee over long periods of time.[47]:
Given the complexity and unpredictability of the loss of shunt phenomenon, comprehensive testing is required to produce reliable conclusions. Effective testing for loss of shunt requires monitoring of consistent locations within a certain mileage range over multiple months.
Mr. Lui believes that the October 11, 2024 directive is an appropriate method to mitigate the risks associated with a possible shunting problem with Venture trains. He then examines the various alternatives that VIA could implement to comply with the directive other than asking its drivers to slow down at every level crossing, and discusses their feasibility.[48]
I note that, with respect to Mr. Ireland’s April 9 report, its methodology and conclusions were not reviewed by CN’s experts. To avoid postponing the hearing, CN instead chose to respond with a sworn statement by Mr. Hoang Tran, Senior Director, Regulatory, System Safety, and Passenger Operations at CN dated April 10, 2025[49].
Mr. Tran explains that he is responsible for compiling and analyzing the data to be submitted to Transport Canada following the Minister’s request for information. He indicates that the files reviewed by Mr. Ireland in his April 9, 2025 report were not final documents, but rather draft spreadsheets. He estimates that he will need another three weeks of work to complete his analysis of the data.
Mr. Tran explains that Transport Canada sent CN a file in which some 11,756 alleged short warning time events were identified by department employees. CN must verify whether each of these alleged events is a genuine short warning time, and, if so, identify their cause and the train involved. He goes on to state that, according to CN’s examinations at this stage, the number of “valid” short warning times is closer to 2,000, but that this estimate is expected to decrease further as verifications of alleged short warning times are carried out.
Mr. Tran is therefore of the opinion that the calculations carried out by Mr. Ireland to determine the proportion of short warning times attributable to CN trains as opposed to VIA trains are therefore unreliable, especially since the analysis of the data used to precisely identify the train involved for each valid short warning time has also not been completed. Mr. Tran also disputes Mr. Ireland’s methodology and calculations for establishing his ratios of short warning times for VIA trains compared to CN trains.
COGNITIVE LOAD OF VENTURE TRAIN DRIVERS
EXPERT MANDATED BY VIA
(MS. CHRISTINA M. RUDIN-BROWN OF HUMAN FACTORS NORTH)
Ms. Rudin-Brown holds a Ph.D. in experimental psychology and extensive experience in the field of human factors, which studies the relationships between human behavior and other elements of a given system, particularly in decision-making. The mandate entrusted to her by VIA's attorneys is to provide her assessment of: 1) the cognitive load required of Venture train drivers in light of the requirements of the October 11, 2024 directive; 2) the risks associated with this cognitive load; and 3) if applicable, the degree of urgency of the measures that must be taken to reduce this load to an acceptable level.
Using three separate methods, Ms. Rudin-Brown compared the cognitive load of Legacy train drivers (who are not subject to the October 11, 2024 directive) with that of Venture train drivers, who must follow this directive and therefore the requirements of REFC 103.1 f). She first counted the tasks to be performed. There are five demanding tasks (high-level task steps) for Legacy train drivers compared to ten such tasks for Venture train drivers, because the latter must slow down upon arrival at the level crossing, determine whether the signals have been working for the required time, prepare to slow down or even stop the train if the signals are not working as expected, and then accelerate again. She then inserted the sequence of these different tasks into diagrams that show that the tasks of Venture train drivers are more mentally demanding.
Finally, and most importantly, she interviewed a dozen experienced VIA train drivers and asked them to assess their task performance in light of certain criteria defined by a methodology used by NASA, depending on whether they were required to comply with the directive or not. Ms. Rudin-Brown concluded that the cognitive load of Venture train drivers who were required to comply with the directive exceeded acceptable standards. She wrote:[50]
Based on subjective workload ratings from, and discussions with, twelve experienced LEs[[51]], it is apparent that the mental workload associated with the approach to crossing restrictions is significantly greater than for the same task without crossing restrictions. Of particular concern is that subjective ratings on the NASA [Task Load Index] increased by more than double and approached the maximum value in some of the workload subscales.
Whether these ratings indicate an acceptable level of workload was assessed using the Bedford Workload Scale, which has been used by other organizations (e.g., NASA) to judge task acceptability. Importantly, the average Bedford Workload Rating of 6.4 for the "approach to crossing" task while operating the crossing restrictions exceeds NASA's maximum rating of 3 for critical or frequent tasks, and the maximum rating of 6 for infrequent, non-critical tasks.
According to her, this cognitive overload leads to increased risks of mental fatigue and deficient performance in performing the tasks required for safe train operation. She adds that "the elevated workload could also result in inattentional blindness, where visual cues or hazards, such as warning signals, are not detected or identified because the [locomotive engineer's] attention is directed exclusively at the approach to crossing task." Regarding the timeframe within which action would be required to reduce cognitive load to an acceptable level, the expert states that this is "urgently needed," given the significant excesses of NASA's standards. She clarified that, in her opinion, the question is not whether an accident will occur in relation to the increased workload imposed by the directive of October 11, 2024 but rather when such an accident will occur[52]:
It is my opinion that it is not a matter of whether a workload-related incident will happen during or near in time to the approach to crossing task under the crossing restrictions but when.
