48 comments

[ 4.8 ms ] story [ 135 ms ] thread
I think Thiel misses a key point: getting the right answer isn't necessarily as important as the appearance of justice. Human beings are fundamentally emotional creatures, especially in situations when they feel they have been wronged or are accused of wronging someone else. The whole process of a trial with a human judge and a human jury doesn't just resolve a dispute, but it serves as an emotional palliative allowing people to get beyond the dispute.

And that is, of course, one of the key functions of the legal system as a system of dispute resolution: allowing society to continue to function by helping people get over disputes.

I think this is a very good point, and I totally agree about the importance of emotion and the appearance of justice in our legal system. But I'm not sure it follows that the appearance of justice is at odds with getting things right. A better system would get things right and feel like a fair system to the participants.

My personal opinion is that you can get both with a more transparent system. In theory, a transparent system would drive consistency, and consistent outcomes would hopefully converge around correct outcomes.

I think the danger is when it comes to automation/rigid rules. People want to feel like their problems are unique and being considered in all their unique complexity. That makes achieving transparency or automation hard, because a system of hard and fast rules is more amenable to transparency than a system of fuzzy fact-specific inquiries.
But at what cost? There is a tremendous opportunity for pretty-good, bargain-basement, automated legal services.
Is there?

There are two kinds of clients in the legal field: rich ones and poor ones. Rich clients don't seek the cheapest legal services in the way they don't look for the cheapest office space or the cheapest office furniture. No board member wants some guy in India poring through his e-mails to do an investigation or diligence for a transaction. For rich clients, saving money and hiring "pretty good" isn't such a great thing if their opposing party spends more money and hires "pretty great."

Poor clients are different. Poor clients do want the lowest possible price for the given result. But poor clients tend to have problems that are hard to outsource. How is an attorney sitting in India supposed to represent you in your divorce, or help you get out of a traffic ticket?

Correct - and that is why I think that automation will not get us all the way there. (At least not in the 20-50 year range; I have no idea what the world will look like in 1000 years.)

And I think your point suggests that lawyers will continue to be valuable for many years to come, as well. There are so many rules that can be applied in a given case - including seemingly contradictory ones - that there is room for reasonable disagreement about which rule should be applied. A skillful lawyer can make a compelling (but not obvious) argument that rule A should be applied and not rule B.

Thiel spends half of the article banging at that exact point - the current system is stuck at appeasing emotions and appearances resulting in a grossly unjust system, resulting in evil deeds done in the name of 'justice'. This should be fixed.
The legal industry is perfectly suited to be utterly destroyed by startups. It's pure information work, it can be done from offshore, it is generally disliked by the public, and as a whole is pretty much entirely technologically inept.

Certification regimes present a barrier to full automation, but with so many young and unemployed lawyers, you could easily get around that as well. Just do most of the stuff on the computer and let Mr. cheap-young-JD sign off on the last mile work on his phone. Each contract he goes through knocks another $100 off his student loan bill. Use the law schools' oversupply to break the legal industry and bring cheap legal services to all.

The main threat will be regulation. Putting lawyers out of work will mean they will try to rewrite the law to ban you. Depending on how you structure it, you might want to ally with a few firms in each vertical to rain fire upon their competitors. As long as you divide and conquer, and some lawyers are profiting from the new regime, it will make it much more difficult for the ABA to achieve consensus on the imminent threat your startup undoubtedly poses to the health of the republic.

EDIT: Even more interesting...as a dotcom you may also be able to market your services in a way that normal lawyers cannot.

http://en.wikipedia.org/wiki/Legal_advertising#In_the_United...

So I'm a lawyer and you should take my opinion with a grain of salt, but I think you've misread the market.

There has been an oversupply of legal gradutes for decades now. Today, half of law graduates don't get a job, and 90% can't get a job at a big firm. Yet, big firms are still paying $160k+bonus to people fresh out of school, because that's what they have to pay to get kids out of the top 15-20 schools or the tippy-top kids from the lower-ranked schools. Why do they continue to do this when lowering their standards for pedigree could cut their costs massively?

