Vinay Gupta: So, what I’m going to be talk­ing about today is whether we can get rid of the world’s third-oldest pro­fes­sion, which is bureau­crat. And specif­i­cal­ly the bureau­crats that han­dle resource allo­ca­tion for our soci­eties. Everybody thinks of bureau­crats as being kind of a neu­tral force. But I’m going to make the case that bureau­crats are in fact a very strong­ly neg­a­tive force, and that automat­ing the bureau­crat­ic func­tions inside of our soci­ety is nec­es­sary for fur­ther human progress. And this is a fair­ly auda­cious kind of a claim, so I’m gonna take it very slow­ly and we’ll see whether you agree with where I’m going, kind of one step at a time.

I’m also going to talk quite a lot about blockchains and Bitcoin and stuff like this. So I want to start by ask­ing does every­body here know rough­ly called Bitcoin is? Okay, so most of you’ve got a rough idea of Bitcoin. What about blockchain as some kind of tech­nol­o­gy which is sep­a­rate from Bitcoin? Yeah? Okay. So, smart con­tract? Few peo­ple for smart con­tracts. So I’ll take the tech­ni­cal part fair­ly gen­tly, and if we get into ter­rain that nobody under­stands, please raise your hands, let me know, and I’ll explain the con­cept. But I’m going to try and make sure that I go slow­ly enough that all of this comes along because the the­sis is rad­i­cal so the deliv­ery should be conservative.

So, blockchain ver­sus bureau­crat. Four top­ics that we’re going to cov­er. What is a blockchain? What is a bureau­crat? What is trans­paren­cy? And what are our options?

First top­ic. Basically we’ve gone through rough­ly forty years of real­ly full-on inno­va­tion in soft­ware. 1970s, we spent basi­cal­ly twen­ty years get­ting every­body’s head around data­bas­es. And what it pro­duces is the mod­ern orga­ni­za­tion­al form that we all know and hate, which is the giant silo. And the giant silo runs every­thing in your soci­ety. It’s your health­care data­bas­es, your edu­ca­tion data­bas­es, your gov­ern­ment data­bas­es. It’s every com­pa­ny you deal with. Its most sci­en­tif­ic research projects. It’s con­tent man­age­ment sys­tems. You know what this looks like.

1990s, we come along and we spend basi­cal­ly twen­ty years crank­ing through what hap­pens when we begin to intro­duce net­works into the mix. So, the Internet gets start­ed. The first access that I have to the Internet is on a machine that has a text-only screen that looks like this, in about ​’92 or ​’93, using a tech­nol­o­gy called Usenet, which to this day is still the best infor­ma­tion deliv­ery expe­ri­ence I’ve ever had. Because Usenet in those days was only filled with smart peo­ple. And it was a text-only medi­um so peo­ple real­ly worked on their lan­guage. It’s an amaz­ing medi­um but very nar­row and specific.

The prob­lems that we have in soci­ety are large­ly tech­nol­o­gy cre­at­ing eco­nom­ics cre­at­ing cul­ture. So the prob­lem of data­bas­es talk­ing to oth­er data­bas­es over the net­work has nev­er been ade­quate­ly solved. Every few years there’ll be anoth­er gen­er­a­tion of pro­posed solutions—EDI, XML, [WEDI?], SOAP, REST, Ajax, XML-RPC, autonomous agents. 

This goes round and round and round and round. It nev­er comes to a real con­clu­sion. Because data­bas­es were nev­er made to inter­op­er­ate. Each data­base has a mod­el of the world embed­ded with­in it and that’s the mod­el of the world of the orga­ni­za­tion that owns the data­base. And data­bas­es store what that orga­ni­za­tion con­sid­ers to be fact.

On the net­work, every­thing that you hear on the net­work is basi­cal­ly hearsay. It’s gos­sip. And so tak­ing exter­nal stuff (which is basi­cal­ly gos­sip and it’s in some­body else’s world mod­el) into the silo at the heart of your orga­ni­za­tion is basi­cal­ly much more like get­ting a brain trans­plant than it is like learn­ing some­thing. Because our tech­nol­o­gy is very crude. 