EXPERT COMMISSIONED BY CN
(MR. JASON KUMAGAI OF 30 FORENSIC ENGINEERING)
CN commissioned Mr. Kumagai—who has over 30 years of expertise in human factors—to review Ms. Rudin-Brown's report and provide his opinion on the validity of her findings and conclusions. According to Mr. Kumagai, Ms. Rudin-Brown's report is affected by numerous methodological deficiencies that completely invalidate its results. These limitations can be briefly summarized as follows:[53]
- The task list for Venture trains includes tasks that are not considered for Legacy trains but are nevertheless likely to occur during the operation of these trains;
- The comparison of Ms. Rudin-Brown's tasks ignores the technological differences between Venture and Legacy locomotives, the former being more modern than the latter;
- The Bedford questionnaire should only be used for specific tasks and not as an overall workload rating;
- Ms. Rudin-Brown used a modified version of the NASA Task Load Index, which diminishes the value of the exercise;
- The number of train drivers interviewed was limited; moreover, they necessarily had little experience operating a train in accordance with the directive's requirements and could be biased toward it;
- The assessment of task performance was purely subjective and based on participants' recollections, without an objective assessment of performance at the time the tasks were performed.
After detailing these limitations and their influence on the results, Mr. Kumagai proposes the methodology that should have been followed to obtain conclusive results.[54]
ANALYSIS
As mentioned in the introduction, VIA is requesting that the Court issue an interlocutory injunction ordering CN to cease implementing the October 11, 2024 directive and its amendments. VIA is also requesting that CN be prohibited from implementing special instructions under Rule 103.1(f) that would only apply to passenger trains with fewer than 32 axles and not equipped with a shunt enhancer.[55]
An interlocutory injunction is a measure intended to “preserve the subject matter of the litigation so that effective relief will be possible when the case is finally decided on the merits.”[56] The criteria to be weighed by the Court when deciding whether or not to grant such an injunction are well established:
1) does the application disclose a serious issue to be tried?[57];
2) will the person seeking the injunction suffer irreparable or serious harm[58] if his or her request is not granted; and
3) does the balance of convenience test favour the granting of the injunction?[59]
These criteria must be assessed globally and not in isolation because they are interrelated and influence each other.[60] Furthermore, it is well established that the decision to grant an interlocutory injunction is one that is discretionary. Indeed, the interlocutory injunction is a measure that comes from the common law, more precisely from the equitable jurisdiction of the English courts, and this implies that the Court always retains the power not to grant an injunction if it is of the opinion that such a measure is not appropriate in the particular circumstances of the case.
In this regard, it is important not to confuse permanent injunctions with interlocutory injunctions. In Quebec, developments have led to the injunction (whether negative or mandatory) becoming the preferred procedural means for obtaining specific performance of an obligation[61]. Since in Quebec civil law, the creditor has the choice of remedy in the event of non-performance, and can demand specific performance “in cases that permit it”[62], the discretion of a Quebec judge to refuse a permanent injunction in such a case is more limited than that of the common law courts. Thus, Justice Frédéric Bachand, writing for the Court of Appeal in the case of Syndicat des copropriétaires du 310, 320, 330 et 340 Boulevard industriel c. 9322-0549 Québec inc[63], determined that equitable principles that are incompatible with Quebec’s rules of specific performance (for example, the “clean hands” or laches theory) have no application in matters of permanent injunctions aimed at implementing obligations falling under private law[64]. However, Justice Bachand is careful to point out that this analysis does not concern the case of interlocutory injunctions[65]:
I digress to clarify that the preceding analysis is limited to the specific situation of permanent injunctions aimed at obtaining specific performance of a private law obligation. Entirely different considerations come into play when a party seeks to obtain the forced performance of such an obligation by means of an interlocutory injunction (including an interim interlocutory injunction), and it is well established that the rules of English law concerning the criteria for granting such a procedural measure can then play a certain supplementary role.
He relies in this regard on the decision in Groupe CRH, where the Court of Appeal clarified that, although the interlocutory injunction is codified in articles 510 and 511 of the CCP, it remains that "the issuance of an interlocutory injunction remains a discretionary power of the same kind as that exercised in equity in common law jurisdictions"[66] and that it is therefore appropriate to refer to common law principles as a supplementary measure, as the Supreme Court of Canada had already indicated in Trudel v. Clairol Inc. of Canada[67]. In short, with regard to interlocutory injunctions, the considerations that may lead a common law court to refuse to grant them in certain circumstances where this measure would seem inappropriate are also applicable as a supplementary measure in Quebec law.
One of these considerations that seems relevant to me in this case is the fact that an administrative process, not only adequate but better placed than the Court to resolve the debate, is currently underway.