You can say that it's because law firms are stupid and invested in the status quo, but corporate clients surely aren't. Why do firms persist in putting so much stock in hiring Harvard/Yale/etc graduates, and why do clients keep hiring such firms? If there was really an opening in the market you could drive a truck through, what has stopped anyone from taking advantage of it thus far?

All of this can be called "legal services." But the extremes, you two are essentially talking about different markets.

Corporate clients will continue to hire the best lawyers and pay high fees when and because the company is on the line.

Most individuals, on the other hand, do not need a Yale Law grad billing $900 an hour to set up a living trust or LLC.

Obviously there's a whole spectrum in between. Just flagging that "Tech will utterly destroy the legal industry" and "We'll always have top lawyers starting at $160K" may be talking past each other a bit.

I don't disagree re: two markets, but even leaving aside big corporate clients there are barriers to outsourcing work. Individuals tend to have one-off problems. They need someone to represent them in a divorce, they need someone to do a will, etc. It's not clear how some "law center worker" sitting in India is supposed to do those things. If there was a market for it, you think someone would have already set something like this up in Nebraska. You can't quite pay $2 an hour for Nebraskans, but you could probably pay $10-15 and get U.S. licensed attorneys in a cheap cost of living state. Heck, document review work in major metro areas these days only pays about $20-$25/hour and those jobs get tons of applicants.
From what I've overheard from lawyer friends, it seems educational pedigree is way culturally overrated in the legal industry.

Apparently, if you have ten years experience in a specialized role, where you went to undergrad still matters. That's just crazy. So in light of that, it's possible that everyone's just putting so much stock in those graduates for no good reason at all.

Or maybe it's for the same reason as McKinsey and the other management consulting firms. They're billing these fresh grads out at 5-10X what they're paid, and those clients want to feel like they're paying for the best and brightest.

As for whether there's an opening you could drive a truck through.. it depends on how much of the work is 'routine' vs how much actually requires a seasoned lawyer to get done.

Yes. Also, there's a spectrum of legal work, from bread-and-butter conveyancing and wills, to appellate litigation.

The work that is truly straightforward can be automated... and is. For example, in Australia at least, you can get a DIY wills pack.

The essential need for humans in legal work is integration: relating a fact situation with relevant laws. The fact situation is not pre-recorded according to a grammar - and the semantics of the law are not precisely defined. Both aspects and their connection are far more flexible and ambiguous and uncertain than any IT integration. Possibly similar complexity though.

tl;dr we'll have automated IT integration long before we have automated legal work.

  Why do they continue to do this when lowering their 
  standards for pedigree could cut their costs massively?

  If there was really an opening in the market you could 
  drive a truck through, what has stopped anyone from taking 
  advantage of it thus far?
The key reason is that in this scenario, all the intelligence is being implemented in the computer, with a few Stanford JDs at the company making sure the code is up to snuff at a high level.

Then your less-skilled JDs can review the documents mass produced by software, in a sort of legal assembly line. They may not be able to follow chains of complex reasoning, but that's ok: engineers + Stanford JDs already made sure the legal templates and algorithms worked on representative data.

In other words, we are making use here of the old joke - "What do you call the guy who got the lowest GPA in medical school? Doctor."

Similarly, the less-skilled JDs who can't get jobs today do have one very important asset: they passed the boolean threshold and are licensed to practice law. Hollow out everything else, replace it by software, and use this army as scalable last-mile reviewers with the thinking done by code. Their value-add is that they absorb the last mile liability, as they are doing final review before release to client.

This is no different than the way that Intuit replaces the CPA in many situations. Kind of a TaskRabbit for law. By giving jobs to (a) young tech-savvy jobless students and (b) a few top attorneys from the very best law firms to write the contracts, you can do a pincer attack and massively reduce legal costs.

For the kind of simple contracts that could be handled by software, people already use form contracts. And remember, you don't need a lawyer at all to draft contracts.