So the result of this is that even in a net­worked soci­ety, all of your per­son­al infor­ma­tion is in silos. And all of the silos belong to peo­ple that won’t talk to each oth­er. So it’s extreme­ly hard for you to make a change in your life and have that change prop­a­gate through your envi­ron­ment. When you move house you have to reen­ter your address in sev­en and a half thou­sand dif­fer­ent sys­tems. And each one of those orga­ni­za­tions prob­a­bly has 300 copies of your address. And the result is this kind of end­less night­mare of human beings inputting the same data over and over and over again into tire­less machines, and thus entire­ly a tech­no­log­i­cal arti­fact. It does­n’t real­ly kill any­body? So it’s not nec­es­sar­i­ly impor­tant, but it’s a sign of deep, under­ly­ing per­va­sive dysfunction.

So, last ten years or so there’s been anoth­er wave of kind of redis­cov­ery of cryp­tog­ra­phy, large­ly around this thing called Bitcoin. And the blockchain tech­nol­o­gy that under­lies Bitcoin is sim­ply a data­base that works like a net­work, or a net­work that looks like a data­base. And Bitcoin takes that con­cept and builds a sin­gle appli­ca­tion on top of it which is a cur­ren­cy. And it proves that the tech­nol­o­gy works. It proves that it can store infor­ma­tion secure­ly. Everybody is very hap­py with it…apart from the peo­ple that think that only cen­tral banks should be able to issue cur­ren­cy. And you get this enor­mous kin­da Cambrian explo­sion of exper­tise in soft­ware, because enough peo­ple make enough mon­ey doing Bitcoin stuff that they can then go off and work on their own projects.

And one project that came about that way is the thing called Ethereum, which imple­ments a thing called a smart con­tract. And a smart con­tract is where we take a piece of soft­ware and we store it in a blockchain data­base. So the blockchain ensures that every­body sees the same ver­sion of the soft­ware. And because com­put­ers are pre­dictable, if every sin­gle one of us runs the same piece of soft­ware in the same envi­ron­ment, we’ll all get the same results. And the result of that is that if you’ve got soft­ware inside of a blockchain, and you’ve got a kind of pre­dictable, secure way of run­ning the soft­ware, what comes out of this is con­sen­sus about what the soft­ware does.

So we can take a piece of code. We can all run the piece of code. We can all agree on the con­clu­sion. And as a result we can per­form com­pu­ta­tion that’s done at a soci­etal lev­el rather than with an indi­vid­ual run­ning the com­pu­ta­tion and then telling you what the results are and ask­ing you to trust them.

So this is a very, very, very ear­ly imple­men­ta­tion of a kind of kind of gen­uine­ly demo­c­ra­t­ic soft­ware. Because every­body shares in the com­pu­ta­tion, every­body shares the result, and you can have con­fi­dence across the whole of a soci­ety that this is what the soft­ware does and this is what the soft­er says. 

And that real­ly has­n’t hap­pened before. The old mod­el is that the soft­ware runs in the silo and the peo­ple that run the silo tell you what the soft­ware says. If the com­put­er says no, that’s the end of the sto­ry. In open source soft­ware, this is mod­i­fied a lit­tle bit. If the com­put­er says no, at least you could exam­ine the source code and find out why it said no. But you still have all the issues around secu­ri­ty and trust about the hard­ware it runs on and the rest of this kind of stuff.

So the notion that you could have soft­ware that runs for the ben­e­fit of a soci­ety that the entire soci­ety can par­tic­i­pate in and audit, this is new stuff. And I think that that has got­ten obscured a lit­tle bit under­neath the ear­ly press around blockchains and Bitcoin. There is a lot of con­fu­sion about what does it all mean and how does it all fit togeth­er. But these under­ly­ing lessons have become clear­er as the stuff grows more mature. 

Now, is every­body rough­ly with me so far? Okay, good. I just want­ed to check because this blockchain stuff is real­ly new and real­ly com­pli­cat­ed sometimes.

Second top­ic, what is a bureau­crat?. Would any­body in the audi­ence like to define what a bureau­crat is for me? Yeah, see you’re going to wind up with my def­i­n­i­tion, which is a bureau­crat is sim­ply an impe­r­i­al col­lec­tor of tax. Most of the cul­tures that have bureau­crats are cul­tures where the gov­ern­ment is fund­ed by tax­a­tion. And the job of the bureau­cra­cy is to iden­ti­fy who should pay how much of the tax; who has paid their tax; who has not paid their tax; and who you need to send the boys round to to get the tax off. This is why I say bureau­crat is the world’s third-oldest profession. 

