Where Should Cases Go? Or Where Should Patent Law Be Made?

"Gedanken ohne Inhalt sind leer, Anschauungen ohne Begriffe sind blind." — Immanuel Kant (Thoughts without content are empty, intuitions without concepts are blind.)

Recent reactions to my earlier contribution have raised a familiar concern: the debate on UPC case distribution would remain too abstract, insufficiently grounded in data, and lacking concrete proposals. The concern is understandable. Practitioners expect actionable levers, institutions are judged on outputs, and a system as recent as the UPC invites immediate operational scrutiny. Yet the objection risks reversing the order of analysis. One does not begin with measures. One begins with the nature of the problem. Otherwise, what appears as pragmatism quickly becomes a competition of preferences. The question is not whether something must be done, nor whether successful divisions should be penalised for their success. It is whether the current distribution of cases affects the conditions under which UPC law is formed. That is not an abstract question. It is the most consequential one.

The UPC is a success. It has attracted litigation, delivered decisions, and rapidly established credibility. The concentration of cases before certain local divisions reflects experience, predictability, and user confidence. This should be acknowledged without hesitation. But success does not end the inquiry. It reframes it. A court is not a market, even if users choose where to litigate. A market allocates demand. A court produces law. If the practical formation of that law is concentrated within a limited number of procedural environments, the issue is no longer merely where cases go. It becomes where law is made, under which conditions, and through which repeated practices. At that point, success becomes structure — and structure demands scrutiny.

The prevailing account treats case distribution as the natural outcome of rational user choice. This account is partly correct. But it is incomplete. Choice is structured. It reflects language, speed, professional habits, perceived judicial culture, and accumulated precedent. Once a division becomes predictable, it attracts more cases; because it attracts more cases, it becomes still more predictable. The mechanism is self-reinforcing. It requires no intervention. It produces its own equilibrium. That equilibrium, however, is not neutral. What is privately rational may be institutionally problematic. A system driven entirely by user preference may stabilise around a limited number of centres of gravity. Left unexamined, those centres will shape the emerging body of law — not by design, but by repetition. Early doctrinal patterns, once consolidated, are rarely reversed.

Calls for more data and more concrete measures are legitimate. Transparency can inform choices. Procedural adjustments can improve allocation. But neither data nor measures answer the prior question. Data describe the system. Measures adjust it. Neither determines how a common legal order is formed. Without a clear conceptual framework, empirical observations risk remaining blind; without engagement with empirical reality, conceptual analysis risks remaining empty. The issue is not whether data and measures matter. It is that they do not suffice — and that treating them as sufficient forecloses the more fundamental inquiry.

If the issue is institutional, practical responses must be visible, proportionate, and compatible with user choice. They should not suppress success, but prevent success from becoming dependency. The following proposals illustrate possible directions.

• Europeanising case preparation. More frequent designation of a non-local judge as rapporteur in high-volume divisions would rebalance the structuring phase of proceedings.
• Targeted judicial rotation and secondments. Short, practice-oriented exchanges across divisions would diffuse expertise and reduce structural asymmetries before they harden.
• Meaningful operational transparency. Publication of comparable indicators — timelines, procedural patterns, interim relief outcomes — would anchor forum choice in data rather than reputation alone.
• Capacity-building in less-used divisions. Visible support through mixed panels, mentoring in real cases, and shared practices would progressively strengthen confidence across the system.
• Calibrated transfer mechanisms linked to timelines. Where a division departs materially from the Court's own time targets, limited transfer tools could restore temporal coherence without removing party autonomy.
• Pooling of judicial expertise across divisions. More dynamic use of judges from the central pool would reinforce the perception — and the reality — of a genuinely common court.
• Light-touch harmonisation of case management. Establishing reference practices — without rigid uniformity — would reduce the procedural divergences that currently reinforce forum-driven expectations.

These measures are institutional safeguards. Their purpose is not to override choice, but to ensure that choice operates within a genuinely common court.

The usual objection is that intervention risks reducing attractiveness and pushing cases back to national courts. This risk exists and must be taken seriously. Any measure that creates uncertainty, increases transaction costs, or signals institutional instability could undermine what the UPC has built. But it is not the only risk — and it may not be the greater one. Persisting concentration shapes expectations, litigation strategies, and early doctrinal patterns in ways that compound over time. A body of law formed predominantly within a small number of procedural environments gradually reflects those environments — their pace, their culture, their assumptions. That is not a failure of intent. It is a structural consequence of design by default. A court may remain formally European while becoming operationally centred. Such a perception, once established, is not easily undone — and the corrective measures required at that stage would be far more disruptive than proportionate action taken early. Waiting is therefore, in itself, a structural choice, with its own costs and its own risks.

The UPC was not created merely to offer an additional forum. It was intended to address the historical fragmentation of European patent enforcement. Its legitimacy therefore cannot be measured solely by the number of cases it attracts. It must also be assessed by the conditions under which it produces law — the diversity of judicial perspectives it brings to bear, the procedural coherence it maintains across divisions, and the genuine commonality it embodies in practice. A European patent court is not simply a court used by European litigants. It is a court capable of producing European law under European conditions. This is the real issue behind case distribution. Not arithmetic. Not correction. Not distrust of the divisions that have earned their standing. But the deliberate, ongoing construction of a common judicial space — one that does not merely reflect demand, but actively sustains the conditions for a shared legal order. The success of the UPC is not in question. What remains open is whether that success will become the foundation of something genuinely European — or simply the expression of preferences that happen, for now, to converge.