Daniel sent us this one, and it's the kind of question that sounds simple until you actually try to answer it. He studied law back in Ireland, had full access to LexisNexis and the proper research tools as a student, and then, like most people who don't go into practice, that access just evaporated the day he graduated. Now he's in Israel, trying to research a tenancy issue, something about whether a landlord can refuse to let you clean the property before handing it back, and he's hitting the wall.
That wall is real. The legal system runs on precedent, but the library is locked behind a paywall that assumes you're billing clients at four hundred dollars an hour.
So the question is, if you're just a citizen trying to understand what the law actually says about your situation, are you just out of luck? And Daniel also raises this other thing, which is that what we see in published judgments, especially for stuff like tenancy disputes that mostly play out in small claims or magistrate courts, is just the tip of the iceberg. Most rulings never get published at all. So you're not just locked out of the library, you're locked out of a library that's missing most of the books anyway.
That's exactly the problem we're going to dig into today. The gap between "the law is public" and "actually finding the law is private." And specifically, what a tenant can do when they need to research their own case and they don't have a law firm budget.
Where do we even start with this? Because I think most people assume that if a court rules on something, that ruling is out there somewhere, and if you're determined enough, you can find it.
That assumption is wrong in two directions at once. First, even published judgments are often behind paywalls that make them practically inaccessible to non-lawyers. And second, most judgments, especially at the lower court level where tenancy disputes actually happen, are never published at all. So you're searching for something that may not even exist in any searchable form.
Let's start with the paywall problem, because that's the one people run into first. Daniel mentioned LexisNexis. What are we actually talking about in terms of cost?
LexisNexis and Westlaw are the two giants. For a solo practitioner, we're talking thousands of dollars a year. These are tools designed for law firms doing hundreds of searches a day and billing that cost back to clients. If you're a tenant trying to look up one question about cleaning access before returning keys, you're not going to drop three thousand dollars on a subscription.
That's the thing. It's not that the information is classified or secret. It's that the access model is built for a completely different user. It's like if the only way to read a book was to join a book club that costs five grand a year and assumes you're reading fifty books a week.
Daniel's experience is the normal lifecycle for almost everyone who studies law but doesn't practice. You get institutional access as a student, the university pays for the subscription, you have the full arsenal, and then you graduate and the door slams shut. Suddenly you're on the outside of a system you were just inside of.
That's a weird feeling. You know the tools exist, you know how to use them, you know the information is in there, and you just can't get to it anymore. It's not a knowledge problem, it's a gate problem.
And the gate isn't even malicious, it's just that nobody designed these systems for citizens. They were built for professionals, priced for professionals, and the idea of a non-lawyer wanting to do their own legal research just wasn't part of the business model.
Which brings us to the second problem Daniel raised, the iceberg. Even if you had full access to every paid database on the planet, you'd still be missing most of what actually happens in courts.
This is the part that I think would surprise most people. We assume that court judgments are public records and therefore publicly accessible. And technically they are public. You could theoretically walk into a courthouse and request a copy of a specific ruling. But practically, if it's not published in a database, it might as well not exist for research purposes.
Tenancy law is the perfect example of why this matters. These disputes are almost always in the lower courts, small claims, magistrate courts, whatever the jurisdiction calls them. The amounts of money are relatively small, the legal questions are fact-specific, and nobody is appealing these cases up to a supreme court where they'd get published and create visible precedent.
Daniel's question about whether a landlord can refuse to let you clean before returning home, that's not going to the Supreme Court of Israel. That's a small claims case. The judge hears it, makes a ruling, and if nobody appeals, that's it. There might be a written decision, there might not be. It might get published on the court's website, it might not.
You've got this whole body of law that actually governs how these disputes get resolved, and it's essentially invisible. Tenants and landlords are both arguing in the dark about what the law actually is in practice, because the practice isn't recorded anywhere they can see it.
That creates a feedback loop. The only cases that create visible precedent are the ones where someone had the resources to appeal. Which means the law develops based on who can afford to keep fighting, not based on what's actually common or what affects the most people.
The picture is pretty bleak. Paid databases you can't afford, and even if you could, they're missing most of the cases you'd actually want to read.
