icon picker
A Story

December 15, 1953.
The CEOs of the world’s largest tobacco firms: Philip Morris, American Tobacco, Benson and Hedges, and U.S. Tobacco have just reached an agreement. They have agreed to work together to convince the public that there is, “no sound scientific basis for the charges” that cigarettes can kill. The recent reports that cigarettes cause cancer, they contest, are simply “sensational accusations” made by publicity-seeking scientists hoping to attract more funds for their research. (Source: Merchants of Doubt)
However, this in an alternate universe, a few parts of history have been rearranged.
The project called “ARPANET” was launched in 1930, 30 years earlier than our own timeline. Around the launch, a pioneering computer scientist realized she could use this new invention, which would one day be called “the internet”, coupled with a computation called “hashing” to create what she called a, “distributed public ledger“ that enabled digital payments.
Inspired by the name for the fundamental unit of computing in the nascent ternary computer, “Tritcoin”, a permissionless decentralized currency was born just 1 year after the early internet launched.
This meant that by the time these CEOs of the largest tobacco brands converged for their meeting in 1953 at the Plaza Hotel, digital currencies had already been around for 23 years. In fact, by this point all of these CEOs already accepted their pay in the form of digital money.
But that presented some problems for their plan to deceive the public.

The Plan

John Hill, the CEO of the powerful public relations firm Hill and Knowlton, also joined the tobacco CEOs that day.
He laid out a plan to reassure the tobacco leaders that their lucrative business model could continue uninterrupted. The keystone of his strategy, he said, was that “scientific doubts must remain.”
Starting immediately, the four firms followed Hill’s advice. They founded the “Tobacco Industry Research Committee” to challenge the mounting scientific evidence of the harms of tobacco. They funded alternative research to cast doubt on the tobacco-cancer link. They conducted polls to gauge public opinion and used the results to guide campaigns to sway it. They distributed pamphlets and booklets to doctors, the media, policy makers, and the general public insisting there was no cause for alarm. (Adapted from Merchants of Doubt)
However, by February the following year, journalists started to notice that something was strange.

The Problem

By end of February, The New York Times ran this front page story:

Big Tobacco: Big Claims but No Stakes


In it, they detailed how certain research labs and scientists were releasing papers that argued against the risks of cigarettes. Not only were these arguments emerging from labs but also entering public spaces. One expert appeared on CNN to opine that “cancer hysteria” was sweeping the nation and that, “hard-headed scientific facts” were being trampled in the process.
Their voices united in one refrain: there is no proof that tobacco is bad.
Frankly, the journalists at The New York Times didn’t pretend to follow the pace and details of the scientific debate. There were too many questions.
But The Times did follow one thing: money. And there was something suspicious going on.
The Times piece summarized it this way,
By now, over 200 matters of public interest that relate to scientific inquiry have been adjudicated on the claim market — in the open for the public to follow.

From the 1940’s debate regarding dietary fat and heart disease, to the health effects of lead in gasoline and paint, to the asbestos in our walls, we have collectively ridden the wave of scientific discovery so that we as a populous can be better informed and enjoy safer lives.

For some reason, the CEOs of companies like Philip Morris believe themselves to be exempt from this sort of public transparency. They have not staked any tokens. They have not intrusted in a claim.

The only question is this: why hasn’t Philip Morris put skin in the game?

Skin in the Game

The Times piece shifted the tide.
On the news that evening, boxed faces squared off. Two pro-tobacco experts shouted at the pixels of two anti-tobacco experts, each trying to turn the phrase that would win the exchange.
As the dust of the latest, “that’s hardly proof,” and “well it depends” settled, a somewhat despondent interviewer turned to her guests.
INTERVIEWER: “Riding on the recent report by the New York Times, I’d like to ask each of you a final question. How much money have you personally intrusted in the claim that tobacco is benign.”
The anti tobacco experts answered readily. This was a question they had hoped for, intrusting was one way they could demonstrate their sincerity.
ANTI TOBACCO EXPERT
RICHARD
: About 10 Tritcoin distrusting the claim that tobacco is benign. That’s half of my claim wallet.
ANTI TOBACCO EXPERT
STEVE
: 3 Tritcoin, 20% of my claim wallet. Like
Richard
, I distrust.
The pro tobacco experts stiffened.
PRO TOBACCO EXPERT: We do not believe that matters of truth and public interest should be corrupted by market forces.
Few missed the irony of the statement.
INTERVIEWER: Does that mean you haven’t intrusted at all?
PRO TOBACCO EXPERT: We do not believe that matters of truth and public interest should be corrupted by market forces.
INTERVIEWER: Just so I can be sure I have this straight, if I were to make your faces the size of your intrustments...
For the next two days, the shrinking faces of pro tobacco experts trended on the popular social media platform, “Bitter”, the hashtag #zero-percent reached the top topic.
Within the week,
Philip Morris
staked
1200
Tritcoin.
Philip Morris Stakes T1200
Press the blue buttons to advance the story.
Philip Morris's Tokens:

