Daniel sent us this one. He's been thinking about the causes of action episode we did — the idea that not everything on the statute books gives you a right to sue — and he wants to go deeper into the other big confounding factor: precedent. Specifically, two questions. One, how does the balance between case law and statute law actually differ across common law countries? And two, what happens when judicial interpretation of a statute gets so far from the original text that the two are basically in opposition? He wants concrete examples from the UK, the US, and Israel. And honestly, the Israel piece of this is perfectly timed.
It really is. The Knesset spent much of twenty twenty-four and twenty twenty-five wrestling with exactly this tension — whether to amend the nineteen seventy Tenants' Protection Law or replace it entirely. And the single biggest argument against starting fresh was: if we tear up this statute, we lose more than fifty years of precedent that has defined what terms like "protected tenant" and "reasonable rent" actually mean in practice. You'd be throwing away thousands of judicial decisions that everyone has relied on for decades.
The thing Daniel's pointing at is genuinely a live question right now. The law you read on the books is not the law that applies to you. The gap between statute and precedent can be enormous. And sometimes the two are actively fighting each other.
That's what makes precedent the most interesting part of any legal system. Statute law gives you the skeleton — the broad architecture. But precedent fills in everything else. It's the connective tissue, the muscle, the fine motor control. The problem is, sometimes the muscle starts pulling the skeleton in a direction it was never designed to go.
Which is a mildly disturbing image, but I'll allow it.
The rental law debate in Israel crystallizes this perfectly. You've got one camp saying: keep the old statutory language, preserve the precedent, just update the numbers. And you've got the other camp saying: the precedent has so thoroughly warped the original text that only a clean legislative slate makes sense. That's not an obscure legal philosophy debate — that's a practical question about whether millions of tenants and landlords know what their rights actually are.
Let's step back and frame this properly. What exactly is the relationship between a statute and the cases that interpret it, and why does that relationship look so different depending on which common law country you're standing in?
In any common law system, you have two sources of law operating simultaneously. You have statutes passed by a legislature, and you have judicial decisions that interpret those statutes and also develop purely judge-made common law where no statute exists. The interaction between these two sources is what Daniel's getting at, and it's messier than most people realize.
The messiness varies by jurisdiction. So where do the big differences show up?
The UK, the US, and Israel give us three really distinct flavors. The UK has the strongest culture of stare decisis in the common law world. Their Supreme Court only formally claimed the power to overrule its own past decisions in nineteen sixty-six, and even then it's exercised that power sparingly. The US Supreme Court, by contrast, overrules itself regularly — sometimes dramatically, as we saw with the death of Chevron deference last year. And Israel sits in this fascinating middle ground, partly because its legal system is a hybrid of Ottoman law, British mandate law, and Knesset legislation, with precedent stickiness that is unusually high in some areas and surprisingly flexible in others.
Before we get into the mechanisms, I want to sit with that rental law example for a second, because it's the perfect illustration of what's at stake. The Tenants' Protection Law was passed in nineteen seventy as a temporary measure after the Six Day War. It was supposed to be short-term. It's still here. It's been extended again and again, and around it has grown this massive body of precedent — court decisions interpreting what counts as a protected tenancy, what counts as reasonable rent increases, what happens when a protected tenant dies and their children want to stay.
That's the "sedimentation" process. Every new case adds another layer of interpretation. After fifty-plus years, you're not just reading the statute anymore. You're reading the statute through fifty years of judicial gloss. A phrase like "protected tenant" in the original nineteen seventy text might have been a few lines. Today, that phrase carries the weight of hundreds of decisions defining its boundaries. If the Knesset writes a new rental law from scratch, all of that falls away. Every boundary, every definition, every settled expectation — gone. The courts would have to start building it all over again, and nobody knows what the new building would look like.
Which is why the pro-amendment camp was essentially arguing: we'd rather keep imperfect statutory language than lose the interpretive certainty that's grown up around it.
That sounds bizarre if you think of law as just whatever the statute says. Why would you keep bad language on purpose? But if you understand that the real law is the precedent, not the text, it makes perfect sense. The text is just the anchor point. The precedent is what people actually rely on when they make decisions — landlords deciding whether they can raise rent, tenants deciding whether they have a right to stay, courts deciding who wins.
