Why would a dispute over a tiny sliver of farm lane in North East England, triggered by an obscure legal principle with a Latin name hardly anyone knows, be of interest to a reader of this article in Ohio?
The answer is simple: everywhere we have modern civilisation, we have law. And the effects of law can be life-changing. Children, homes, careers, liberties, and even lives themselves can be at stake. Everyone therefore has an interest in the structural properties of law — including its failure modes.
Small cases can be unusually revealing because they strip away the noise. Cases that cleanly expose the boundary conditions of law can become diagnostic of the system itself. The better we understand how it goes wrong, the easier it becomes to put it right.
My previous article was a court report on that very land dispute. As we arrived at the courthouse in Newcastle last week and signed in with the ushers, my friend Andrew Stephenson asked which judge was sitting on his case.
The answer was a familiar name: it was the same judge as in a previous case I had attended as observer, where dozens of people had come as public witnesses. I was so impressed by the judge’s handling of that situation that I wrote to him afterwards, thanking him for his commitment to public justice.
I am deliberately not repeating his name here, because the whole point is that this is not a personal matter. This is a man whom I respect, and his demeanour this time was similarly professional and welcoming. He was clear about the boundaries of my role as a lay adviser, including that if I didn’t follow the rules I would be asked to leave. It was said in a way that was matter-of-fact, not threatening. This is necessary and appropriate management of a court, and it is reassuring.
The outcome of the case wasn’t the one we sought: the claim was struck out. The judge was clear about his reasoning, even if we disagreed with it. Court correspondence since the hearing has reinforced the clarity of the decision. There is no suggestion here of incompetence, malevolence, or negligence.
It is easy to write polemics about rude or aggressive judges whose oppressive orders appear self-evidently biased or lacking any foundation in fact or statute. What is more challenging — and more productive — is to examine what happens when the “iniquitous decrees” come with honesty and diligence.
If injustice requires no bad actor, then good actors are not sufficient protection against it.
As we entered the room for the hearing, the usher apologised that while the skeleton arguments had been put before the judge in time, they had not yet been read. The judge took a few minutes at the start of the hearing to skim them both. He clearly had the overall shape of the dispute in his head; the issues lay in the specifics of the decision architecture and the correct path through it.
This front-end experience was mirrored by a back-end one. The same judge was needed in what I understood to be another case involving an infant. (It could instead have been that he needed to collect his own child from daycare; there was some ambiguity.) So the end of the hearing became a little rushed as he gave his ex tempore judgment. A ninety minute slot was, on the face of it, still a reasonable amount of time for the hearing.
The reason I am telling you this is that it sets up the rest of what I have to say. There is a finite amount of judicial mind and cognition on offer. The same has been true since the beginnings of systems of justice, which can be traced back through centuries and civilisations.
Meanwhile, facts and arguments can multiply almost without bound. They therefore have to be compressed into symbolic objects upon which adjudication can operate. This compression is necessarily lossy: some fidelity to reality must be surrendered. The important question is what gets lost.
Does the process omit essential aspects of reality?
Does it introduce fresh unreality of its own?
Or does it preserve the essential structure of reality despite the compression?
This case is useful as a teaching aid precisely because land has no opinions or feelings. There is no “he said, she said,” as there can be in child custody or divorce cases. The soil simply is. And the law ultimately requires its ownership to be determinable — even if, in the degenerate case, it remains officially unclaimed — rather than leaving it perpetually suspended in some metaphysical state of uncertainty.
By tracing a short and relatively simple chain of logic, we can see how reality itself can become lost in its legal representation. This is a quality failure in the output of the system.
As a quality control engineer, this fascinates me more than the specifics of the land dispute. Variability and error are ordinary. What is illuminating is how much survives despite AI removing cognitive scarcity, and management theory telling us how to put that cognition to productive use.
That the gap is readily discernible is precisely what makes the example worth studying. If we can see the defect clearly here, we can ask what quality-control mechanism is missing that would ordinarily detect and correct it — and what happens in harder cases where the same kind of defect may be much less visible.
So we aren’t looking at this as a personal failure of the judge: bad adjudication.
Nor are we looking at this as a reasoning failure of the law: bad doctrine.
What we are examining is a structural failure in the load-bearing machinery of the justice system: bad architecture.
What absent quality control mechanism made this outcome possible at all?
There are relatively few essential features of reality in this case:
There is a lane.
An 11kV electricity cable runs beneath part of the lane; it connects to a wind turbine.
The soil through which that cable passes has an ownership status.
Northern Powergrid (NPG) claims a legal right to keep its cable there.
That claimed derivative right depends upon rights relating to the land through which the cable passes.
The court was asked to determine rights arising from that situation.
Those ingredients are the underlying reality. Everything else — title documents, registered boundaries, leases, presumptions such as ad medium filum viae, pleadings, precedents, skeleton arguments and submissions — belongs to the representational machinery by which the legal system attempts to determine what follows from it.