Here's the thing, and this is where we're going to spend most of today, it's not hopeless. There are free tools out there that are genuinely useful, and there are workarounds that most people don't know about. The picture is bad, but it's not nothing.
Let's get into what actually exists. What free tools are out there right now, and what can they actually do for someone in Daniel's position?
The landscape is a patchwork, and it varies enormously by country. But there are some genuine bright spots. The gold standard, in my opinion, is CanLII, the Canadian Legal Information Institute. Launched in two thousand, it now covers over three million Canadian court and tribunal decisions, all free to access. Not a limited preview, not a trial, not a stripped-down version. The full text of decisions, searchable, free, for anyone.
Three million decisions, free. That's the model that should exist everywhere.
And it's not just Canada. The UK has BAILII, the British and Irish Legal Information Institute, free since two thousand, covering case law and legislation for the UK and Ireland. Australia has AustLII. These are part of a global movement called the Free Access to Law Movement, and they're comprehensive for their jurisdictions.
If Daniel were in Canada or the UK or Australia, he'd be in pretty good shape.
He'd be in excellent shape. And even in the US, there's Google Scholar's case law database, which launched in two thousand nine. It covers US federal and state court opinions going back to seventeen fifty. Free, searchable, and surprisingly powerful once you learn how to use the search operators.
That's before the United States existed.
Colonial court opinions, yeah. It's an extraordinary resource. But, and this is a big but, it's limited to US law. If you're in Israel, or France, or Brazil, Google Scholar's case law database isn't going to help you. And even within the US, it has gaps for very recent cases and for lower court decisions.
What about Israel specifically? Daniel's here, he's dealing with Israeli tenancy law, what does he have access to?
The Israeli Courts website does publish Supreme Court and District Court judgments for free. And there's a decent amount there. But Magistrate Court judgments, which is where small claims and most tenancy disputes happen, are published inconsistently at best. There's no single comprehensive free database for Israeli case law at the lower court level.
He's in the worst of both worlds. No free comprehensive database, and even the paid ones don't have the lower court rulings he'd actually need.
That's the situation. And it's not unique to Israel. Most countries are in a similar position. The free legal information institute model has spread to about sixty countries, but coverage and quality vary enormously. Some are excellent, some are barely functional, and many countries have nothing at all.
Before we get deeper into what's missing and what you can do about it, I want to understand the mechanics a bit. When you're actually using one of these databases, what are you actually doing? How does legal research work when you're not a lawyer?
There are really two main approaches. One is keyword searching, which is what most people naturally do. You type in "landlord refuse entry cleaning" and see what comes up. And that can work, but it's crude. The better approach is something called citation chasing. You find one relevant case, even if it's not exactly on point, and then you follow the citations. What cases did this case cite? What cases have cited this case since? You're tracing a web of legal reasoning rather than just hoping your keywords match.
It's like the difference between googling a symptom and following a chain of medical studies that reference each other.
And the medical analogy is useful because in both cases, the amateur approach gets you something, but the professional approach gets you something much more targeted. The difference is that in medicine, the studies are mostly on PubMed and they're free. In law, the studies are behind paywalls.
The search itself has its own language. Boolean operators, all of that.
AND, OR, NOT, quotation marks for exact phrases, proximity operators that let you search for words within a certain number of words of each other. These are powerful tools, but they have a learning curve. And the free databases vary in how well they support advanced search syntax.
I feel like this is a good moment to actually walk through what a search looks like in practice. If Daniel sits down at his computer and tries to research this cleaning access question, what happens?
Let's take Google Scholar as an example, even though it's US law, just to show the process. He types in something like "landlord refuse access cleaning premises." What comes up is mostly appellate court decisions, mostly about much bigger issues, constructive eviction, habitability, security deposits. He might find a case or two that mentions cleaning in passing, but nothing that directly addresses his specific question.
Because his question is too small and too specific to have generated an appeal.
Now if he were in Canada and did the same search on CanLII, he'd have a better chance of finding lower court or tribunal decisions, because CanLII actively collects and publishes those. Ontario's Landlord and Tenant Board decisions, for example, are on CanLII. Not all of them, but a meaningful selection. So he might find a case where a landlord tried to deduct cleaning costs from a security deposit and the tribunal ruled on whether that was reasonable.