In the­o­ry, once you’ve got bureau­crats that are work­ing for democ­ra­cies, the bureau­cra­cy is work­ing for the peo­ple as a whole. In prac­tice, if you have an insti­tu­tion which has been impe­r­i­al tax col­lec­tion for 4,000 years and then you put a cou­ple hun­dred years of bureau­cra­cy on top of it, not very much changes. Which is why every time you deal with bureau­crats in cen­tral gov­ern­ment or in big com­pa­nies you feel like you’re deal­ing with an impe­r­i­al enti­ty rather than with some­thing that rep­re­sents your interests. 

And there is no good bureau­cra­cy, because the orga­ni­za­tion­al form of Hell is the pyra­mid and every bureau­cra­cy is a pyra­mid. All that it lacks on top is an emper­or, and even that seems to keep com­ing back. Look at the increas­ing­ly impe­r­i­al machi­na­tions of the American pres­i­den­cy, right. This kind of struc­ture in the soci­ety nev­er worked for the peo­ple, and it was nev­er designed to work for the peo­ple. This is an old pow­er struc­ture that we tried to repur­pose for democ­ra­cy and that in fact has nev­er being suc­cess­ful­ly repur­posed for gen­uine­ly demo­c­ra­t­ic goals. Reasonable? Maybe.

So once we accept that bureau­cra­cy is the orga­ni­za­tion­al form of Hell and that it defines impe­ri­al­ism, you could begin to under­stand that there is an incen­tive for tear­ing down the bureau­crat­ic struc­tures around us. But we’d like to do that with­out also tak­ing down the soci­ety. Because a post-bureaucratic democ­ra­cy might be a democ­ra­cy that actu­al­ly feels like a democ­ra­cy, rather than a democ­ra­cy that feels like impe­ri­al­ism where you get to choose your emper­ors. This is the hope, anyway.

The agency prob­lem is the kind of tech­ni­cal eco­nom­ics descrip­tion of this. The bureau­crat always works for the bureau­cra­cy. The bro­ker always works for the bro­ker­age or for their own inter­est. Finding some­body that rep­re­sents your inter­ests at any kind of scale larg­er than the indi­vid­ual is extreme­ly dif­fi­cult. So if you’re inter­est­ed in this ques­tion of who do the bureau­crats work for, agency prob­lem is where you look for the dis­cus­sions of that.

Okay, now trans­paren­cy. There’s a lot of dis­cus­sion about things like open data, trans­par­ent demo­c­ra­t­ic process­es, trans­paren­cy in gov­ern­ment, free­dom of infor­ma­tion. You know, all these kind of efforts inside a gov­ern­ment to be more trans­par­ent. But I think that trans­paren­cy, when you boil it right down, is actu­al­ly a ques­tion of juris­dic­tion. You can have a cer­tain amount of secre­cy inside of gov­ern­ment to pro­tect things like a pro­cure­ment process. But what you want is for the legal process­es that over­see the gov­ern­ment to be ful­ly trans­par­ent. And if you have a trans­par­ent legal sys­tem, usu­al­ly it gives you the abil­i­ty to get the trans­paren­cy in the oth­er areas where you need it, by force if nec­es­sary. You can sub­poe­na information.

So what is trans­paren­cy? Law you can read. Open courts, where you can actu­al­ly have peo­ple go in and lis­ten to the court mak­ing its deci­sions. The abil­i­ty to always name who makes a deci­sion. So if you’ve ever worked in a con­text where you have health insur­ance, the health insur­ance peo­ple will nev­er take per­son­al respon­si­bil­i­ty for mak­ing a deci­sion. The deci­sion is always the sys­tem’s deci­sion, and you can­not extract the name of the indi­vid­ual that made the choice. And in the­o­ry this is about pro­tect­ing these peo­ple from reprisals or it’s about cor­po­rate respon­si­bil­i­ty. But in prac­tice it allows you to shuf­fle some­body who makes ter­ri­ble deci­sions at one job into anoth­er job with­out any­body ever hav­ing to get fired. The bureau­cra­cy serves the bureau­crats rather than the peo­ple that it pur­ports to serve.

