A Call for Moderation in Canada’s Divisional Regime
Canadian divisional practice has quietly but decisively shifted. What was once a pragmatic safety valve in prosecution has increasingly been framed as something suspect—tolerated when forced, frowned upon when voluntary, and often viewed through the lens of double patenting risk rather than treaty compliance or prosecutorial reality. Compared to peer jurisdictions, Canada is drifting toward a posture that is unusually restrictive.
This post argues for recalibration. Not permissiveness without limits, but balance: a regime that respects Canada’s international obligations, acknowledges how inventions are actually drafted and examined, and preserves public notice without turning divisionals into a presumptive vice.
The Paris Convention: The Forgotten Starting Point
Any serious discussion of divisionals has to begin with the Paris Convention. Canada committed—long ago—to a system that permits divisional filings, including voluntary divisionals, subject to patent office control. That bargain is not optional. It reflects an international consensus that unity is examined, not divined at filing.
Unity objections are the mechanism by which patent offices police overreach. Divisionals are the mechanism by which applicants respond. The two are inseparable. Treating divisionals as inherently problematic risks collapsing that balance and, more importantly, risks placing Canada out of step with the treaty framework that underpins global patent practice.
Drafting Reality: Seeing the Forest Before the Trees Exist
It is unrealistic to expect inventors and drafters to identify every divisible invention at the drafting stage. Complex technologies do not present themselves as neatly bounded claim sets. Reasonable, skilled practitioners can—and frequently do—disagree about unity until examination pressure reveals fault lines.
That disagreement is not gamesmanship. It is a function of how innovation unfolds and how claims interact with prior art. A regime that implicitly assumes bad faith whenever claims later separate into multiple inventions misunderstands both invention and prosecution.
Amendment Late in Life: Not a Moral Failing
A related theme has emerged in recent years: the suggestion that amending claims later in prosecution, particularly in light of potential infringers, is somehow improper.
That framing is historically thin. When Canada adopted deferred examination, the model explicitly contemplated that applicants would wait, learn, and then shape claims with a clearer view of the commercial and competitive landscape. Deferred examination was not an accident; it was a policy choice grounded in efficiency and fairness.
Using that very feature of the system should not be retroactively cast as sharp practice. The patent system has always tolerated—indeed expected—iterative refinement.
The Other Side of the Ledger: Public Notice and Proliferation
None of this is to deny the countervailing concern. Endless divisionals can erode public notice and burden the register. At some point, proliferation defeats the very signaling function that claims are meant to serve.
This concern is real, and it deserves weight. A system that allows serial, marginally distinct divisionals without discipline risks undermining confidence in the patent bargain itself.
But acknowledging this does not justify treating divisional practice as inherently improper. The solution lies in calibrated limits, not categorical suspicion.
Bargain Theory as a Unifying Frame
Recent case law underscores why a calibrated approach is needed. There are many decisions pointing in this direction and they need not be canvassed in detail here. However, NCS Multistage Inc. v. Kobold Corporation, 2025 FCA 187, and Shionogi & Co., Ltd. & ViiV Healthcare Company (Re), 2025 CACP 11, illustrate the tension well. While applicants are entitled to pursue divisional protection, claims that are not patentably distinct from earlier claims will fail—not because divisionals are improper, but because the bargain has already been struck once. At the same time, the broader trend toward restrictiveness in Canadian divisional practice is unmistakable, to the point that Canadian patent agents routinely caution applicants to adopt a highly conservative posture in Canada compared to peer jurisdictions.
Bargain theory provides a useful lens. The patent system is an exchange: disclosure in return for time-limited exclusivity. Divisionals sit at the margins of that exchange, mediating between full disclosure and administrability.
The real question is not whether divisionals are good or bad. The question is how much divisional practice is consistent with the bargain. That inquiry necessarily involves trade-offs—between notice and fairness, between efficiency and completeness, between early certainty and prosecutorial reality.
Framing divisionals as presumptively abusive short-circuits that analysis.
Deference Where It Belongs: Unity Objections
One final note of caution. In recalibrating divisional doctrine, Canada should not drift toward routine challenges to CIPO’s unity determinations themselves. High deference to unity objections is both sensible and stabilizing. Unity is the front line; divisionals are the response.
Undermining confidence in unity determinations risks compounding uncertainty rather than resolving it.
A Call to the all Stakeholders in the Canadian Patent System
Canada’s current trajectory places its divisional regime increasingly out of step with international practice. The emerging emphasis on divisional filings as inherently improper lacks balance and risks distorting the patent bargain.
The better path is moderation: respect for treaty obligations, realism about drafting and prosecution, discipline against abuse, and principled deference to administrative expertise. Divisionals are not the problem. Losing sight of the balance is.