Even then, he's seeing a fraction of what's actually been decided. Which brings us back to the iceberg.
This is where the numbers get staggering. Studies across common law jurisdictions consistently estimate that eighty to ninety percent of lower court decisions are never published in any accessible database. Not behind a paywall, not hard to find — just not there at all.
Eighty to ninety percent. So for every ten disputes that get resolved in small claims or magistrate court, maybe one or two leave any kind of searchable trace.
That's probably generous. Think about what that means for someone in Daniel's position. He's trying to research whether there's precedent for a landlord refusing to let a tenant clean before handing back the keys. That's a hyper-specific fact pattern. The stakes are maybe a few hundred shekels in dispute. It's never going to be appealed to a court that publishes its decisions. So even if the exact same dispute has been heard fifty times in Israeli magistrate courts, there might be zero published records of any of them.
You're not just locked out of the library. The library doesn't even have the book you need. It was never written down.
Or it was written down in a ruling that sits in a physical file in a courthouse somewhere, accessible only if you know the case number and file a formal request. But you can't know the case number because you can't search for the case because it was never indexed.
It's a perfect catch twenty-two. You need to know what you're looking for to find it, but you can't know what you're looking for because it was never made findable.
This is the thing I think most non-lawyers don't realize about how the legal system actually works. We have this idea that the law is a body of rules and you just look up the rule that applies to your situation. But in a common law system, the law is mostly what judges have decided in previous cases. If those decisions aren't published, the law itself becomes invisible.
Which means the practical law, the law as it's actually applied to people like Daniel, is fundamentally different from the law as it appears in databases and textbooks.
The published law is the law of appeals, the law of big money disputes, the law of parties who could afford to keep fighting. The unpublished law, the iceberg beneath the surface, is where most people actually encounter the legal system. And it's a black box.
That asymmetry cuts both ways, but it cuts unevenly. A big landlord with a portfolio of properties, they've been through enough disputes that they have institutional knowledge. They know what magistrates tend to rule, even if those rulings aren't published. They don't need to look it up because they've lived it. The tenant doing this for the first time has none of that.
That's the structural inequality baked into unpublished judgments. It's not just an access to information problem, it's a knowledge asymmetry that systematically favors repeat players. Landlords, insurance companies, employers — anyone who goes to court regularly builds up a mental database of how things actually go, while the one-time litigant walks in blind.
Daniel's question about cleaning access, that's not really a question about cleaning access. It's a question about whether a citizen can meaningfully inform themselves about their legal position before making decisions.
And the answer is, it depends enormously on where you live and what level of court your issue falls into. But for most people in most places, the answer is that you can inform yourself partially. You can get a sense of the landscape. But you're almost certainly missing most of the relevant data.
Which is a pretty fundamental design failure for a system that's supposed to be public.
And it's worth noting that this isn't necessarily anyone's malicious intent. Courts are under-resourced, transcribing and publishing every small claims ruling would be expensive, there are genuine privacy concerns in some cases. But the cumulative effect is a legal information ecosystem that's public in name and private in practice.
Given all of that, the question becomes, what do you actually do? Daniel's sitting there with his tenancy issue, he knows the databases are incomplete, he knows most rulings are unpublished, he knows the system is stacked against him. Does he just give up?
No, and this is where it gets interesting. There are workarounds. They're imperfect, they're time-consuming, and they require a certain kind of stubbornness, but they exist. And the first one is the most obvious thing that almost nobody thinks to do — you can actually go to court. Not as a party, just as a spectator.
Most court proceedings are open to the public. You can walk into a small claims or magistrate court, sit in the gallery, and watch how judges actually handle these disputes.
That's such a deceptively simple idea that I think most people would never consider. You're not researching the law, you're just going to watch it happen.
And you learn things you'd never get from a database. How does this particular judge react to certain arguments? What kind of evidence do they find persuasive? How much leeway do they give to tenants who represent themselves versus landlords who bring a lawyer? That's all invisible in published decisions, even when those decisions exist.
It's almost ethnographic. You're studying the culture of the courtroom rather than the black letter law.