The oth­er impor­tant part of this is a pub­lic dis­cus­sion about what is true. So, in many cas­es you wind up in a posi­tion where there is no pub­licly agreed-on def­i­n­i­tion of true, in the areas where there’s a con­tro­ver­sy. And the great moth­er of all these prob­lems is glob­al warm­ing. But you can also see it around vac­cines. America has a recur­rence of ter­ri­ble dis­eases that we’ve got­ten rid of gen­er­a­tions ago like, I don’t know, whoop­ing cough, because peo­ple have sim­ply start­ed to refuse to vac­ci­nate their chil­dren based on a bunch of botched sci­ence and a whole bunch of kind of New Age pro­pa­gan­da. And the result of that is a ter­ri­ble destruc­tion of pub­lic safe­ty, caused by what’s essen­tial­ly a new super­sti­tion. It’s like they’ve rein­vent­ed witch­craft and they called it vaccination.

So you need some abil­i­ty in a trans­par­ent soci­ety to have a gen­er­al assent about what is true, based on a pub­lic exam­i­na­tion of the facts, that does­n’t become polit­i­cal­ly cor­rupt­ed. And polit­i­cal cor­rup­tion of sci­ence is how this is being attacked right now.

Finally, all of this does no good if you nev­er learn any­thing. So, when things go wrong you have to have mech­a­nisms inside of the soci­ety that don’t repeat the same error end­less­ly. And usu­al­ly that means you need some kind of mech­a­nism to draw con­clu­sions like they do for inquiries after an air crash, and then apply those con­clu­sions to the soci­ety that you’re oper­at­ing in. So you don’t repeat, you always have some kind of account­abil­i­ty so that when things go wrong you can point the way.

My sug­ges­tion is that blockchain tech­nol­o­gy, par­tic­u­lar­ly smart contract—so soft­ware in blockchains—has all of the attrib­ut­es nec­es­sary to pro­vide trans­par­ent gov­er­nance. You can have a piece of soft­ware. It’s stored in pub­lic. Your decision-making process­es is inside of the soft­ware are com­plete­ly trans­par­ent. The peo­ple that put the soft­ware into the sys­tem are named indi­vid­u­als and the audit process­es are also all done on a blockchain. The pub­lic analy­sis of fact is in the form of a thing called an ora­cle, which is where you take infor­ma­tion that is not intrin­si­cal­ly inside of a com­put­er like tem­per­a­ture of the ocean, and you agree on the fact before it’s placed into the soft­ware. And then final­ly the out­puts are all vis­i­ble, so it’s pos­si­ble to go over the entire set of out­puts of the sys­tem and look for what actu­al­ly happened.

So if we have soft­ware that can imple­ment trans­par­ent gov­er­nance, the ques­tion then is what can we do with it? And I have a kind of rad­i­cal sug­ges­tion, which is real­ly just some­thing to play with. But I think it gives some scope of the new pos­si­bil­i­ties. And if you could do this spe­cif­ic instance there must be 800 oth­er things you can do which are small­er, eas­i­er to imple­ment, or less controversial.

So, my sug­ges­tion is this: We could imple­ment basic income on a blockchain. You could take a small soci­ety, or even a big soci­ety, and you could decide that every cit­i­zen will have so many tokens which are issued to them every day or every month or every year by virtue of their cit­i­zen­ship. And you could say that these are euros, but this cre­ates a prob­lem of financ­ing these things. So you could sug­gest that you’re going to use some kind of elec­tron­ic currency.

So you have a basic income unit which is issued into the soci­ety, in known scarce quan­ti­ties. And what that unit is worth is allowed to float on the mar­ket and becomes worth what­ev­er peo­ple decide it’s worth. And you can increase or decrease the amount issued, depend­ing on social pol­i­cy decisions.

Blockchains are ide­al for this. You want to iden­ti­fy every cit­i­zen that’s enti­tled to pay­ments: you log them into the blockchain sys­tems. You want the pay­ment to be trans­par­ent: the soft­ware issues the pay­ment and every­one can see what the pay­ment was. There’s no ambi­gu­i­ty in this. Because every­body is known to be receiv­ing exact­ly the same size of pay­ment, there’s no social stig­ma asso­ci­at­ed with the pay­ment. So hav­ing it be ful­ly trans­par­ent might have less of a neg­a­tive impact than you would get if it was say a means-tested ben­e­fit. And I think that is a—you know, it’s a jump, but I real­ly want­ed to show some­thing which is a lit­tle sci-fi, just to see what we think.