This is an important element that may lead a court to exercise its discretion to refuse to grant an injunction, as the Labrador and Newfoundland Court of Appeal noted in Nunatukavut Community Council Inc. v. Nalcor Energy[68]:
Chief among the discretionary considerations that must be addressed is whether there is an alternative remedy available. This is because equitable remedies are generally regarded as supplementary to other available remedies. It is only when such supplementation is needed that the appropriateness of an injunction enters the picture. In most cases, this will involve a consideration of whether the claim can be properly remedied by an award of damages. (…) In addition, question as to whether effective protection of the rights that have been or are threatened to be interfered with can be achieved either directly or indirectly by other mechanisms such as by invoking some other sort of statutory remedial process (…).
In his reference work on injunctive relief, Robert Sharpe[69] notes the reluctance of courts to issue injunctions when the interest asserted by the plaintiff can be adequately protected by administrative processes carried out by specialized bodies[70]:
[…] deference has been shown by the courts to the specialized expertise and procedures of various administrative agencies and where the interest the plaintiff seeks to protect falls squarely within that competence and some non-judicial remedy is available to protect the right, the courts have tended to refuse injunctive relief.
In the same vein, the author mentions that granting an injunction may be inappropriate, particularly when the specialized body in question has powers allowing it to issue orders similar to injunctions:[71]
In many areas, injunctive relief from the courts may be inappropriate because jurisdiction has been specially assigned to an administrative board or tribunal. Often, that assignment of jurisdiction will include the power to make restraining orders very similar to injunctions […]
An example of the application of these principles can be seen in the case of Association des sociétés de téléphone du Québec inc. v. Attorney General of Canada,[72] where Justice Stratas of the Federal Court of Appeal was seized of an application for a stay[73] of certain CRTC decisions. Justice Stratas stated that, even if a Court had the jurisdiction to issue a stay or an injunction, “it may decide not to exercise it” because “other effective administrative remedies may exist”[74] and that, moreover, “another court may have greater expertise or be better placed to decide the issue”[75]. Justice Stratas thus dismissed the application for a stay submitted to him, without further consideration of its merits, on the grounds that it could, under the relevant legislative provisions, be addressed to the Governor in Council, which had the powers required to grant it, was prepared to exercise them and was better placed than the Court to decide the question of its appropriateness.
Shortly thereafter, in Strickland v. Canada (Attorney General)[76], the Supreme Court of Canada confirmed that when making a discretionary decision[77], the court may choose not to grant the requested remedy – even if the applicant would otherwise have been able to establish the merits of his or her claim – particularly when there is an “adequate alternative” to which he or she could have resorted. In doing so, it identifies the factors to be taken into consideration in exercising this discretion. Relevant considerations include the convenience of the other remedy, the nature of the other tribunal that could rule on the issue and its ability to provide relief to the applicant, the existence of an adequate and effective remedy before a tribunal already seized of the dispute, the speed, relative expertise of the other decision-maker, and the economical use of judicial resources and costs[78].
The Court clarifies that "for an alternative remedy or forum to be adequate, it is not necessary that the procedure or remedy be identical" to that available through the court process; rather, the question is whether the alternative remedy adequately resolves the applicant's grievance in light of all the circumstances of the case. [79] In this regard, the factors to be considered are not limited (the Supreme Court expressly rejects the notion of "check-boxes") and the court considering whether to exercise its discretion to dismiss the application before it in the presence of an alternative remedy must consider not only the existence of that alternative remedy, but also the relevance and appropriateness [of the court procedure] pursued. This "requires a balance of convenience type analysis" and the balancing of relevant factors "should take into account policy objectives and considerations underlying the legislative scheme in question." [80]
In this case, in its argument plan, CN, referring to the process initiated by the Minister of Transport in December 2024, is of the opinion that the Superior Court should “refrain from issuing an interlocutory injunction [which] would have the effect of short-circuiting a parallel procedure before another more appropriate forum”. It goes on to write that “if the law provides a specific procedure to deal with a specific issue, and that law provides an effective remedy to the applicant’s concerns, then it is appropriate to follow that procedure rather than invoking the inherent power of the courts to issue an injunction”[81].
Considering all the relevant circumstances, I agree that it would be inappropriate for the Superior Court to issue the requested interlocutory injunction, given that the Minister of Transport has already taken up the specific issue in dispute here, that she has all the necessary powers to protect VIA’s interests (and those of the public in general) and is prepared to exercise them if necessary, and that, moreover, the Department of Transport of Canada, because of its expertise in railway safety, is clearly in a better position than the Superior Court to resolve the issue.