That's the Israel piece — a live legislative debate where the precedent cost of rewriting a statute became an explicit, quantified factor. But before we get to what actually happened with that bill, I want to understand the mechanisms. How does precedent actually build up around a statute, and why do different countries have such different relationships with their own precedents?
The first thing to understand is the sedimentation process. A statute gets passed. Someone sues, claiming the statute means X. A court rules — yes, it means X, but with these caveats. Someone else sues, claiming the statute also means Y. Another court rules — no, it doesn't mean Y, and here's why X was the right reading all along, and by the way here's a nuance about X that the first court didn't consider. Each decision cites the earlier ones. After a decade or two, you've got what lawyers actually read — which is not the statute, but the leading case that interprets the statute.
The shadow statute phenomenon.
That's exactly what to call it. In practice, the statute becomes almost ceremonial. You cite it in your brief because you have to, but the real argument is about what the precedent says. If you're a property lawyer in the UK, you don't spend your time reading the Law of Property Act nineteen twenty-five. You read the cases that have interpreted it over the last century. The statute is the starting point, but the precedent is the destination.
That's the sedimentation mechanism. What makes it different across countries?
Two big variables. First, how strictly courts are bound by their own past decisions — horizontal precedent. Second, the philosophical stance the judiciary takes toward statutory interpretation — are they trying to find the original meaning of the text, or are they trying to adapt it to contemporary circumstances?
Let's take horizontal precedent first, because the differences there are stark.
In the UK, the Supreme Court has operated under the Practice Statement of nineteen sixty-six. Before that, the House of Lords considered itself absolutely bound by its own previous decisions. The Practice Statement said: we can overrule ourselves, but we'll do it rarely, and only when the previous decision is causing real problems. And they've meant it — the UK Supreme Court overrules itself maybe once or twice a decade. The US Supreme Court overrules itself multiple times per term. Chevron last June is a perfect example — they wiped out forty years of precedent in a single ruling, and it wasn't even that unusual for them.
Israel is the most complicated of the three. The Supreme Court sitting as the High Court of Justice can strike down Knesset legislation that violates the Basic Laws, Israel's quasi-constitutional framework. That power was established in the nineteen ninety-five Bank Mizrahi decision, and it means the Israeli Supreme Court has a constitutional override mechanism that neither the UK nor the US has in quite the same form. The UK has parliamentary sovereignty — courts can't strike down primary legislation at all. The US has judicial review based on a written constitution. Israel's system is somewhere in between — the Basic Laws function like a constitution, but they were passed by the Knesset as ordinary legislation, and the court's power to enforce them is itself a product of precedent.
The Israeli Supreme Court built its own constitutional review power through precedent, without a formal constitution. That's a dramatic example of precedent creating something the original statutes didn't explicitly provide for.
It's one of the most dramatic examples in any common law system. The Knesset passed the Basic Laws in the early nineties. Did the members who voted for them think they were creating full judicial review? That's heavily debated. But the court, through its own precedent, said: yes, that's what these laws mean, and we have the power to enforce them against the Knesset itself.
Which brings us to the second variable — interpretive philosophy.
Here the UK is especially interesting because of the declaratory theory of precedent. This is the idea — widely acknowledged to be a fiction — that judges don't make law, they merely declare what the law has always been. When a court interprets a statute in a new way, under the declaratory theory, it's not changing the law — it's discovering the true meaning that was always there, hidden in the text.
Which is obviously nonsense, but it's a useful fiction if you're a court that doesn't want to be seen as legislating from the bench.
The UK has maintained this fiction more stubbornly than the US. American courts, since the legal realist movement of the early twentieth century, have been more open about the fact that judges make law. Holmes, Cardozo — they all acknowledged it explicitly. But it also creates a different dynamic — because if everyone acknowledges that judges make law, then overruling precedent is less of a big deal. You're just saying the previous judges got it wrong. In the UK, overruling precedent is harder precisely because the fiction says the previous judges were just declaring the law — so how could they have been wrong?
Israel inherited the British declaratory theory through the mandate period, but then developed its own approach.