And for the kind of hyper-specific, low-stakes question Daniel is asking, that might actually be more useful than reading appellate decisions about completely different fact patterns. If you sit through twenty small claims tenancy cases, you'll start to see patterns in how judges think about cleaning, about security deposits, about what counts as reasonable. Even if none of those rulings are published anywhere.
The downside being, of course, that this requires time. A lot of it. And you might sit through three days of sessions and not hear a single case about cleaning access.
That's the trade-off. It's free, it's educational, but it's not efficient. Which brings me to the other workarounds. Freedom of information requests. In Israel, and in many other countries, you can file a formal request with the courts for specific unpublished judgments. You need to know what you're asking for, which is the catch, but if you have a general sense that a particular court handles a lot of tenancy cases, you can request rulings on specific topics.
Does that actually work?
The success rate varies wildly. Sometimes you get exactly what you asked for. Sometimes you get a letter saying the judgment doesn't exist in a form that can be released. Sometimes you get ignored. It's not a reliable system, but it's a tool that exists and most people don't know about it.
What about university legal clinics? Daniel mentioned he studied law in Ireland, but even without that connection, most universities with law schools have some kind of community legal clinic.
These clinics are often staffed by law students supervised by practicing lawyers, and they frequently have access to the paid databases. They're not going to do your research for you as a general rule, but if you're actually facing a legal issue and you qualify for their services, they can pull cases and give you advice grounded in real precedent.
It's not direct access, but it's a bridge. Someone with institutional access does the lookup for you.
In some jurisdictions, public law libraries also provide free access to LexisNexis or Westlaw terminals. You have to physically go there, you can't use them from home, but the access exists. Most people just don't know the library has it.
That pattern keeps coming up. The information isn't always inaccessible, it's just that finding out how to access it requires knowledge that most people don't have. It's gatekeeping by obscurity.
That's before we even get to the emerging AI tools. There's been a wave of legal tech startups trying to democratize this. Casetext, which was acquired by Thomson Reuters in twenty twenty-two for six hundred and fifty million dollars, built an AI legal research assistant. The promise is that AI can read and summarize case law faster than a human, potentially making legal research more accessible.
The catch is they're still mostly paid products.
Some have limited free tiers, but they're limited. You might get a few searches a month, or access to a subset of cases. The free tier is a taste, not a solution. The AI isn't the bottleneck, the access to the underlying database still is.
The AI democratization story is real but incomplete. It makes the research faster for people who already have access, but it doesn't necessarily solve the access problem itself.
Though I think that's changing. As these tools compete, some are starting to build on open access case law rather than proprietary databases. If you combine an AI research assistant with something like CanLII's free database, you start to get something powerful at zero cost. But we're not quite there for most jurisdictions.
For Israel specifically, none of these AI tools are going to help with unpublished magistrate court rulings anyway. The data doesn't exist to feed into the AI.
Which circles us right back to the iceberg. All the technological solutions in the world can't surface decisions that were never digitized, never transcribed, never published in the first place. The problem isn't just access to what exists, it's that most of what matters was never made to exist in accessible form.
Ontario's Landlord and Tenant Board is a perfect case study. They publish a selection of their decisions, roughly five to ten percent of total rulings. So a tenant researching a specific issue might find three relevant cases out of a thousand actual disputes.
Three out of a thousand. And those three might not even be the ones most similar to your situation. They're just the three someone decided were worth typing up.
The selection isn't random either. The board tends to publish decisions that clarify a point of law or address a novel issue. Routine disputes, the ones that make up the vast majority of cases, they don't get published because there's nothing legally interesting about them. But for a tenant, the routine is exactly what they need to understand. What normally happens in cases like mine.
The selection bias is built in at the publication stage, before you even get to the access problem. The visible law is systematically skewed toward the unusual.
That creates a distorted picture. If you're a tenant researching on CanLII and you find three Ontario cases about your issue, all of which went in the landlord's favor, you might conclude the law is against you. But what if the board actually rules for tenants on that issue eighty percent of the time, and those rulings just weren't published?
You'd be making decisions based on a completely inverted picture of reality. That's worse than having no information. At least with no information you know you're guessing.
And this is where the practical workarounds become not just helpful but essential. Court observation, FOI requests, legal clinics, pro bono services, these aren't just alternatives to database research. In many cases they're the only way to access the actual law as it's practiced.