Parliament adopted the Railway Safety Act (RSA, or LSF en francais). As its title indicates, this Act aims to ensure safety in railway operations. Its primary objective is “to provide for and promote the safety and security of the public and personnel in railway operations and the protection of property and the environment.”[82] This Act grants broad powers to the Minister of Transport to intervene in railway safety matters. According to section 32.01 of this Act, the Minister may, if he or she considers it necessary for railway safety, issue an order to a railway company "ordering it to cease any activity that could compromise railway safety or to follow any procedure or take the necessary measures specified in the order." Such an order is not limited in time. Furthermore, according to section 33(1) of the RSA, the Minister may, in an emergency, issue ministerial injunctions (negative or mandatory) by sending a notice to a railway company to this effect:
The Minister may, by sending a notice to this effect, order the railway company concerned to cease, in whole or to the extent provided for in the notice, the use of railway installations or equipment of a specific type, or any practice relating to their maintenance or operation that, in the Minister's opinion, may imminently compromise railway safety. It may similarly order the Minister to implement a certain practice regarding this maintenance or operation when a failure to do so poses such a risk.
Such a ministerial injunction may be in effect for a period of one year (maximum duration of six months, which may be extended for another six months by the Minister).[83] The RSA specifies that ministerial orders and injunctions "may be assimilated to orders of the Federal Court or a superior court" and that "their enforcement shall be carried out according to the same terms and conditions." Thus, the order under section 32.01 is similar to a permanent injunction of a superior court, while the "ministerial injunction," of a temporary duration and in cases of emergency, is more similar to an interlocutory injunction or a safeguard order of a superior court. Furthermore, as already mentioned, the RSA gives the Minister the power to issue an order requesting a railway company “to provide him, in the form and within the time period provided therein, any information or document that he considers necessary to verify compliance with this Act and its implementing regulations”.
This is precisely what the Minister of Transport did in December 2024 when she issued Order # MO 24-01 by which she asked CN to provide her with the relevant data in order to assess whether there is indeed a safety issue (“an active safety issue”) with the shunting of Venture trains and this, with the aim of determining whether measures must be put in place with regard to one or the other of the parties “to mitigate any risks”[84]. We therefore see that the legislative regime put in place by the legislator in matters of railway safety offers the applicant a process which is fully capable of adequately responding to the protection of the interests which it seeks to protect by its request for an interlocutory injunction.
VIA claims that the directive issued by CN regarding Venture trains is disproportionate and unreasonable and that its application increases rather than reduces the risk of rail accidents and that its application must therefore be suspended immediately.
If the Minister concludes that this is the case, she has all the necessary powers to order CN to cease applying the directive to Venture trains. If she deems it necessary, the Minister can even act urgently, by simply sending a notice ordering CN to cease this practice, which order will have the same enforceability as an injunction order from the Superior Court.
In short, the remedies provided for in the RSA are entirely adequate to protect the interests that VIA seeks to protect by its application for an interlocutory injunction.
I would also point out that VIA itself indicated to Transport Canada that orders or injunctions "under Section 33 and 32.01 of the Railway Safety Act would represent a much more expedient way to resolve the issue (…)"[85]. Not only does the administrative regime established by the legislature in matters of railway safety provide an adequate remedy, similar to the injunction in the Code of Civil Procedure, but, as we have seen, the Minister has taken up the issues that are the subject of this debate and Transport Canada is currently in communication with CN in order to obtain and analyze data to determine whether there is indeed a safety problem with the Venture trains, whether CN's measures are justified from a railway safety point of view and whether other measures (ministerial orders or injunctions) should be implemented in this regard.
However, as mentioned above, this analysis process is still ongoing and Transport Canada has not yet reached definitive conclusions and has not ordered anything at this time. When an administrative process likely to provide an adequate remedy to the applicant is underway, the courts must be even more cautious before issuing an interlocutory injunction that could interfere with it. It is important to avoid a judicial decision harming the administrative process, especially when, as in this case, it involves an organization with specialized expertise on the subject of the dispute and which says it is ready to act to exercise its powers[86].
It would be rather embarrassing, for example, if the Superior Court issued the injunction requested here by VIA (and therefore prohibited CN from applying the directive during the proceedings) and the Minister of Transport subsequently concluded, following her investigation, that this directive is necessary to ensure railway safety. The risk of such decisional inconsistency should be avoided by allowing the administrative process provided for in the RSA to follow its course, without judicial interference until a decision is rendered by the Ministry[87].
Furthermore, it seems clear to me that in this case the federal Ministry of Transport is clearly better placed than the Superior Court to decide this dispute. Essentially, VIA claims that the directive is unjustified since the data relied on by CN does not show that there would be a shunting problem with Venture trains, which would be confirmed by the various tests carried out since 2021. For its part, CN is of the opinion that the data establishes that there would be a shunting problem with 24-axle Venture trains, and relies in particular in this regard on the conclusions of the National Loss of Shunt Committee's white paper. As described above, the experts disagree on the value to be given to a particular test or to a particular data item, and even on the validity of the Loss of Shunt Committee's conclusions.
Furthermore, VIA claims that the application of the directive increases rather than reduces the risk of rail incidents and relies on the expertise of Ms. Rudin-Brown on this subject, the validity and value of which is contested by Mr. Kumagai, CN's expert. I would point out that both have considerable expertise in the area of "human factors." These are clearly questions of rail safety, and such questions, within the administrative regime established by the RSA, must be decided by specialists, either the Department of Transport or the members of the Transportation Appeal Tribunal of Canada.[88] Of course, it often happens that ordinary courts are required to rule on the probative value of conflicting expert reports in various fields, but there is a complexity in the present debate which seems to me to be out of the ordinary.
