Crypto’s Founding Fallacy
How mistakes in the “smart contract” genesis block weaken the whole chain.
Regardless of how strong or immutable their code, today’s “smart contracts” (and, by extension, entire crypto ecosystems) are built on a shaky conceptual foundation.
Crypto’s founding fallacy is over-reliance on a doctrinally-flawed conception of contract and law advanced by an early crypto theorist — Nick Szabo.
Some key features and consequences of this fallacy include:
Thesis A: Nick Szabo’s understanding of contract rests on a misunderstanding of contract formation and enforcement basics, such as:
(A)(1) impossibility of “coding around” third party enforcement;
(A)(2) usage of legal forms to create alegal processes (which actually results in legal relations that are orders of magnitude more complex than traditional contracts, while offering a fraction of traditional contract protections & benefits)
(A)(3) incorrect usage of Legalese to “write your own law” results in more, not less, legal scrutiny (→ more, not less, transaction costs).
Thesis B: Crypto’s reliance on Szabo’s “smart contract” theories & nomenclature harms crypto far more than it helps crypto.
Thesis C: A better, cheaper, more secure path forward is available: simply embracing legal indeterminacy.
1. Goodbye “Smart Contracts”
A good starting point to capture the turmoil around “smart contracts” is this exchange, just a few months ago—
The fact that crypto theorists are getting woke to the messy reality of “smart contract” legal theory is very encouraging. This heightens the likelihood that regulators and other stakeholders will tread more carefully, too.
2. Goodbye #SzaboLaw
Most recently, Vlad Zamfir wrote a broader repudiation of Szabo’s approaches to CryptoLaw. It’s a complex but logical and readable argument, so we encourage everyone to get familiar with it.
Here it is:
Zamfir identifies several different fallacies in what he defines as “Szabo’s law” (this is an initial read/restatement of Zamfir’s position — Zamfir himself doesn’t call SzaboLaw a “founding fallacy,” offering a more blunt assessment instead).
Szabo’s law is simple: Do not implement changes to the blockchain protocol unless the changes are required for the purpose of technical maintenance.
There are many ways to think about “smart contract” vis-a-vis “SzaboLaw.” The most immediately useful is to understand both as manifestations of a more general Szaboian theory of law and state.
In Zamfir’s framing, SzaboLaw is simply the legal foundation for immutable smart contracts running on immutable blockchains. There’s a catch, however. The only way global public blockchains can accomplish a fully immutabilist objective is to maintain an inherently unstable and unsustainable (& ultimately futile) anti-legal/alegal stance.
As Zamfir points out, SzaboLaw has been a useful heuristic for highlighting bigger stakes in blockchain global governance debates. For instance, it has allowed us to articulate the authoritarian threats posed by strict immutabilist ideology much better.
We can ask better questions, such as: can/should autonomous software systems (like truly immutable blockchains) interface with existing/legacy dispute resolution protocols (like, Law) or should they continue to strive for sui generis autonomous and alegal posture with respect to the aforementioned legacy systems? In either case, how?
3. Hello, Law is Politics!
There are no clear answers to the questions above, except perhaps for … Politics. Buterin may call this a platitude; but the Law = Politics point is a powerful standalone tool because it allows us to identify and study emerging trends in blockchain governance. Concretely, it allows us to see how CryptoLaw is used by actors to pursue particular political objectives.
The tenor of the immutabilist response to Zamfir’s critiques shows the critiques struck a nerve. Perhaps immutability-proselytizing maximalists are starting to realize that the state (through Law) is far more decentralized and adaptable than many crypto networks, and so they shoot the messenger who brings the news they do not want to hear.
This is exactly why we need better analysis on the implications of Szabo’s thought on law and state. This is especially vital given that Szabo seems to be pursuing politics of censorship against critics. CleanApp is doing its part to document Szabo’s theoretical and doctrinal infelicities in this context, but the gospel of legal immutability can only be dispelled with community effort.
It is good to see leading crypto theorists take part in this debate. Vitalik Buterin’s responses outline many specific potential examples of crypto legal tension and suggest fruitful avenues for research. We added a word, or two.
Zamfir’s critique is also trending on #CryptoTwitter at the highest levels.
We will unpack more of Zamfir’s argument in the days and weeks ahead. For now, heeding Vitalik Buterin’s call to identify specific tensions between SzaboLaw and Law, we will narrow the focus here to smart contract theory.
But before diving in, we’ll restate (yet again) why we’re intervening in this debate.
4. Fallacy Spotting
Analysis that tries to explore any social project’s core assumptions is, by definition, controversial.
The architects and beneficiaries of the project under analysis have much to lose from admitting their project is built on sand.
But everyone knows that a house buidl on sand will not stand.