Israeli courts still use some of the language of the declaratory theory. But in practice, the Israeli Supreme Court under Aharon Barak in the nineteen nineties and two thousands embraced a very expansive theory of purposive interpretation. The idea is that courts should interpret statutes in light of their purpose, not just their text, and that purpose can evolve over time. That gives Israeli courts enormous room to develop precedent that diverges from the original statutory language.
We've got three different approaches. The UK clings to the fiction that precedent and statute are always aligned. The US admits they can diverge and overrules precedent more freely. And Israel has built a system where the court's interpretive power is so expansive that precedent can effectively rewrite statutes in the name of purpose.
Each approach has its own failure modes. The UK's fiction means that when precedent really has diverged from statute, the system has a hard time admitting it and fixing it. The US approach means you get circuit splits — the same federal statute means different things in the Ninth Circuit than it does in the Fifth, and until the Supreme Court resolves the split, you literally have different law in different parts of the country. Israel's approach means the Knesset sometimes feels like it's in a tug-of-war with the Supreme Court over who actually gets to decide what the law is.
Let's make this concrete. Give me an example from each jurisdiction where precedent clearly diverged from the original statutory text.
The UK example taught in every law school is R v R from nineteen ninety-one. The House of Lords had to interpret the Sexual Offences Act nineteen fifty-six, which criminalized rape. The statute didn't explicitly address marital rape — for centuries, there had been a common law rule that a husband couldn't rape his wife, based on the idea that marriage implied consent. By nineteen ninety-one, that rule was obviously outrageous. So the Law Lords said: we are interpreting the nineteen fifty-six Act, and we find that the marital rape exemption no longer exists. But the nineteen fifty-six Act didn't abolish the exemption. Parliament had never voted on it. The Lords abolished it by interpreting a statute that was silent on the question, in a way clearly contrary to what the nineteen fifty-six Parliament would have intended.
Everyone accepts that was the right result, but the mechanism is worth examining. The court effectively amended the statute without Parliament.
And that's the declaratory theory doing heavy lifting. The Lords said they were just declaring what the law always was. Nobody seriously believes that. What they were doing was updating the law to match contemporary values, using statutory interpretation as the vehicle. The statute didn't change. The meaning did.
The US example?
Bostock v Clayton County in twenty twenty. Title Seven of the Civil Rights Act of nineteen sixty-four prohibits employment discrimination "because of sex." For decades, courts interpreted "sex" to mean discrimination against someone because they were a man or a woman — not discrimination based on sexual orientation or gender identity. Then the Supreme Court, in a six-to-three decision written by Neil Gorsuch, said: actually, discrimination against someone for being gay or transgender is discrimination because of sex, because you can't define sexual orientation or gender identity without reference to sex.
The argument from the dissent was: the nineteen sixty-four Congress was definitely not thinking about sexual orientation or gender identity when it wrote those words.
Gorsuch's response was essentially: it doesn't matter what Congress was thinking. The text says what it says, and the logical implication is that if you fire a man for being attracted to men, but you wouldn't fire a woman for being attracted to men, you're discriminating because of sex. That's a purely textualist argument that produced a result the drafters would not have anticipated. And Congress has never amended Title Seven to codify or reject that interpretation — so the precedent now is the law, even though the statutory text hasn't changed.
The Israel example brings us back to the rental law.
When the Knesset's housing committee took up reform in twenty twenty-four, they faced a dilemma. If they replaced the law entirely, all that precedent becomes irrelevant — the new statute would use new language, and the old interpretations wouldn't apply. But if they just amended the existing law, they'd preserve the precedent but be stuck with statutory language drafted for a temporary post-war housing crisis that doesn't reflect the modern rental market.
What did they do?
The compromise was a hybrid approach. They kept the core structure of the nineteen seventy law — preserving the precedent — but added a new section that explicitly overrode certain court interpretations that the Knesset disagreed with. It's a rare example of the legislature saying to the courts: we see what you've done with our statute over the last fifty years, and on these specific points, we're taking it back.
That's fascinating — a legislative override clause that surgically removes specific precedents while leaving the rest of the interpretive edifice intact.