Indeed, on the one hand, the amount of raw data that must be analyzed in this case to draw conclusions seems astronomical and, on the other hand, the methods followed or the calculations established by one or the other to draw conclusions remain fairly obscure, at least for a court of ordinary law. To be convinced of this, one need only examine Mr. Ireland's report of April 9, 2025 and the computerized files that accompany it, or to note that the experts from CN and Transport Canada themselves have difficulty adequately reading the relevant raw data, which requires elaborate exchanges between them over several months, in particular to clarify the question of what are the real "short warning times" that should be taken into account in assessing the situation.
In short, the question of the conclusions to be drawn from the relevant raw data seems fairly complex here, even for the experts. Furthermore, as I have already mentioned, the parties did not see fit to have the experts testify before me. I am well aware that this is not usual in the context of an interlocutory injunction, but it remains possible[89] and could have been considered here in light of the complexity of the issues and the expert opinions submitted to the Court.
In these circumstances, it seems to me that the respective interests of the parties (and the safety of the population in general) are clearly better served by a decision from the experts of the Ministry of Transport than by an analysis of the situation by the undersigned, which could only be approximate, while I would have to decide on the record, without having had any explanations regarding many of the concepts that are not self-evident used by the experts in their reports, nor having had the benefit of the examinations and cross-examinations of the latter, who are all at the cutting edge of their ultra-specialized discipline.
However, and this seems fundamental to me, this is a matter of public safety in relation to operations that are likely to cause catastrophic accidents, whether they be collisions between a train and a vehicle or collisions between two trains, with all that this implies in terms of loss of life, very serious injuries and material damage. While being aware of the significant harm that maintaining the directive causes to VIA's operations, and the possibility that CN's directive increases the risk of incidents rather than reducing them, it would seem inappropriate to me to play the sorcerer's apprentice here in relation to an issue directly affecting public safety, when I have only heard the parties for two days, without hearing any experts, with a technical file of several thousand pages, and this, when an adequate administrative system for processing VIA's request exists and is currently underway with rail safety specialists who are looking into the situation.
The balancing of the respective advantages of the administrative process over the judicial process seems to me to clearly lean in favour of the former rather than the latter. I add that the Superior Court can only grant or reject the requested injunction without being able to consider other measures, unlike the Minister of Transport, who can adapt the content of an order or a ministerial injunction to alternative solutions that may be identified along the way and that may prove better suited to the situation. In First City Financial, cited above, the Ontario High Court, rather than issuing an interlocutory injunction, preferred to let the Securities Commission deal with the issue of disclosure of information between two rivals in the context of a takeover of a company, precisely because of the greater flexibility offered by the administrative system compared to the judicial process. Justice Reid writes:
It seems to me that the Commission is better fitted by its jurisdiction, its nature, and its expertise to maintain a watchdog type of supervision over the required level of disclosure in this kind of contest than is the Court. It is more flexible both in its ability to receive and deal with complaints. If disclosure is insufficient in the Commission's opinion, it is not caught in the "yes" or "no" situation that the Court is by reason of its nature and its jurisdiction.
The federal legislature has established a sophisticated, efficient and flexible administrative system to ensure railway safety in the country. This process relies on the expertise of the Department of Transport in this area. The Minister can issue the necessary orders for railway safety and even ministerial injunctions in an emergency and adapt them to the situation before him. Appeals are provided for before the Transportation Appeal Tribunal of Canada[90], whose members must have expertise related to the transportation sector in question when they hear a case, as already mentioned.
There is, it seems to me, a clear legislative desire to leave railway safety issues in the hands of specialists who also have all the flexibility required to ensure such safety without imposing standards that would be unduly restrictive for railway operations. This legislative will in such a matter seems entirely appropriate and allows for the rapid and effective adoption of solutions which may be required in the area of railway safety, a hyper-regulated, ultra-specialised and, by necessity, complex field.
I also emphasize that this dispute, in its essence, is not truly legal. Certainly, if the directive of October 11, 2024 is unreasonable, CN does not have the contractual right to adopt it because it can only adopt reasonable rules according to the contract that binds the parties. However, this legal determination is entirely dependent on the real question that arises, which is a pure question of railway safety: is the directive of October 11 necessary from the point of view of railway safety, as CN claims, or is it unjustified and more dangerous than anything else by creating a cognitive overload for the train drivers who must apply it, as VIA claims? Since these are first and foremost technical questions relating to railway safety, it is better to leave it to the Ministry's experts, who are already seized of the matter, to establish what is required here or not to ensure railway safety, without having to suffer interference that could prove "embarrassing" from the ordinary courts.