The FOI route is interesting to me. Daniel's in Israel. If he wanted to file a freedom of information request for unpublished magistrate court rulings on tenancy cleaning disputes, what does that actually look like?
In Israel, you can file a request under the Freedom of Information Law with the Courts Administration. You'd need to be reasonably specific, you can't just say "give me everything about tenancy." But you could request all unpublished Magistrate Court rulings from the past three years in a particular district that address disputes about cleaning deductions from security deposits, for example.
Then you wait.
Then you wait. The law requires a response within thirty days, but extensions are common. And the response might be that no such rulings exist in a releasable form, or that the cost of retrieving them exceeds what the administration is obligated to provide for free, or that privacy redactions make the request impractical.
It's a tool, but it's not a reliable one. You can't build a legal strategy around hoping the FOI comes through.
Which is why, for someone in Daniel's position with a specific tenancy issue, I'd probably recommend starting with a legal clinic or pro bono service. They have the institutional knowledge, they've seen the unpublished patterns, and they can tell you whether your question even has a legal answer or whether it's just a negotiation position.
That's the thing we keep circling. Daniel's question about cleaning access before returning home, it may simply have no precedent anywhere. Not because the law doesn't address it, but because it's too fact-specific and too low-stakes to have generated a written ruling that anyone preserved.
That's almost certainly the case. The legal answer to "can my landlord refuse to let me clean" is probably "it depends on what your contract says, what's reasonable, and what the judge had for breakfast." That's not a satisfying answer, but it's an honest one. And knowing that your issue falls into that category is itself useful information. It tells you this is a negotiation, not a legal argument with a clear answer.
It also tells you that spending weeks trying to find precedent is probably not the best use of your time. Sometimes the most valuable legal research result is learning that there's nothing to find.
Given all those gaps and limitations, what can you actually do? Let me give you four concrete steps, and these are things someone can act on this week.
I'm ready.
Step one, start with the free databases that are comprehensive for their jurisdictions. If you're dealing with US law, Google Scholar's case law database is the obvious starting point. It's free, it covers federal and state opinions back to seventeen fifty, and the search is solid. For Canada, CanLII is the gold standard, three million decisions, all free. For the UK, BAILII. These aren't watered-down previews, they're the real thing.
For everyone else?
That's step two. Search for your country's legal information institute. The Free Access to Law Movement has spread to about sixty countries, and most people have no idea their country even has one. Just type your country name plus LII or free legal database into a search engine. You might be surprised. Some are excellent, some are bare-bones, but it's worth the thirty seconds to check.
Because the worst case is you find nothing, and you're exactly where you started.
Step three is where the mindset shifts. If you search these databases and find nothing relevant to your specific issue, that result itself is useful information. It tells you your problem is probably too fact-specific or too low-level to have generated published precedent.
The absence of results isn't failure, it's intelligence.
It's intelligence about what kind of dispute you're actually in. A dispute with clear precedent is one kind of fight. A dispute where the judge is going to decide based on equity and reasonableness because nobody has ever appealed this exact question, that's a different kind of fight. You prepare differently, you argue differently, you negotiate differently.
Go to court. Small claims sessions are open to the public. Sit in the gallery and watch. You'll see how judges actually rule on issues like yours, even if those rulings are never published anywhere. You'll see what arguments land and what arguments annoy the judge. You'll see how tenants who represent themselves are treated versus tenants who bring a lawyer. None of that is in any database.
It's the ultimate workaround for the iceberg problem. If the rulings aren't written down, go watch them happen in real time.
You only need to do it once or twice to get a feel for the room. You don't need to become a courthouse regular. One morning observing small claims tenancy cases will teach you more about how your issue is likely to be handled than a week of frustrated database searches.
The meta-lesson in all of this, and I think this is where Daniel's question really lands, is that legal research for non-lawyers is possible but it requires accepting that you're working with incomplete information. The goal isn't to find the answer with a capital A. It's to understand the range of possible outcomes well enough to make informed decisions.
Lawyers have the same problem, by the way, they just have better tools for managing the uncertainty. But the fundamental limitation, that most lower court rulings are invisible, that affects everyone. The difference is lawyers know it and factor it into their advice. Citizens doing their own research need to know it too.