The out­come of this is that you get trans­par­ent spend­ing. So, because all the spend­ing is done on a blockchain, you can see what the stuff is spent on. And at this point you have a deci­sion to make. You can require that cit­i­zens are iden­ti­fied on the blockchain by their name, or you can pro­tect their name and have them iden­ti­fied on the blockchain by a pseu­do­nym, or an alias, or a ran­dom­ized iden­ti­ty number. 

So, in one of these paths you get com­plete trans­paren­cy about indi­vid­ual spend­ing, and in the oth­er path you only get trans­paren­cy about the gen­er­al aggre­gate of spend­ing. And then there are ques­tions about can peo­ple be deanonymized suc­cess­ful­ly. The answer to that is it depends. It’s a com­plex tech­ni­cal ques­tion. But gen­er­al­ly this is one of these social choice deci­sions where you’ve got to decide whether you want full trans­paren­cy or not.

With a trans­par­ent spend­ing data­base, you could then do real­ly detailed analy­sis of how the mon­ey from basic income is being used. And I think every­body involved in the basic income debate, whether they’re for it or against it, could agree that basic income is a real­ly good thing to get com­plete knowl­edge about.

So you take this new tech­nol­o­gy which is kind of untest­ed at a large scale. (Blockchains.) You take a set of require­ments for a sys­tem to be trans­par­ent and imple­ment a blockchain tech­nol­o­gy that is faith­ful to those. Which is pret­ty much all of them but you want to be clear. You need a much more detailed spec­i­fi­ca­tion than the one that I’ve out­lined. And then you take a new social pol­i­cy instru­ment like basic income and you imple­ment it on top of the blockchain to give you a lab­o­ra­to­ry in which you can see exact­ly what’s hap­pen­ing inside of a social inno­va­tion, to decide whether or not you want to scale it, whether you want to con­tin­ue it, just to under­stand what it is.

And I think that there’s an enor­mous amount of room for tak­ing prob­lem­at­ic areas of soci­ety and deploy­ing these kind of lens solu­tions to them. We’re going take this prob­lem. We’re going to imple­ment a solu­tion that we can mon­i­tor every sin­gle aspect of. We’re going to run it on an exper­i­men­tal basis. And we’re going to agree all the way through what the rules of the game were, to make sure that peo­ple aren’t cheat­ing in some way to make the idea look like a bet­ter idea than it is.

The soft­ware can be audit­ed by all par­ties before you put it into the sys­tem. And in this instance the code lit­er­al­ly is the law. There’s no dis­tinc­tion between who imple­ments the soft­ware and who makes the law. The law and the code are one. So you need to have a legal process by which the author­i­ty of the state or the author­i­ty of a cor­po­ra­tion or a sort of char­i­ty, even, is trans­ferred direct­ly into a piece of soft­ware. And those mech­a­nisms already exist. You can get there through bind­ing arbi­tra­tion. You can get there through what’s called Ricardian con­tract, designed by a friend of mine called Ian Grigg. There are lots of approach­es to that prob­lem but you just need to pick one and imple­ment it.

Does this gen­er­al­ly sound…even slight­ly pos­si­ble? Yeah, right? So I sort of sat down as I was writ­ing the slides for this think­ing, can it real­ly be that easy? Is it even remote­ly rea­son­able that you could just come along and take a bunch of new social pol­i­cy; write a bunch of soft­ware; put the soft­ware into the new social pol­i­cy; and then do the analy­sis in real time as the trans­ac­tions were done?

And I kept look­ing at it like that can’t be right. There must be some catch. It must be anti-democratic, or bla­tant­ly ille­gal, or it must give you cor­rupt­ed data or some­thing. And I kept look­ing at it and I could­n’t fig­ure out why it was­n’t a rea­son­able thing to sug­gest. So here we are talk­ing about it. And if any­body can tell me why it’s not rea­son­able, let me know. Come up and tell me after­wards or talk about it on Twitter.