I add in this regard that the Minister of Transport, although advised by VIA of its concerns regarding the urgency of lifting the application of the directive due to the risks that it could create by cognitive overload for Venture train drivers, did not deem it appropriate at this stage to issue an emergency ministerial injunction.
I will therefore dismiss the application for an interlocutory injunction, because I believe that the applicant has an adequate remedy within the administrative process provided for in the RSA, a process which has already been initiated and is currently underway.
It is therefore neither necessary nor appropriate to further consider the other criteria for granting an interlocutory injunction since, even assuming that these criteria can be met here (on which I make no pronouncement), I am of the opinion that it would be inappropriate to grant the requested interlocutory injunction, preferring to leave it to the Department of Transport to consider this complex issue of rail safety in order to determine, if necessary, and without judicial interference at this stage, what measures should be taken in this regard.
FOR THESE REASONS, THE COURT:
REJECTS the application for an interlocutory injunction filed by the plaintiff VIA Rail Canada Inc. in the application entitled "Originating Application for a Permanent Injunction, Interlocutory Injunction and Safeguard Order" dated March 4, 2025. With legal costs in favor of the defendant.
-Serge Gaudet, Judge of the Superior Court
Bogdan Catanu, Jean-Christophe Martel, Laurent Crépeau, Mathias Heilke, Omar El Baba,
Woods LLP,
Counsel for the Plaintiff Via Rail Canada Inc.
Vincent Rochette,
Jean-Simon Shoenholz, Maya Angenot, Dominic Dupoy, Heather Clark, Florence Méthot,
Norton Rose Fulbright Canada LLP,
Counsel for the Defendant Canadian National Railway Company
Hearing Dates:
April 10 and 11, 2025
Footnotes:
[1] CN, which was once a federal Crown corporation, was privatized in 1995: CN Commercialization Act, S.C. 1995, c. 24.
[2] Declaration of Nicholas MacKenzie, para. 5 (RDDII, Vol. 1, Tab. 15).
[3] The 2009 Transport Service Agreement, as amended.
[4] VIA claims, in particular, that CN's restrictions were imposed without warning on the eve of Thanksgiving, its busiest weekend of the year.
[5] RSC 1985, c. F-7.
[6] Judgment of February 19 of Justice Catharine Moore (RDDII, Vol. 1, Tab. 3)
[7] RSC 1985, c. 32 (4th Supp.).
[8] I understand that VIA, faced with this alternative by the Court’s management, preferred to proceed with the interlocutory injunction in an expedited manner rather than first submitting a safeguard order and then submitting its application for an interlocutory injunction.
[9] Unfortunately, the way in which the exhibits were classified (for example, all of the exhibits filed before the Federal Court became Exhibits P-1 and P-2, while all of CN’s exhibits before the Federal Court became Exhibit P-3, each of Exhibits P-2 and P-3 totaling several thousand pages) made the list of exhibits accompanying the application to institute proceedings particularly difficult to use, even with a computer. I therefore asked the parties to provide me with additional tools to facilitate my task. On April 14, VIA's attorneys therefore sent me four large, stubby notebooks entitled "Collection of Documents Cited in the Originating Application" (hereinafter "RDDII"), listing each document or excerpt from a document or examination mentioned in the said originating application, in order of appearance. For their part, CN's attorneys provided a new version of their detailed argument plan, the references to which were more specific than those in the original version. I thank the attorneys for acting expeditiously to produce these additional documents, without which it would have been virtually impossible to draft these reasons within a reasonable time.
[10] Section 103.1(a) of the Canadian Rail Operating Rules (CROR).
[11] Ibid.
[12] For example, in medical matters, there may be a "shunt" of the blood circulation.
[13] Section 103.1(a) of the Canadian Rail Operating Rules (CROR). 12.5(b) of Transport Canada's Grade Crossing Standards (VIA Compendium of Legislation, Tab. 6).
[14] What those in the industry call "trued" wheels.
[15] Reference.
[16] This "cab car" is useful for steering the train when the locomotive is pushing rather than pulling the train, as Venture trains are bidirectional, unlike Legacy trains.
[17] CN Compendium, Vol. 1, Tab. 11.
[18] S.C. 1996, c. 10.
[19] See the references to paras. 45 et seq. of VIA's Application to Initiate Proceedings and the arguments filed before the CTA at tabs 21 and 14 of Volume 1, RDDII.
[20] RDDII, Vol. 2, Tab 44.
[21] Statement of Mr. Quintal, para. 4 (RDDII, Vol. 2, Tab. 36).
[22] RDDII, Vol. 2, Tab. 41.
[23] RDDII, Vol. 2, Tabs. 46 and 47.
[24] RDDII, Vol. 2, Tab. 55 (p. 5).
[25] RDDII, Vol. 3, Tab. 57.
[26] Better known in the railway industry by its acronym “CROR” for “Canada Rail Operating Rules”.