It's exactly the kind of thing that makes comparative law so interesting. The UK has done something similar — the Protection of Freedoms Act twenty twelve was Parliament's response to the European Court of Human Rights ruling in Gillan v UK, which found that Section forty-four stop-and-search powers under the Terrorism Act two thousand violated Article eight privacy rights. Parliament didn't just grumble about judicial overreach. It passed a new law that explicitly narrowed the statutory language to match the judicial interpretation. The statute caught up to the precedent.
Which is the healthy version of this dynamic. The unhealthy version is when the statute and the precedent just sit there in tension, each pulling in a different direction, and nobody resolves it.
That happens more often than you'd think, because resolving it requires legislative action, and legislatures are busy. Or, in some cases, they're deliberately avoiding the issue because the precedent has created a politically convenient outcome that nobody wants to take responsibility for codifying.
To pull this back to Daniel's questions — the balance between case law and statute law differs by country along several axes. How strictly courts bind themselves, how honest they are about making law versus declaring it, and how much interpretive flexibility they claim. And when statute and precedent diverge, you get everything from quiet judicial updating of outdated laws to explicit legislative override clauses to decades of unresolved tension.
The Israel rental law is the perfect case study because it shows all of these dynamics compressed into a single legislative debate. You've got the sedimentation of fifty years of precedent. You've got the declaratory fiction — Israeli courts saying they're just interpreting the nineteen seventy text. You've got the legislature pushing back with an override mechanism. And you've got the practical question that affects real people: if we change the words, do we destroy the certainty that people have relied on for decades?
Let's go deeper into the knock-on effect when the accumulated interpretation starts pulling away from the original text.
The sedimentation process — think of a river delta. Each case is a layer of silt. One case says "reasonable rent" means rent comparable to similar properties in the same neighborhood. The next case says: actually, it also has to account for the tenant's ability to pay. A third case adds: and the landlord's maintenance costs matter. After fifty years, "reasonable rent" isn't a phrase anymore — it's a multi-factor test that no single judge ever designed. It just accumulated.
Nobody along the way said, wait, are we still anchored to what the nineteen seventy Knesset actually voted on?
That's the thing about sedimentation — it doesn't require anyone to step back and check. Each judge is just deciding the case in front of them, citing the most recent precedent, adding their own small gloss. The drift is incremental. No single decision feels revolutionary. But stack up fifty years of increments and you've traveled a long way from the original text.
Once precedent gets thick enough, it changes how legislators behave — not just how courts behave.
The legislative inertia problem. If you're a Knesset member looking at the Tenants' Protection Law, you know that rewriting it means resetting fifty years of interpretive clock. Every landlord, every tenant, every judge — they all have to learn the new language from scratch and wait for cases to percolate up and define what the new terms mean. That's years of uncertainty. So you're strongly incentivized to just keep the old statute, no matter how creaky the language has become. The precedent is too valuable to lose.
Which is a perverse incentive. You're preserving statutory language you know is bad because the judicial gloss has made it usable.
It's not just Israel. The UK's Law of Property Act nineteen twenty-five is still on the books with much of its original language, and a huge amount of modern property law is just precedent interpreting that century-old text. Nobody wants to open that Pandora's box. The US has the same dynamic with the Administrative Procedure Act of nineteen forty-six — the statutory text is largely unchanged, but the law of administrative procedure is almost entirely judge-made at this point.
The shadow statute isn't a bug, it's the system working exactly as designed, just with the embarrassing parts papered over.
It's worth being precise about what "shadow statute" means in practice. If you're a lawyer preparing a case, you don't start by reading the statute. You start with the leading case — the one that synthesized all the earlier decisions into a workable framework. The statute is something you check at the end, almost as a formality, to make sure the precedent hasn't obviously wandered off the reservation.
Which brings us to another knock-on effect — strategic behavior by litigants. If you know the precedent varies by jurisdiction, you choose your battlefield accordingly.
The US is the extreme case here. If you're bringing a nationwide class action and you want favorable precedent on securities fraud, you file in the Second Circuit — New York, generally favorable to plaintiffs. If you're a defendant trying to kill a class action, you fight to get the case heard in the Fifth or Eleventh Circuit. Forum shopping isn't a loophole — it's a core litigation strategy, entirely driven by the patchwork of circuit precedent.