Because as far as I can tell, it means that you could try basic income at a whole of soci­ety lev­el with very lit­tle risk and very lit­tle cost. And you could mon­i­tor exact­ly what was hap­pen­ing. And giv­en that half of Europe has an enor­mous prob­lem with access to liq­uid­i­ty, it seems like you could have an injec­tion of liq­uid­i­ty into the soci­eties with­out nec­es­sar­i­ly caus­ing main­stream eco­nom­ic dis­rup­tion. And if it does­n’t work it’s fair­ly easy to turn off. It was quite surprising.

Now. What this kind of sug­gests is a method of gov­ern­ment inno­vat­ing. Because right now we’re in a posi­tion where gov­ern­ments have ceased to inno­vate. In fact, they’re all liv­ing inside of this bizarre fan­ta­sy world which is left over essen­tial­ly from the end of World War II. The state has­n’t real­ly got­ten to grips with moder­ni­ty in any fun­da­men­tal way. Anytime you have a deci­sion being made, the basic struc­ture and process that makes that deci­sion is the same struc­ture and process they would’ve used after World War II. Things are a bit dif­fer­ent pre World War II—there’s much more diver­si­ty in how gov­ern­ments oper­ate. After World War II there’s a strong con­sen­sus. It comes out this is pret­ty much how a state should be run, and we kin­da fall into this pattern.

The prob­lem is that that pat­tern is very very poor­ly suit­ed to exact­ly the kind of com­plex moder­ni­ty that David Orban was talk­ing about, all of this kind of strat­e­gy up expo­nen­tial curves busi­ness. You know, rapid rate of change plus 400 year-old demo­c­ra­t­ic design, plus bureau­cra­cies that date back to World War II, gives you this prob­lem where the rate of change of the world mas­sive­ly exceeds the rate of change for democ­ra­cy to be able to adapt to the world.

And the result is an increas­ing trans­fer of pow­er from demo­c­ra­t­ic to non-democratic insti­tu­tions. The only rea­son that the cor­po­ra­tions wind up with such enor­mous pow­er is because the gov­ern­ment is unable to solve prob­lems with­out them, and basi­cal­ly lets them run around doing what­ev­er they like the rest of the time. 

So, because the cor­po­rate forms can evolve rapid­ly, because you’ve actu­al com­pe­ti­tion between forms, there’s this con­tin­u­al push and this con­tin­u­al accel­er­a­tion. And what it pro­duces is a much bet­ter under­stand­ing of how to deliv­er change than you get if you’re inside of a bureau­cra­cy where your boss changes every four years, using a pro­ce­dure that’s 400 years old, and the insti­tu­tion does quar­ter­ly report­ing but real­ly only mea­sured on a four-year cycle for elec­tions. You see this kind of mis­match of speeds, right. It’s a prob­lem of tempo.

So if we could devel­op mech­a­nisms for weav­ing the exist­ing super­struc­ture of democ­ra­cy in place, but per­form­ing exper­i­ments inside of trans­par­ent sub­sys­tems, it’s pos­si­ble that you could use blockchains and sim­i­lar tech­nolo­gies to build exper­i­ments at small scale with the man­dat­ed author­i­ty of the state but with­out going through cor­po­ra­tions as the implementers. 

So, open source soft­ware com­mis­sioned by a gov­ern­ment, built by a uni­ver­si­ty, deployed across a pop­u­la­tion with a social man­date behind it, gives you a kind of test pool for new con­cepts that can be tried in an envi­ron­ment where nobody can accuse any­body of hid­ing cards up their sleeve. 

And that sort of notion that we’re lack­ing, that abil­i­ty to test new ideas in soci­ety, is behind a lot of the crazy think­ing that comes out of Silicon Valley. So the seast­eading move­ment by Patri Friedman (or rather Patri Friedman ran ini­tial­ly) was this idea that you could build kind of lit­tle arti­fi­cial islands out in the oceans. You could stick a flag in them and declare they were sov­er­eign. And then you could exper­i­ment with new forms of gov­ern­ment. This one will have an elec­tion every fif­teen min­utes and your phone will do arti­fi­cial­ly intel­li­gent vot­ing for you. This one over here will have six-person MP groups rather than indi­vid­ual MPs, and they’ll rotate off the job. This one over here will ran­dom­ly select who the MPs are, and they’ll have a tenure which is decid­ed ran­dom­ly. This one over here, you’ll just have sor­ti­tion every four years, you’ll ran­dom­ly select… You know, all of these kind of dif­fer­ent approach­es could be tried in these kind of micro­cosms, and we could poten­tial­ly find some­thing that works bet­ter than four or five-year elec­toral democracy.