[27] As already indicated, this “minimum required time” is in principle 20 seconds for level crossings not equipped with gates. Where there are gates, if the maximum authorized speed is greater than 15 miles per hour, the gates must be observed by the train driver to be in a horizontal position for at least 5 seconds before the arrival of the train and if the maximum authorized speed is less than 15 miles per hour, the gates must be observed in a horizontal position when the train arrives at the level crossing. See paragraph (a) of section 103.1 REFC. [28] I understand that, in these graphs, a good shunt should show a fairly smooth linear downward line, while a problematic shunt will show more or less significant vertical variations in this downward line, which is therefore not as smooth.
[29] RDDII, Vol. 3, Tab. 62.
[30] Ibid., p. 65.
[31] I would point out that, on November 27, 2024, a second version of the directive was issued to adjust the text to regulatory changes relating to the “minimum time required” depending on whether the level crossing has light and sound signals or barriers. See Mr. Panetta’s sworn statement of December 12, 2024, para. 65 et seq. (RDDII, Vol. 1, Tab. 8). A third version of the directive was issued on February 20, 2025, to slightly modify the number of level crossings covered (RDDII, Vol. 4, Tab. 96). These two additional versions of the October 11, 2024 directive, however, do not significantly modify its nature or scope for the purposes of this debate.
[32] This is not contested by CN. See Mr. Lui's expert report of January 30, 2025, p. 32.
[33] RDDII, Vol. 3, Tab. 73.
[34] RDDII, Vol. 3, Tab. 9.
[35] What the parties' witnesses call the "White Paper" (RDDII, Vol. 3, Tab. 79).
[36] The technical details of these devices are provided in Appendix B of the report: idem.
[37] Art. 36(1) RSA.
[38] Tab. 9 of Vol. 1 of the CN Compendium.
[39] Ibid. (emphasis added).
[40] RDDII, Vol. 3, Tab. 87.
[41] Sworn statement of Mr. Hoang Tran dated February 13, 2025, RDDII, Vol. 4, Tab. 94.
[42] See VIA's Expert Report (to which he added John Ireland's additional expert report dated April 9, 2025) and CN's Expert Report.
[43] Report of March 3, 2025, pp. 21-22.
[44] Exhibit P-87.
[45] Ibid. These data were not explained to me, nor are they explained in Mr. Ireland's brief report. [46] Report of January 30, 2025, pp. 1033 to 1036 and pp. 1041 to 1044.
[47] Ibid., p. 1040.
[48] Ibid., pp. 1045 to 1048.
[49] Tab 9a of the CN Book of Sworn Declarations.
[50] Report of March 3, 2025, p. 12.
[51] i.e. train drivers ("locomotive engineers" in English).
[52] Report of March 3, 2025, p. 13 (italics in original).
[53] Report of January 30, 2025, pp. 1662 to 1676. A reading of the report shows that the summary at the beginning of Mr. Kumagai's report is faithful to the detailed content of the report.
[54] Ibid., pp. 1677-78. [55] I recall that, with respect to the merits of the case, VIA is seeking the same conclusions by way of a permanent injunction, the award of compensatory (undetermined amount) and punitive damages, and reserves its rights to seek any alternative order to order CN to take corrective measures if it is shown that its trains are not adequately shunting. See the conclusions of the Originating Application for a Permanent Injunction, Interlocutory Injunction and Safeguard Order of March 4, 2025.
[56] Google Inc. v. Equustek Solutions Inc., [2017] 1 SCR 824, para. 24.
[57] In certain cases, the prima facie case may be more stringent. Thus, in the case of a mandatory interlocutory injunction, the Supreme Court requires a strong prima facie case: R. v. Canadian Broadcasting Corporation, 2018 SCC 5. Furthermore, if the granting of the injunction amounts to a practical final disposition of the dispute, the judge will have to conduct a more careful examination of the right invoked by the applicant: see Groupe CRH inc. v. Beauregard, 2018 QCCA 1063, para. 28.
[58] This criterion is that indicated in article 511 of the Code of Civil Procedure.
[59] Google inc., supra, para. 25; RJR-MacDonald inc. v. Canada (AG), [1994] 1 SCR 311;
[60] Canadian Society for the Prevention of Cruelty to Animals v. Longueuil, 2022 QCCA 1690, para. 25.
[61] On the introduction of the injunction in Quebec law in the 19th century as a provisional measure and its complex relationship with specific performance in civil law, see G. Massé, “L’exécution des obligations via l’astreinte française et l’injonction québécoise,” (1984) 44 R. du B. 659.
[62] Art. 1601 C.c.Q.
[63] 2023 QCCA 892.
[64] Ibid., para. 44.
[65] Ibid., para. 46 (italics in original).
[66] Groupe CRH, supra, para. 25.
[67] 1975 2 RCS 236, p. 246.
[68] 2014 NLCA 46, para. 65.
[69] I note that Mr. Sharpe, who taught and was Dean at the University of Toronto Faculty of Law, was appointed a judge of the Ontario Superior Court in 1995 and then of the Ontario Court of Appeal in 1999.
[70] Sharpe, Injunctions and Specific Performance 2024, loose-leaf ed., para. 3.22.