In the UK, where there's only one jurisdiction, the strategy takes a different form.
The UK version is the leapfrog appeal. If a case raises a point of statutory interpretation where the Court of Appeal has already settled on an interpretation that one side thinks is wrong, that side can apply to skip the Court of Appeal entirely and go straight to the Supreme Court. The Court of Appeal is bound by its own precedent, so there's no point arguing there — you need the Supreme Court to overrule it. The leapfrog mechanism is an explicit admission that horizontal precedent can create dead ends.
Israel takes yet another form. Because the Supreme Court sits as both an appellate court and a constitutional court, litigants can sometimes choose which door to knock on. Do you frame your case as an ordinary statutory interpretation appeal, where precedent is sticky? Or do you frame it as a constitutional challenge under the Basic Laws, where the court has shown more willingness to break new ground? The choice of framing can determine which body of precedent applies.
Which is exactly the kind of strategic ambiguity that makes the rental law compromise so revealing. The Knesset looked at this whole ecosystem — the sedimentation, the legislative inertia, the strategic behavior — and said: we need a mechanism that preserves the good parts of precedent while giving us a scalpel for the bad parts.
The override clause they developed is worth describing in detail. The bill kept the core statutory language of the nineteen seventy law intact — preserving all the precedent built up around terms like "protected tenant" and "key money." But it added a new section that explicitly listed several court interpretations and said, in effect: these specific holdings are no longer good law. Not because the courts were wrong on the law as it stood, but because the Knesset is changing the policy going forward.
It's a legislative reset on specific points, without the collateral damage of a full statutory rewrite. And it mirrors what the UK did with the Protection of Freedoms Act after Gillan. The European Court of Human Rights said Section forty-four stop-and-search powers were too broad. Parliament responded by narrowing the statute to match the judicial interpretation — but also by adding new safeguards that went beyond what the court had required. In both cases, the legislature treated the precedent as a starting point for reform, not an obstacle to it.
Which is a healthier dynamic than what we often see in the US, where Congress just lets precedent sit there, sometimes for decades, neither codifying nor rejecting it.
Title Seven is the classic example. The Supreme Court in nineteen eighty-six said sexual harassment is sex discrimination under the nineteen sixty-four Civil Rights Act. The Court in twenty twenty said discrimination based on sexual orientation and gender identity is also sex discrimination. Both decisions were controversial. Both could have been overruled by Congress with a simple statutory amendment. Congress did nothing. So the precedent stands, and the statutory text — "because of sex" — carries a meaning that no one in nineteen sixty-four would have recognized.
That's not necessarily a criticism. Sometimes legislative inaction is tacit acceptance. The point is that the precedent, not the statute, is doing all the work.
Which connects to the comparative insight Daniel was asking about — the different drafting philosophies. The UK's Human Rights Act nineteen ninety-eight was deliberately written in broad, principles-based language, explicitly delegating interpretive authority to the judiciary. US statutes, by contrast, are famously hyper-detailed — the Affordable Care Act ran to thousands of pages — precisely because Congress doesn't trust courts to fill in the gaps.
The UK writes statutes expecting precedent to do the heavy lifting. The US writes statutes trying to preempt precedent from having any room to maneuver.
Both approaches create their own precedent dynamics. The UK's broad drafting gives courts enormous interpretive flexibility — precedent can evolve dramatically without anyone amending the statute. The US's hyper-detailed drafting means that when courts do find ambiguity, the precedent battle is over tiny phrases, and the stakes feel artificially high because so much turns on a single word.
Israel, as usual, is somewhere in between — the Knesset drafts with more specificity than the UK but less than the US, and the courts have claimed more interpretive flexibility than either. Which is why the override clause mechanism is such an Israeli innovation. It's a response to a system where the courts have been unusually assertive and the legislature is now pushing back with unusual precision.
What do we do with all this? Let's make it practical. If you're someone who needs to predict how a legal question will actually be resolved, the single most useful thing I can tell you is: don't read the statute. Read the leading cases.
That sounds like terrible advice if you haven't been following the last twenty minutes, but it's how the system works. The statute is the starting point. The precedent is the destination. If you read the Tenants' Protection Law as written in nineteen seventy, you will not understand Israeli rental law. You have to read the cases that have interpreted it. Same with Title Seven in the US — "because of sex" is four words, but the law of employment discrimination is thousands of pages of judicial opinions.