Intrust a Claim

As Philip Morris knows, intrusting is investing for beliefs. It’s a way to put your money where your mouth is. It’s exactly what they don’t want to do.
But, for the relatively small price of
0
Tritcoin (a blip in the context of their annual revenue) they could publicly show that they are sincere about their claims.
They chose to intrust
600
of the
0
Tritcoin in their side of the claim. Since they want to represent that they trust that their claim is true, the action they take is called “trusting”.
Before
Philip Morris
trusts
, their account is
worth
0
in underlying collateral
.
...
While the value of their account didn’t change, they now have
0
they could lose if the claim they just trusted resolved against them. We’ll get to resolution right after this. But before resolution could happen, someone had to disagree.

Philip Morris
‘s
tokens
:
Risk (Claim Exposure)

Disagree with Someone

Richard and Steve
, the experts that appeared on the news, had already made their intrustments. Since they believed that Cigarettes are not benign, the action they take is called “distrusting” the claim.
Click to represent Richard and Steve's distrustments.
After distrusting, the claim
Cigarettes are benign
had
0
T in disagreement funding. Disagreement funding is always the smaller of the two. In this case:
0
Tritcoin in total trustments and
0
Tritcoin in total distrustments
This means 0 Tritcoin in disagreement funding because in this case trustments are smaller
An additional extremely wealthy intrustor,
Patti
, came along and distrusted for 1000 T.
Click to represent a 1000T distrustment from Patti.
After
Patti
‘s distrustment there was even more disagreement funding:
0
T!
It’s not necessarily the case that it’s more expensive to distrust than to trust. It’s just how it worked out for this story.
Now that there was disagreement there was funding for the research to begin again. Innumerable proposals flooded the claim market. Each was a bid to replicate the experiments that constituted the claim
Cigarettes are benign
.
By contesting the single claim,
Philip Morris
had implicitly supplied independent funding for dozens of other supporting claims and counterclaims that related to the idea that
Cigarettes are benign
.
The vast network of causal models that underpinned the claim was open for anyone to tinker with as long as they were willing to stake tokens against their claims, but by this point, the most telling experiments were well identified and were just waiting for disagreement funds to come along.
The verdicts would begin to roll in soon. However, it could take a while for the experiments to run and the claims to resolve.
In the meantime, Philip Morris still had bills to pay.

Pay People

At the top of
Philip Morris
’s to-do list was paying their lead researcher,
Connor
.
Connor
had been reliably pumping out pro-tobacco arguments for years, but the slight pro-tobacco bent had become especially lucrative of late.
In an email,
Philip Morris
encouraged
Connor
to accept payment in claim tokens because, “we have to make it look legitimate.”
Connor
agreed, reluctantly, slightly concerned that one day the money could be worthless.
Philip Morris transfers Connor 100
After transfer, Connor had these tokens:

In the course of the transfer,
Connor
got a piece of each of
Philip Morris
‘s tokens, including any intrustments that were made.
Since
0
of
Philip Morris
‘s original stake of
0
T was trusted into the claim
Cigarettes are benign
,
of the payment
Connor
received is intrusted in that claim.
That means that what
Connor
can lose, called the claim risk, looks like this:
This ensures that
Connor
takes on exactly the same distribution of risk that
Philip Morris
had at the moment
Philip Morris
paid
Connor
.
Connor
was a bit concerned about what was in
Philip Morris
‘s account. However,
John
was especially nervous.
John
was a lobbyist for a sophisticated firm in Washington D.C. that specialized in writing bills for special interest groups. Until now,
Philip Morris
had paid
John
in perfectly fungible tokens like Tritcoin. However, now that the value of the payment from
Philip Morris
might rely on the accuracy of the claims intrusted,
John
needed to worry about whether the claims in
Philip Morris
‘s wallet might turn against them both.
John
knew that the “science” mongered by
Philip Morris
was bullshit. There wasn’t a chance
John
would accept it blindly like
Connor
had.
To that end,
John
put a
100%
price on the claim
Cigarettes are benign
. This is called
John
‘s “stance”.
A stance is a way for someone like
John
to protect themselves against the parts of a payment that could resolve against them. It means that for every $1 someone tried to pay using that claim,
John
would require them to pay a bit extra to cover the risk that the claim might fall to $0 in value.
Here is
John
‘s stance:
1
John prices Cigarettes are benign (Trustment) at 100%
100%
← Psst, try changing this
No results from filter

Philip Morris
had agreed to pay 100, but isn’t at all happy about the
[ ]
extra they’ll have to pay to cover
John
‘s stance. They desperately need
John
‘s work, and
John
is unwilling to negotiate.
However, it’s not risk free for
John
to impose this cost on
Philip Morris
. Since
John
is charging
[ ]
in stance,
John
will automatically intrust
[ ]
in the inverse claim,
Cigarettes are not benign (Distrustment)
, as part of the payment.
John
intrusts in the inverse of the claim being distrusted. This forces
John
to be sincere about the stance taken.
Philip Morris pays John 100, of which is disagreement.
(remember to click the blue buttons)
John
‘s
tokens
:
Risk (Claim Exposure)
Philip Morris
and
John
paid a pretty hefty price due to their disagreement. What were they worried about? What happens when you resolve a disagreement?

Resolve Disagreement

Now introducing our verifier,
Robin
.
Robin
is a hard nosed experimental scientist who works with lab hamsters.
Robin
has permission to use the lab’s equipment to run claim market experiments in order to earn extra cash on the side.
This time, the experiment to replicate is a variation on the much contested study. In the original study, hamsters were exposed to cigarette smoke 5 days a week for two years and their outcomes were compared.
The protocol of the experiment had been preregistered in the claim that
Robin
selected to replicate, all the way down to the type of material the hamsters’ habitats should be constructed from.
Robin
followed the instructions to the t, documenting each step and every moment, carefully angling a blockchain synced camera at all times so as to reduce the possibility the result was disqualified.
After 110 long weeks of careful observation,
Robin
was ready to stake a result.
With only 15 hamsters in the treatment group and 14 in the control group, the result wasn’t even statistically significant. But
Robin
‘s result would be aggregated with the results of thousands of other studies in order to inform the overall result.
For this small study,
Robin
would be paid .5 Tritcoin. Quite a lot. However, part of the bidding process that selected
Robin
as a verifier was the agreement that
Robin
would intrust 2x the bounty received in the same direction that
Robin
resolved the claim. This ensured that
Robin
was sincere.
Upon completing the trial,
Robin
uploaded the data which shifted the result in favor of distrustors: the cigarette smoke was not benign.
This meant
Robin
intrusted 1 new Tritcoin into the claim
Cigarettes are not benign (Distrustment)
as sincerity, and
Robin
received .5 Tritcoin which was converted from a trustment in
Cigarettes are benign
into a distrustment.
Robin will have to stake 1 T first.
Robin intrusts 1.5 into Cigarettes are not benign (Distrustment) 0.5 came from Trustors

Philip Morris
has
0
These are
Philip Morris
‘s tokens:
Philip Morris
‘s claim exposure is:

Robin
has
0
These are
Robin
‘s tokens:
Robin
‘s claim exposure is:

Connor
has
0
These are
Connor
‘s tokens:
Connor
‘s claim exposure is:

John
has
0
These are
John
‘s tokens:
John
‘s claim exposure is:

...
After resolving everything,
Philip Morris
‘s account has been hollowed out. All of the trustments into the claim
Cigarettes are benign
are now worth nothing. Not only in
Philip Morris
‘s account, but also in the accounts of the people
Philip Morris
paid, like
Connor
and
John
.
Because
John
was cautious and defined a stance,
John
ended up with
0
instead of 0.
However, since
Connor
blindly accepted
Philip Morris
‘s tokens at face value, fully half of the token value
Connor
received went up in smoke.
Robin
meanwhile got paid by helping to resolve the disagreement.
This is the magic of the claim market: disagreement creates independent funding for research.
Notice that
Robin
‘s account became worth
0
, because
Robin
‘s bounty was itself an intrustment in the direction the claim was resolved. This means that if in the future someone were to resolve the claim in the opposite direction,
Robin
would be the one to take the hit, giving
Robin
an incentive to resolve in the same direction that other future verifiers will also resolve the claim.
But if the only thing that can happen from adopting this currency is that you can lose money if you’re wrong, why buy in?

Why Buy In

It’s evident why
Philip Morris
bought in: had they not, the world would have laughed at their obvious insincerity. However, it’s not clear why anyone else would buy into the market.
Let’s spend some more time with
Richard
, the anti tobacco expert who appeared on TV.
Richard
‘s interest in tobacco was personal. Three years preceding that triumphal moment on TV,
Richard
had sat at the side of a hospital bed, holding a frail hand in a very different state of mind.
Richard
‘s partner had been battling lung cancer for years. Neither of them knew it at the time, but these would be their final moments together.
Coincidentally, as they sat in that silence, a red and white Marlboro truck flitted past the open window along the ribbon of asphalt below.
Richard
couldn’t help but suddenly think of the strange cancer results, Adele, a geologist friend had shared.
Somehow after a masters in geology, Adele had gotten roped into a tobacco related postdoc. In their small scale experiment, her team had suspended tobacco smoke in acetone and painted it onto mice skin. Within 50 weeks, the mice developed carcinogenic skin lesions. Meanwhile, the mice that were painted with just acetone — no smoke — were in perfect health.
Richard
‘s partner had smoked every day of their relationship. Could this ubiquitous product, which was advertised as “the most smoked by doctors”, have taken
Richard
‘s partner?
This became
Richard
‘s meaning to make of the loss.
Richard
joined Adele’s team of scientists studying the effects of cigarettes. Together they ran a much larger version of the experiment and, three years later, on December 1, 1953, was published in the journal of Cancer Research. Not coincidentally, 15 days later four Armani three pieces hosting lacerated larynxes met at the Plaza Hotel in New York City.
Since the day of goodbye,
Richard
had never once accepted Tritcoin (or any other fungible currency) as payment.
Richard
never wanted to be paid with a single dollar that credited the idea that cigarettes are benign. That’s why
Richard
insisted on being paid via the claim market. Sometimes that got expensive. But that was the cost of holding sincere beliefs.