[71] Ibid.
[72] 2012 FCA 203.
[73] A stay and an interlocutory injunction are remedies of the same nature: Manitoba (AG) v. Metropolitan Stores Ltd, [1987] 1 SCR 110.
[74] He relies in this regard on Canadian Pacific Ltd. v. Matsqui Indian Band, (1995) 1 SCR 3 and C.B. Powell Ltd. v. Canada Border Services Agency, 2010 FCA 61.
[75] See Reza v. Canada, 1994 2 SCR 394.
[76] 2015 SCC 37.
[77] This case involved an appeal for judicial review. The Federal Court had been seized of an application seeking a declaration that the federal child support guidelines were invalid because they were not provided for in the Divorce Act. The Federal Court dismissed the application on the grounds that the provincial superior courts, which had the power to rule on this issue, were better placed to resolve the issue, as the Federal Court had virtually no expertise in family law. The Federal Court of Appeal and then the Supreme Court upheld this decision.
[78] Ibid., para. 42.
[79] “As Brown and Evans put it, ‘in each context the reviewing court applies the same basic test: is the alternative remedy adequate in all the circumstances to address the applicant’s grievance?’”: topic 3:2100 (emphasis added).” I note that the French version of the judgment mentions "en toutes circonstances" but that this is an awkward translation of the expression "in all the circumstances" which rather means in the context in which it is used "in light of all the circumstances." The following paragraph confirms this, since the Court states that the "list of factors is not limited, as it is up to the courts to identify and weigh them in the context of a given case" (emphasis added).
[80] Ibid., paras. 43 and 44.
[81] CN Argument, para. 39.
[82] Art. 3(a) RSA.
[83] Art. 33(2) and 33(6) RSA.
[84] CN Compendium, Vol. 1, Tab. 9.
[85] Email dated February 4, 2025, from Mario Péloquin, President of VIA, to Transport Canada (tab. 10 of the CN Compendium, p. 1).
[86] See Burnaby (City) v. Trans Mountain Pipeline ULC, 20-14 BCSC 1820; First City Financial Corp. Ltd v. Genstar Corp., 125 DLR (3d) 303 (1981, Ontario High Court of Justice). As early as 1923, the Supreme Court of Canada established that courts must be careful not to “embarrass” administrative bodies validly seized of disputes by issuing interlocutory injunctions or declaratory judgments that could interfere with their decision-making process: City of Lethbridge v. Canadian Western Natural Gas, Light, Heat and Power Co., 1923 SCR 652, (see in this regard the reasons of Davies, Mignault, Duff and Anglin JJ.A.).
[87] Should the Minister fail to render a decision within a reasonable time, the applicant could still apply to the Federal Court for mandamus to obtain a decision. Should the decision rendered be unreasonable or otherwise illegal, the Federal Court may review it by way of judicial review.
[88] Members who must themselves be specialized in the transportation field in question, except for the Chairperson and Vice-Chairperson of this Tribunal (s. 12(4) of the Transportation Appeal Tribunal of Canada Act, S.C. 2001, c. 29). See ss. 32.1 and 32.2 RSA for appeals to this specialized tribunal.
[89] Art. 106 C.C.P.
[90] Art. 32.1 and 32.2 RSA
On June 20, 2025 the Appeal Court of Quebec issued a judgement on an application for leave to appeal from a judgment rendered during the proceedings on April 23, 2025, by the Honourable Serge Gaudet of the Superior Court. The judgement was rendered after a 40-minute hearing with lawyers representing VIA and CN. The Appeal Court of Quebec is the highest judicial court in the province.
10:07 a.m. Beginning of the hearing. Identification of the case and attorneys.
10:08 a.m. Preliminary remarks by the judge.
10:09 a.m. Arguments by Mr. Catanu.
10:17 a.m. Interventions by the judge and responses by Mr. Catanu (2nd error raised - paras. 231, 235 et seq. DII).
10:26 a.m. Arguments by Mr. Rochette.
10:37 a.m. The judge informs the attorneys that she will be able to deliver her judgment today, during the day. The judgment will be rendered in the minutes of that day and sent to them as soon as it is available.
End of the hearing.
The applicant, VIA Rail Canada Inc. ("VIA Rail"), is seeking leave to appeal the judgment rendered by the Superior Court on April 23, 2025 (the Honourable Serge Gaudet)[1], which dismissed its application for an interlocutory injunction ordering the respondent, the Canadian National Railway Company ("CN"), to cease applying (1) restrictions at all mileages of certain grade crossings[2]; and (2) special instructions under section 103.1(f) of the Canadian Rail Operating Rules[3]. Following the appeal, VIA Rail is asking the Court to issue these orders or refer the matter back to the Superior Court.
The dispute between the parties arose on October 11, 2024, when CN imposed restrictions on VIA Rail's new "Venture" trains. The latter then filed an application for an interlocutory injunction to order CN to lift these restrictions alongside a request for a permanent injunction and compensatory and punitive damages.