But, the prob­lem with this is the assump­tion that you need to be able to take ter­ri­to­ry, have sov­er­eign­ty, get land and all the rest of this, makes it incred­i­bly dif­fi­cult to get these kind of exper­i­ments in gov­er­nance going. And, who watch­es the watch­men? What do you do when one of these things turns into a total dis­as­ter? Are these peo­ple going to then go back to their orig­i­nal nation-states and beg for help? You know, you can imag­ine these projects if they go down, the human­i­tar­i­an fall­out could be quite sub­stan­tial because you’d have a crash of gov­er­nance in a tiny lit­tle island soci­ety. And I don’t like the smell of that at all. I mean, I’m quite… I’d love to see the exper­i­ment tried, but I think it’s quite risky. And we cer­tain­ly can’t solve prob­lems quick­ly if we have to estab­lish a new nation-state every time we want to try an experiment. 

So what I quite like is this idea that we could do lots of exper­i­men­tal stuff inside of the body of democ­ra­cy by using trans­paren­cy tech­nol­o­gy to de-risk exper­i­ments in gov­ern­ment. And the notion that these blockchain tech­nolo­gies are a fun­da­men­tal trans­paren­cy break­through, and that you can push for­ward with that set of schemes to try these exper­i­ments in all kinds of dif­fer­ent places and for all kinds of dif­fer­ent things, with the abil­i­ty to doc­u­ment exact­ly what hap­pened for future gen­er­a­tions, seems like a real­ly sound and sen­si­ble way to do exper­i­ments in governance.

And you don’t real­ly hear ​“exper­i­men­tal gov­er­nance” all that often. But it’s clear that we need it, because the exist­ing sys­tems don’t work—we all know that. I mean, does any­body live in a coun­try where they feel like their gov­ern­ment real­ly works? I went to Singapore. Their gov­ern­ment real­ly works. I mean it’s incred­i­bly effi­cient. But they basi­cal­ly run the entire island like a sin­gle cor­po­ra­tion. And pri­vate enter­prise com­plains that they can’t get enough tal­ent into pri­vate com­pa­nies because all the smart peo­ple work for the gov­ern­ment and they just hire them right out of uni­ver­si­ty and give them a job for life. And it seems very unlike­ly that we’re going to suc­ceed in chang­ing our soci­eties in that way, and we prob­a­bly would­n’t like it even if we did.

So we have to get bet­ter at learn­ing inside of democ­ra­cy. And my sug­ges­tion is that blockchain tech­nol­o­gy is an inher­ent­ly use­ful part of learn­ing inside of democ­ra­cies. It came from the lib­er­tar­i­ans. It was dressed up in some very weird clothes when it arrived. But if you get a good sol­id chance to look at it and talk to it and get some sense of what’s going on under the hood, I think it’s pret­ty clear that it’s sim­ply about trans­paren­cy. And trans­paren­cy is a method­ol­o­gy that can be pulled out of any polit­i­cal set­ting and reap­plied every­where else. Because the truth is only one, right. When you get right down to the fun­da­men­tals, a trans­par­ent sys­tem is the same for every­body that uses it, and there’s no rea­son that that should be iden­ti­fied with lib­er­tar­i­an or any oth­er kind of pol­i­tics. It’s some­thing that peo­ple of any polit­i­cal per­sua­sion can apply. 

So that’s the basic the­sis. I think that blockchains are good for democ­ra­cy. And I think that if we use them wise­ly inside of gov­ern­ment, it will give us an oppor­tu­ni­ty to do social exper­i­men­ta­tion in governance—in fact, just exper­i­men­tal governance—in a way that is large­ly de-risked and polit­i­cal­ly neu­tral because of the trans­paren­cy aspect. I will also go on to some kind of subtopics after this, but that is the main talk. Thank you.