You're not failing if you can't find a case exactly on point. You're just seeing the same incomplete picture everyone sees, and the skill is knowing what to do with that incompleteness.
The four steps give you a framework. Check the free databases first. Check if your country has a legal information institute. If you find nothing, treat that as information about the nature of your dispute. And if you really need to understand how these cases go, go watch them. None of these steps requires a law degree or a subscription. They just require knowing they exist.
Here's the thing I keep coming back to. We've got AI legal tools getting smarter every month, some of them edging toward free tiers, and on the surface that looks like democratization. But I wonder if we're just trading one gate for another.
What do you mean?
The paywall gate gets replaced by a prompt engineering gate. If the free AI tool can find the case but only if you know how to ask it exactly the right way, then access hasn't really been solved. It's just been moved.
That's a uncomfortable question. And I think the answer is probably both. The iceberg will shrink in some ways, because AI can process and surface unpublished decisions if they ever get digitized. But it'll also create a new skill divide. The person who knows how to construct a legal research prompt, who understands Boolean logic and citation chains and can guide the AI toward relevant doctrine, they'll get radically different results than someone typing "can my landlord do this" into a chatbot.
The person who knows how to prompt is probably the person who already had some legal training. So the gap reproduces itself.
Unless the tools get good enough that natural language actually works. And we're not there yet. The AI might give you an answer that sounds authoritative, but if you can't verify what it's drawing from, if you don't know whether it's citing a Supreme Court ruling or hallucinating a case that doesn't exist, you're worse off than when you started.
Confident misinformation is worse than acknowledged ignorance.
And that's the real risk with AI legal tools marketed to non-lawyers. The interface feels accessible, but the underlying epistemology, the question of "why should I believe this answer," that's still a black box for most users.
Which brings me to the thing I think this whole conversation has really been about. The law is supposed to be public. That's not a nice-to-have, that's foundational. The idea that citizens can know the rules they're governed by is what separates a legal system from an arbitrary one. And we've built this elaborate information ecosystem where the law is technically public but practically private.
That's the design failure. Not a conspiracy, not malice, just decades of systems built for professionals, under-resourced courts, publication decisions made by administrators who were never asked to think about the tenant trying to research their own case. The cumulative effect is a legal information ecosystem that fails its most basic promise.
Citizens are hacking their way around it. Court observation, FOI requests, legal clinics, free databases stitched together from whatever's available. It's slow, it's incomplete, it's frustrating, but people are doing it because the alternative is walking in blind.
That stubbornness, that refusal to accept that the law is only for lawyers, that's actually what keeps the system honest. Every tenant who sits in a small claims gallery and learns how the room works, every citizen who files an FOI for an unpublished ruling, they're not just helping their own case. They're asserting that the law belongs to everyone.
Daniel's question about cleaning access, it turned out to be a much bigger question about whether a citizen can meaningfully know their own legal position. And the answer is, it's hard, it's incomplete, but it's not impossible. You just have to know that the library is missing most of its books and work around that.
Now, Hilbert's daily fun fact.
Hilbert: In the nineteen sixties, Soviet mathematicians working near Lake Baikal briefly adopted a notation system that used triangles to represent exponents, because their typewriters lacked superscript keys. The system was abandoned after a visiting researcher from Moscow pointed out that equations involving nested exponents became completely illegible, essentially turning every formula into a field of tiny triangles.
A field of tiny triangles. That's what they were trying to avoid.
I have so many questions about what that looked like on paper.
So here's the thought I want to leave people with. As these AI tools improve and the free database movement spreads, the question isn't just whether the iceberg shrinks. It's whether we end up with a system where the law is public, or a system where access is technically free but practically still gated behind skills most people don't have. And I don't think we know the answer yet.
What we do know is that right now, today, the tools exist to get started. They're imperfect, they're incomplete, but they're real. And using them is itself a kind of citizenship.
This has been My Weird Prompts. Our producer is Hilbert Flumingtop. If you found this useful, leave us a review wherever you listen, it helps more people find the show. We'll be back next week with whatever Daniel sends us.
Until then, go watch a court case. It's free and it's fascinating.