As for more complicated contracts, wake me up when software can negotiate risk allocations for you.

  For the kind of simple contracts that could be handled by 
  software, people already use form contracts. And remember, 
  you don't need a lawyer at all to draft contracts.
I actually think this market is far bigger than one might think. Legalzoom is doing a $500M IPO and I only heard of them recently. Rocketlawyer is similar.

A high quality, heavily marketed site that had forms for the top 100 or so legal issues (with endorsements from top lawyers stating that they reviewed the contracts) would IMO be quite disruptive. Prenups, startup incorporation, wills, employment agreements, things like that.

Lawyers are not nearly as necessary for negotiating contracts as they want people to believe.

Lawyers, by reflex, attempt to insert themselves in the middle of processes, and naturally tend to create conflict. (This is according to my experience, and the experience of many others.) If a lawyer can do this, the negotiation will naturally take longer, generating billable hours.

Secondly lawyers all know that the person who is in a hurry to get a contract is going to get a worse deal, and so will NEVER be in a hurry to close the deal. But there are times when "winning" this negotiation is simply not worth the delay.

Now step back and look at it from the point of view of a company. Paying lawyers fees is not good. Creating conflict with someone you want to cooperate with is not good. Making negotiations take longer is not good.

Making this concrete, I was in a recent negotiation for a pay for performance compensation. Because it was pay for performance I could quantify how much delays cost the company versus what I will get paid. Their lawyers' delays cost them several times what I will get paid. On my side I did not let my lawyers negotiate. I told them to identify everything in the drafts that I should object to. Then I negotiated without them. Once there was nothing that made me unhappy, I signed. And what I got was pretty much the deal that we agreed to up front, in person, which I emailed to them months earlier.

Limiting my lawyers' role saved me thousands of dollars. If the company had limited theirs, they would have saved thousands of dollars. And would have had a contract faster.

TD;LR, just because your lawyers want to be involved, and try to convince you that they should be, doesn't mean that they are providing value.

On the flip side of the coin, lawyers can also mediate between the big egos on the business side that can cause a deal to fall apart over relatively small issues (this happens very often--executives are a lot more personally invested in the process than their lawyers are).

Ultimately, the fact is that you don't need a lawyer to negotiate a contract. Yet people use them almost universally in contract negotiations. Is it for irrational reasons? I like to think not, but who knows.

There are very rational reasons to use lawyers. If the other side is using lawyers, you can be sure that they have tried to slip in terrible terms that you really don't want. They won't look that bad, but they will be bad. Relying on yourself to find them is just stupid.

And once you're using a lawyer, that lawyer will naturally be inclined to try to turn the negotiation into a gravy train.

Try this. The next time you are part of a negotiation and want to redraft the agreement, ask yourself who you are doing it for. Are you serving your client's best interests? Or your interests?

Here is another exercise. The next time you see the need to mediate between "big egos on the business side", think carefully about how those egos got unhappy in the first place. I'll give you even odds that the issue they got upset over is one that was introduced by a lawyer. Could the conflict have been avoided entirely if it was negotiated differently? My guess is that, more often than most lawyers would want to admit, the answer is yes.

> And once you're using a lawyer, that lawyer will naturally be inclined to try to turn the negotiation into a gravy train.

So negotiate a fixed fee, with a bonus for the deal closing. Or find a lawyer that you trust to not run up the clock. There are a lot of things you can do as the client to guard against that sort of behavior.

> Here is another exercise. The next time you see the need to mediate between "big egos on the business side", think carefully about how those egos got unhappy in the first place. I'll give you even odds that the issue they got upset over is one that was introduced by a lawyer.

People get mad at their dentists for telling them to floss, but that doesn't mean anything by itself. The business people on a deal are optimistic--they think the deal will go well and is a good idea, otherwise they wouldn't be doing it. Lawyers are pessimists. They want to hedge against all the ways the deal can go wrong. They don't have a personal investment in the deal that can cloud their judgment of all of the ways it could blow up.