The second thing is for anyone thinking about legal reform. The precedent cost is real, and it should be quantified. The Knesset didn't just wave their hands and say "we'd lose some precedent." They had specific debates about specific terms — "protected tenant," "key money," "reasonable rent" — and what it would cost, in years of litigation and uncertainty, to redefine each one from scratch. Preserving imperfect statutory language can be a perfectly rational choice if the precedent built on it is valuable enough.
The third insight is for anyone following legal news. When you see a headline that says "Supreme Court interprets statute to mean X," ask yourself: is this interpretation filling in fine lines, or is it redrawing the brushstrokes? The answer tells you whether the court is doing routine interpretive work or something more significant. Bostock — redrawing brushstrokes. Chevron's fall in Loper Bright — redrawing brushstrokes. Most of what courts do day to day is the fine lines. But when you spot the brushstroke moments, that's when the relationship between statute and precedent is being renegotiated in real time.
The broader point Daniel's question points toward is this: the gap between statute and precedent is not a bug of common law systems. It's a feature. It's what allows law to adapt without requiring the legislature to vote on every small adjustment. But the adaptation has costs — unpredictability, strategic behavior by litigants, and the occasional need for a dramatic legislative override when the gap gets too wide.
The override clause in the Israeli rental law amendment is a perfect example of the system working as intended. Fifty years of precedent built up. The Knesset decided some of it had gone too far. They didn't burn the whole thing down — they used a scalpel. That's the mature version of the statute-precedent relationship. The immature version is what we saw with Chevron — forty years of settled practice, then a judicial wrecking ball with no legislative input at all.
The next time someone tells you what the law is, the useful follow-up is: according to which — the statute or the precedent? Because they might be saying very different things.
One last question to leave you with. As AI-assisted legal research gets better at surfacing every precedent that has ever interpreted a statute, does precedent become even stickier? Or does the ability to instantly find conflicting interpretations accelerate the push for statutory clarity?
I think it cuts both ways. On one hand, when you can pull up every case that has ever interpreted "reasonable rent" in seconds, the sedimentation process accelerates. Judges see the full weight of what's been built, and they're even less likely to disturb it. On the other hand, when you can instantly find the five cases where courts quietly diverged from each other, the contradictions become harder to ignore. The fiction that precedent is a single coherent line gets harder to maintain.
The declaratory theory doesn't survive a search engine that returns conflicting results in point-three seconds.
There's a deeper shift happening too — the rise of purposive interpretation globally. The UK post-Human Rights Act, the EU-influenced jurisdictions, even Israel under the Barak court — they've all moved toward interpreting statutes in light of their purpose rather than their plain text. That's a deliberate invitation for precedent to evolve. If courts are told to ask what a statute is for, not just what it says, then the text really is just a starting point. The divergence between statute and precedent isn't a side effect — it's the design.
Which means the gap Daniel's asking about is likely to get wider, not narrower, in most common law systems. The tools for finding precedent are getting better at the same moment that interpretive philosophy is giving courts more room to develop it.
The override mechanism — the Israeli approach, the UK Protection of Freedoms model — that might become more common as a result. If precedent is going to drift further from text, legislatures need better tools for pulling it back when the drift goes too far. The scalpel, not the wrecking ball.
Here's where I'll land. The next time you read a law, remember — you're reading a draft. The final version is being written in courtrooms, one case at a time. And the authors of that final version may never have stood for election.
Now: Hilbert's daily fun fact.
Hilbert: In nineteen twenty-three, a Danish carpenter's manual published for Greenlandic tradesmen included detailed instructions for crafting a specialized ice chisel with a handle angled at precisely one hundred twelve degrees — the author insisted this was the only angle that prevented the user's knuckles from freezing to the tool during prolonged use in sub-zero temperatures.
I have so many questions about the field-testing process for that.
This has been My Weird Prompts. If you enjoyed this episode, tell someone who thinks the law is whatever they read on a government website. We're at my weird prompts dot com. I'm Herman Poppleberry.
I'm Corn. We'll be back with another one soon.