Andrew has one representation of that reality.
His neighbour’s grant of rights over the land embodies another.
Northern Powergrid has another.
And the court has a further one.
But the reality doesn’t change because of its representation.
The question is whether these representations correspond to that reality — and whether, when they do not, reality remains capable of correcting them.
It is worth pausing for a moment to unpack “reality”, because the word can easily become an intellectual bludgeon. If NPG were writing a PR rebuttal of this very article, they might reasonably assert their own position as being “reality”. The word, by itself, is therefore of limited use in determining contested questions of fact and law.
In this situation, however, we can cleanly tease out the different facets of what we call “reality”.
First, there is empirical reality. The lane exists, and I have traversed it many times myself when going for Sunday dinner with Andrew. The cable is unquestionably underneath it. The land in question was conveyed by identifiable people on particular dates many decades ago. Little of this is in dispute in this case.
Second, there is an antecedent legal reality. At any given time, particular legal rights existed in relation to particular plots of land. NPG was granted rights over some land. The grantor either possessed the legal capacity to grant those rights over the land in question, or did not. The resulting rights are either effective against Andrew, or they are not. Andrew either has particular rights and obligations with respect to the cable, or he does not.
Third, there are rules for deriving consequent legal reality. Statutes, precedent, presumptions, doctrines and interpretative rules tell the system what legal consequences follow when those rules are applied to the relevant facts and antecedent legal state.
Finally, there are representations of all of this. Plans, registers, deeds, witness statements, pleadings, skeleton arguments, submissions, judgments and orders purport in different ways to represent empirical facts, legal states, applicable rules, reasoning, or their consequences.
In the previous article I introduced the idea of a “lex [i.e. law] crash” as a kind of legal accident zone. A lex crash occurs when the necessary relationships between these layers break down:
when representations lose fidelity to the reality they purport to describe,
when legal consequences cease to follow validly from their antecedents, or
when errors in either can no longer propagate backwards to produce correction.
The very essence of a justice system is to maintain fidelity between its representations and the realities they represent, and to reason rationally from antecedent states to downstream legal consequences. When this uncouples, we move from law to “shape of law”.
What makes this case useful is how clearly that chain has become disconnected. The “accident” is unusually easy to see. The boundary between lawful and law-shaped is visible.
Once we can all see “this isn’t real”, the debate moves to “why?”
As noted, the empirical reality is not in serious question in this case. The real issue is the antecedent legal reality:
Who owns the land the cable is under?
Hence, who had the right to grant NPG permission to run an industrial cable under it?
The law provides a route to an answer via the doctrine of ad medium filum viae, under which ownership of land adjoining a road may presumptively extend to the middle of the road, subject to rebuttal. But to start there is to locate ourselves at the wrong level of reality. We are already arguing about the rules.
The essence is simpler.
Either the disputed land is part of Andrew’s estate, or it is not.
There is no “Schrödinger’s soil” that hangs in an indeterminate proprietary state until somebody examines it.
If it is part of Andrew’s estate, then NPG are a trespasser, as they have not sought a grant from him. They proceeded on the assumption that the grant from his neighbour was sufficient. That assumption might be mistaken, but it doesn’t imply wrongdoing.
If it is not part of his estate, then NPG are not trespassing on Andrew’s land. Their grant from his neighbour is sufficient, and Andrew has no claim against them on that basis.
Exactly one of those underlying proprietary states is true; not zero, not both, not something else. The function of the legal machinery is not to create one by observing it, but to determine which one already follows from the relevant history and law.
What happened instead was that the claim was struck out because the court agreed with NPG that the real case was between Andrew and his neighbour; this was a derivative matter.
Worse, the attempt to get an answer to the simple question “who owns the land this cable is under?” was decreed an abuse of the court’s function because his neighbour was not joined to the claim.
Ignore for a moment that the proportionate remedy would be to add his neighbour as a party. Instead, let’s consider whether reality itself is part of the reasoning process.
Now run through each of the two branches: either the land is part of Andrew’s estate, or it is not.
If it is, then he has simply sued a trespasser. One doesn’t need to sue every neighbour or possible rival claimant to the land first. If NPG believe someone has a superior title to the land, it is their job to present it. Instead, they offered no theory of conveyancing history at all, only a modern map that is explicitly a general boundary, not an exact one.
If it is not, then Andrew is simply wrong. His claim should fail because the cable is not under his land. The court could reach that conclusion by establishing the relevant ownership, rather than because the question of ownership was supposedly one he could not ask of NPG.
What matters is this: the proprietary status of the land had no effect on the outcome either way under the court’s model of reasoning.
By reframing Andrew’s assertion of ownership as somehow “depriving” his neighbour of the land, the reasoning creates a logical architecture in which Andrew may have a property right, but cannot operationalise it.
That cannot be a complete model of how property law works; a right that cannot be exercised is empty.