Realy? you think some start up is going to produce some software or that you could represent you effectively in court for non trivial cases?
You seem to have confused me with someone else. But feel free enjoy your straw man argument with an imagined opponent.

My point was simply that "negotiating for you" is not the best example of how a lawyer provides real value to clients. (Though the activity is undoubtably very profitable for lawyers.)

Those are valid points, and I've seen it from both sides - as a lawyer and as a businessperson. But one of the things missing from your comment is that good lawyers are protecting their clients from the possibility of things going wrong in the future.

Contracts are often unclear and are often broken. Sometimes things just don't work out. Regardless of how, litigation often ensues. That's when the value of a lawyer becomes apparent. It's easy to complain about lawyers during the negotiation - they often needlessly inject themselves into the process - but it would be worth considering that a good lawyer is trying to protect you in case things go wrong. Look at the Frank McCourt case for an example of where the lawyering was poor and it had a disastrous impact: http://www.minnlawyer.com/jdr/2010/09/17/legal-malpractice-c...

I agree. I think it would be stupid to sign any moderately complex agreement without a lawyer having been through it, and having informed you about what you are signing. (I certainly wouldn't sign that.) I also think it would be stupid to have a moderately complex agreement that was not drafted very clearly. And if things go south, you definitely need competent legal counsel.

My entire point was that lawyers like to inject themselves into the process of negotiation, often feel like they are generating value, but in reality tend to generate billable hours and negative client value at the same time.

I think the company that you're describing is LegalZoom.
>There has been an oversupply of legal gradutes for decades now. Today, half of law graduates don't get a job, and 90% can't get a job at a big firm.

This is true, and if anyone is curious about this issue and the data behind it, take a look at Paul Campos's recent book Don't Go to Law School (Unless), which you should also thrust into the hands of anyone you know who is contemplating law school.

I'm not sure this can work. Law above a certain low level has more in common with politics than computer code. A good case is a political campaign (just with very few voters), not a program.

Law 'feels' like code, its kind of the "legal fiction" we need to believe in to stay sane, but when it actually comes down to it, its anything but.

What you just identified is precisely what Peter is criticizing here. The whole point is that maybe technology can make this fiction less fictitious.

You may think that the law is necessarily more like politics, and that there's little room for code. But I don't see any reason for that in your comment, which is more just a reaffirmation that the status quo exists.

Good points. I probably didn't word that very well. What I'm getting at is that justice is mostly political not because we couldn't make it more rational, but because we really prefer it that way.

Certainly technology can make this fiction less fictitious in the future, but I'm not sure we'll want to use it that way.

This is surprisingly naive for someone like Peter Thiel. The law is not an axiomatic rule based system, and it's far too complex and fuzzy to be automated.

We haven't even automated medical records for godsakes!

This is surprisingly naive for someone who wants to build a floating city in order to create libertarian paradise?

What would be 'expectedly naive'?

This is surprisingly naive for someone who wants to build a floating city in order to create libertarian paradise?

Was it surprisingly naive for the explorers and founders of the New World to want to build new societies based upon their ideas? They packed up everything they owned and risked their very lives to go to somewhere lacking an entrenched society so they could build something new.

The effects of their experiment have been disproportionately dominant in the culture and growth of the world for centuries.

Now we're in a world where all the land is spoken for by government bodies that enforce their monopolies on governance through force. Imagine if government said, "There can be no new startup companies. You'll have to influence the existing companies from within." How would entrepreneurs feel about being unable to start new enterprises?

A continent generally provides more opportunities to be self-sufficient than a retro-fitted cruise ship.

EDIT: Also, survival rate on those colonization trips? Not so great. And that was with things like DIRT around that you can PLANT things in.

Basically, the 'floating city' idea is just a rebranding of what current cruise ships already offer: Pay exorbitant prices for everything in order to be on the ocean. Ain't no new society coming out of that.

Also, survival rate on those colonization trips? Not so great. And that was with things like DIRT around that you can PLANT things in.

Early colonists didn't know much about desalination and hydroponics.