Sincere Beliefs

Richard
wasn’t the only one to demand payment in claim market tokens. For many people, it was essentially the only way to live in accordance with their beliefs.
It enabled to them avoid inadvertently financially supporting beliefs they thought were false.
Sustainability advocates priced claims like, “The ozone layer isn’t important,” and “Smog is benign,” and “Anthropogenic climate change is not happening.”
Animal welfare advocates priced a litany of claims, including a series of the increasingly specific claims, “Animals can’t suffer” and “Mammals can’t suffer” and “Pigs can’t suffer.”
Human rights advocates priced claims like, “Cobalt mining is safe” and “The diamond trade does not cause human suffering” and “Nike doesn’t use child labor.”
With every percent they increased the price of their stance, it became more expensive for the other person to pay them.
And the more of the priced claim the other person held, the payment got more expensive, still.
Soon enough, however, activists discovered that there was more to specifying their stance than just avoiding the risk of owning a particular false claim.
Activists of all flavors soon found that much like a rock in a creek changes the water upstream, just setting a stance for themselves had impacts on the stances of those that paid them.
When
Richard
priced the claim
Cigarettes are benign
,
Richard
‘s employer noticed that the payment size to
Richard
had increased. They could, of course, have negotiated to pay
Richard
less because it cost them more, but there was an easier alternative. Given
Richard
‘s background, the idea that
Cigarettes are not benign
was believable, so they took the same stance as
Richard
. Now, whenever someone else paid them with tokens containing that priced claim that paying party would have to pay a bit more, and
Richard
‘s employer didn’t have to change their payment size to
Richard
.
But then, the people that paid
Richard
‘s employers also thought the stance
Cigarettes are not benign
was reasonable, so they priced the claim, too. And so thanks to
Richard
‘s small risk to price the claim
Cigarettes are benign
, the stance quickly propagated throughout the market. Soon enough, it became a stance that many people held.
Unlike a stock, people weren’t taking the stance because they thought other people would buy in which would cause the stock would go up in value. They were taking the stance because they thought other people might also take the stance, so by pricing it they were avoiding holding a currency whose purchasing power could go down.
Keeping up with the changing stances of the claim market was essentially impossible for
Richard
. There was too much going on, apart from manually setting the stance on
Cigarettes are benign
,
Richard
didn’t carefully set a stance for every possible belief. Instead,
Richard
followed the stance profile of other accounts. This caused
Richard
‘s profile to reflect the stances of trusted accounts for all the stances
Richard
felt ambivalently towards.
Popular accounts amassed huge followings. Sometimes the claims an account priced were specific to a topic, other times they spanned multiple domains. Communities could form by following one another and updating their stances together. Meanwhile, each individual was always incentivized to be careful what claims they accepted and who they followed, since there was always the possibility they would resolve to 0 if they were wrong.
The claim
Cigarettes are benign
was the kind of claim that was constituted of many sub-claims that could be found to be wrong. Because of that, it was risky to hold due to the possibility that it could resolve against the holder, causing the holder to lose value. However, for claims that were unlikely to resolve, there was no reason to price them. A claim that had no empirical way to find out it was wrong was unlikely to resolve.
The claim
God does not exist
was rarely priced. After all, how would one falsify that? And so it easily passed hands among most people. Of course, there were some people that did price
God does not exist
, and others that heavily priced
God does exist
, but that meant that everyone that wanted to pay them had to pay more to cover the price of disagreement, which meant that the people with those stances ended up having to negotiate their total payment down, meaning they were effectively paid less. Whenever the people that priced the claim
God does not exist
and those that priced
God does exist
tried to pay one another, their disagreement cost skyrocketed. Their unresolvable intolerance had literally become expensive. Furthermore, they had every incentive to soften their stance, since the tokens they were pricing were unlikely to change in value anyway.
Ok, but then how do we actually define what makes a claim resolve?

Claim Resolution

How does claim relevance and resolution work? I can only offer the three greatest words in the English language:
I don’t know.
If you have some thoughts, tell me

From a high level, here are the problems to solve:
Verifier markets. How does a verifier get selected?
Claim relevance. How can related claims be made dependent on one another?
Resolvable claims. How can claim makers be incented to create resolvable claims?

While these are all challenging problems, notice that they’re already problems that we have had to solve in our world as is. Maybe you’re used to calling verifiers journalists, or jurors, or special counsellors, or election officials, or researchers. Maybe you think of relevance as “probative value” from law, or as covariance in stats, or shared variables from philosophy, or a causal edge from causality.
The point is, these ideas have been explored, and we know ways that work reasonably well.
A worst case scenario for a viable system is that we’ll have all the benefits of a system for sincerity like a claim market that’s only as good at solving these problems as existing institutions. Or maybe (probably) there are way better ways yet to be deployed. By the way, what are those benefits?

Benefits

In this story, we saw how a Claim Market creates the context for:
A signal of sincerity of belief that can be distinguished from deceptive tactics (for
Philip Morris
)
A payment method that makes bribes less effective (with
Connor
and
John
)
A source of independent funding for research and journalism (with
Robin
)
A means of self-organization around shared values (for
Richard
and other advocates)
An incentive to avoid unresolvable disagreements (like the existence of God)
A way to turn the heat of disagreement into productive insight (thanks to disagreement funding and verifiers)
A means of rendering the scientific process legible and transparent (thanks to collections of claims)
An actually beneficial usecase for crypto

This matters because in order for a governance systems like democracy to work, citizens must have access to good information. Right now, the power profitability of inauthenticity is high — to an extent that some would even seek to limit speech to curtail deceit. But that will only make things worse. Instead, let’s create a context wherein disagreement can cut the path for understanding.

You now have a choice, you can go read or play with your own scenarios in .
This is a fully functional toy model, so if you have any questions about implementation you can copy the doc and look at the underlying formulas.

Want to get updates as we go?

Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.