And detecting this loss of reality is the very quality-control issue we care about.
Not downstream legal reasoning about logical jumps, circular arguments, or misunderstood case law.
If the reality of the soil and its ownership don’t matter, then how can the result be law?
The defect becomes even easier to see if we push it to its logical conclusion. What cause of action does Andrew have against his neighbour? After all, the neighbour has never made a formal claim over the whole lane. The grant explicitly gave NPG only such rights as the neighbour had authority to grant. The very purpose of the presumption that the middle line is the boundary is to prevent such neighbourly disputes.
It would essentially mean that NPG could deprive Andrew of the practical benefit of his property rights merely by signing an agreement with a neighbour and asserting the resulting uncertainty.
If so, the invariant has moved from “property is an enforceable right against all” to “property rights depend upon the identity of the interferer.”
If permitted, this would destabilise the entire system of property ownership, including in circumstances where NPG themselves depend upon legal certainty. I doubt their industry peers would thank them for it!
To make it absolutely clear, let’s take NPG’s case at its absolute highest, as they might hypothetically see it: Andrew is a misguided serial litigant, prone to “tenacious chihuahua mode”, while NPG are honourable bastions of civilisation acting entirely properly.
(I see it rather differently: Andrew’s persistence is an understandable response to repeated injustice, while NPG’s conduct gives me considerably less confidence in their institutional virtue.)
None of that matters. The law of the soil is indifferent to whether the parties standing upon it are noble or ignoble.
I don’t need to become a barrister to see there’s a fundamental problem, and nor do you. The failure mode isn’t one of law or even morality. It is elementary: the output of the court is not “real” because reality doesn’t affect the outcome.
There is no need for advanced logical visualisation skills drawn from computer science or computational law.
The defect is pre-legal: change the underlying reality, and the outcome stays the same.
Which brings us back to the judge and the finite cognition on offer. It’s easy to condemn him as insufficiently grasping the case, unconsciously favouring corporate interests, or simply being a bit dim in his reasoning. But those accusations are not merely unfair without evidence; they distract from the more basic problem.
How do we maintain the integrity of law as a system of adjudication that derives its authority from reality? How do we know when reality has ceased to constrain its outputs? And how might we apply ideas from quality control to detect and correct such failures?
The deeper problem is that law has particularly nasty structural failure modes.
Let’s assume for a moment that Andrew ceased litigating through financial and emotional exhaustion, and that NPG kept its cable on what is, on the face of it, presumptively Andrew’s land, without grant or payment. The underlying reality has not budged one inch because of NPG’s procedural victory.
But the resulting “legal reality” can now be leveraged into further injustice.
Andrew faces continuing uncertainty over whether he has a dispute with his neighbour, an industrial asset carrying lethal voltages beneath land he claims to own, and potential future liabilities involving his mortgage company, property insurer, or workmen undertaking drainage or repairs at the boundary.
One small upstream defect — one loss of contact with reality — propagates into an interconnected set of downstream consequences that become ever harder to dislodge.
And NPG haven’t erased their own liabilities or reputational risks. Should the eventual outcome be an adverse adjudication against them, if only to satisfy the needs of their own insurers and financial reporting obligations, then their technical defeat of Andrew was for nothing.
By resisting corrigibility, they have taken on the risk that they are also resisting reality. If so, they have picked a fight with truth itself.
That is never a winning strategic proposition in the long run. The price is paid in legitimacy, even if you appear to prevail tactically in the moment.
So the missing quality-control layer around the boundary of law isn’t just about protecting litigants-in-person from corporations with legal teams. It is a necessary prerequisite for the integrity of the system as a whole.
The real question is not whether NPG or Andrew is in the right, but why they are even engaged in conflict in the first place. The historic documents plus general law must already fix an answer. Defending legal uncertainty is an odd posture for a regulated utility.
My invitation is for you, dear reader, to consider the following question:
What quality-control mechanism should have made this conflict unnecessary?
Whether Andrew, NPG’s staff, their legal representatives, or the judge are “good” or “bad” people doesn’t matter. The result is functionally corrupted: without reality as the root, there is only exercise of power, not resolution through justice.
The outcome here fails a kind of type checking for law — like a spreadsheet expecting a number and being given a street address. While it can be appealed, that’s still treating it as “poor adjudication”, when the problem is several layers deeper.
I don’t have the answer as to how to fix the legal system.
But I can see the problem.
It’s not real, only legal.
And a legal “win” that isn’t real is no real win at all.
I am still offering readers the opportunity to chip in towards the cost of obtaining the expedited court transcript — which will feed into my own analysis and articles — as well as the appeal itself. This is also a public test of how many people actually care about quality control in the legal system.
NPG are watching, I expect. The number of people who contribute matters more than the amount raised. The ultimate goal isn’t to “win” but to restore the role of reality in justice. The transcript is simply raw data for developing AI quality-control tools to audit attachment to reality.