Ain't no new society coming out of that

You very well may be correct as far as probability of an outcome, but your cynicism and lack of imagination are so very wrong... and sad.

Sigh. Plenty died of starvation and exposure. Read up on history sometime, disease was one of the lower risk factors the first few winters.

A bunch of programmers on a boat who need to get every single essential supply from the mainland is not a society. It's a tax avoidance scheme.

It might even be profitable for a generation, until they start having kids and incur more costs.

And the funniest part? Small groups of people in scarcity/survival situations tend to organize themselves into something that looks much more like collectivism than it does capitalism.

When you criticize my lack of imagination, do you mean 'imagine' in the John Lennon sense?

Yeah, it's like we don't even speak a similar language.
Except, you know, for the remaining 99% of colonial Americans, who came from multiple nations (i.e., Spain, Portugal, Netherlands, and France) primarily to exploit the vast natural resources for sale back in Europe.

Let's not forget that the American dream is built on the corpses of the many thousands of native Americans whose homelands were usurped by force, or the corpses of the thousands of slaves brought here against their will, or the corpses of the thousands of Rebel soldiers who fought for their rights to keep other men as slaves, or the corpses of the thousands of Mexican soldiers who fought to keep the Southwest out of the the greedy hands of frontiersmen.

Force is, and has been, the defining facet of American society since its inception. American is the dominant culture in the world today for other reasons (thanks Hollywood), but let's not pretend that the official, widespread use of force played anything less than a major role.

You're kind of all over the map spewing half-truths and full-hate there... but I have to point out the worst of your arguments.

You mentioned the thousands of Rebel soldiers who died fighting for slavery. Really, most were just fighting against being told what to do by the central government and didn't even own slaves. You conveniently didn't mention the thousands of WHITE Northern soldiers who died to end slavery and whose side prevailed, thus showing the core character of the nation. Why is it that folks like yourself generalize the American battle over slavery by mentioning only the losers rather than the winners who sacrificed so much for another race? It's kind of hard to have a meaningful discussion on interesting topics when your post is so intellectually dishonest.

Yes. Slavery had little to do with the civil war. You've totally rebutted the parent post. We're not heading straight into crazytown on this thread at all.
Seasteading is about the frontier. It's not about any political system. Lest we forget, there's quite a tradition of utopian thought within the progressive worldview as well. I think it's wrong to dismiss ambitious idealism as unrealistic.

Worst thing that happens - a bunch of people that don't want to live in the US anymore drown in the ocean. Best thing that happens - they experiment with new forms of government, some of which influence people back on land. Either way, we need more experiments in government, not less.

I don't think it's coming from a place of naivety. Thiel's legal pedigree is pretty impressive (Stanford JD, clerk for the Court of Appeals, etc). Maybe "optimistic" is more apropos.
Naive is a particularly strong word here. What part do you think is naive? Thiel is talking about both a 20-50 year vision and a 1000-year vision, and he is pretty non-committal about the 1000-year vision. And I don't think his 20-50 year vision is clearly wrong or naive. Rather - as a lawyer and as an engineer - I actually agree with it.

US law is supposed to be an axiomatic rule based system, but it is not consistent or complete, and in practice it ends up looking sort of arbitrary. But I don't think this means it is naive to assume that automation will make the system more consistent, or that it will help to simplify much of the complexity.

I am not suggesting that law will not always have some complexity and fuzziness - in fact I think it will - but I don't hear Thiel as explicitly saying the opposite - that we will ever get to a point where there is no complexity or fuzziness.

> an axiomatic rule based system, but it is not consistent or complete

Without disagreeing with your larger point, I can't help but point out you're bringing the wrath of Godel down on yourself here.

Comments like this are the reason I love HN.
Well why not start somewhere, create axiomatic TOS for example and proceed with contracts between companies.

There has to be academic groundwork to build something like that, a mixture of CS and law. Maybe some students at a law faculty could create a project together with a few CS students just to get the stone rolling.

A programming language to describe who is responsible for cleaning the flat.

(comment deleted)