Everyone should see the benefit of spotting problem areas as soon as possible. Exploring blind spots in one’s argument may not be pleasant, but it is critically important. Fortunately, whether illegal dumpsites or crypto jurisdictional lacunae, spotting problem areas is what CleanApp is all about.
While we get that everyone gets queasy when core assumptions get shaken, you must understand that our impulse for shaking the crypto tree is an existential one.
CleanApp needs robust global blockchain markets in order to improve global resource use outcomes. Thus, we have invested heavily in accelerating crypto’s global scaling success. In many ways, our success hinges on the success of the broader crypto enterprise. Therefore, our goal in issue spotting is to resolve minor problems before they get bigger. Our analysis does not attack crypto; it offers suggestions for improvement.
CleanApp’s stance in crypto governance debates isn’t concern trolling or virtue signaling: it is basic strategic risk mitigation.
Separately, prior to proceeding, please note that building this type of analysis is extremely time-consuming and painstaking work.
Therefore, before generalizing (or responding with ad hominem attacks or ascribing improper motivations, etc.) we ask that you please familiarize yourself with the bigger analytical project of which this is just one part.
Lastly, we ask that readers exercise baseline respect and discretion for the pro bono nature of this undertaking, and, due to CleanApp’s independent status, the rare lack of conflicts of interests. Both of these give CleanApp analysts a valuable insider-outsider perspective on crypto. If you see the importance of this type of analysis, please reward it with claps and shares so that we can grow our reach.
If you disagree with all or part of our argument, please do so in good faith.
Now a deep breath, and, let’s dive in.
5. Narrowing Scope & Defining Problem
In the past, our arguments against “smart contract” terminology (& here, here, here, here, etc.) stemmed from empirical observations.
Usage of “smart contract” terminology only sows (1) chaos, (2) incoherence, (3) unnecessary complexity, (4) regulatory blowback, (5) added costs, (6) confusion, and (7) reduced innovation.
But in 2019, it is no longer acceptable to blame legislators or “ignorant people” (Szabo’s phrasing) for their misunderstanding of “smart contracts.” The problems with “smart contracts” (SCs) and #CryptoLaw go far beyond terminology.
To see why there’s so much confusion and blowback against SCs, we need to delve deeper into SC genesis blocks, such as:
Concretely, we must reverse-engineer a “smart contract” from its techy noun usage all the way down to its theoretical and legal base forms — as conceived by Szabo. This is key because Szabo’s doctrinally-flawed presentation of contract law is crypto’s founding fallacy.
6. Auditing the “Smart Contract” Chain
Block 8: Usage of “Smart Contract” (2019)
Here’s how the term “smart contract” is typically used today in the broad Blockchain/DLT/crypto community.
Smart contract (noun): a computer script that allows two or more parties to form a self-enforceable agreement; a computerized protocol that executes the terms of a contract; a blockchain script that permits automation and verification of contractual performances; etc.
Smart contract (verb): to enter into a smart contract.
Block 7: Ethereum Yellowpaper (2014)
The Ethereum Yellowpaper (authored by Gavin Wood, one of Ethereum’s co-founders) gives a more formal intro to “smart contracts.” The Yellowpaper is also the constitutive act for the very field of CryptoLaw, so it’s important and worth a read in its entirety.
For purposes of tracing the intellectual roots of “smart contracts,” the Yellowpaper teaches:
Early work on smart contracts has been done by Szabo [1997] and Miller [1997]. Around the 1990s it became clear that algorithmic enforcement of agreements could become a significant force in human cooperation. Though no specific system was proposed to implement such a system, it was proposed that the future of law would be heavily affected by such systems. In this light, Ethereum may be seen as a general implementation of such a crypto-law system.
Wood also offers this technical Ethereum-specific definition of contract—
Contract: Informal term used to mean both a piece of Ethereum Virtual Machine Code that may be associated with an Account or an Autonomous Object.
The Ethereum Yellowpaper’s building block is the Ethereum Whitepaper.
Block 6: Ethereum Whitepaper (2013)
Vitalik Buterin’s Ethereum’s Whitepaper is titled: A Next-Generation Smart Contract and Decentralized Application Platform (2013).¹ With 129 mentions of ‘contract,’ the paper is replete with “smart contract” usage along two paths: (1) legal; (2) alegal.
Smart Contracts = legal. On the one hand, the Ethereum Whitepaper expressly envisions “smart contracts” serving traditional legal functions:
[A decentralized autonomous organization, or DAO] essentially replicates the legal trappings of a traditional company or nonprofit but using only cryptographic blockchain technology for enforcement.
Smart Contracts = alegal. On the other hand, Buterin intuited that the ‘contract’ part of “smart contracts” would cause confusion, so much so that he expressly qualified usage of “contract” terminology as follows:
Note that “contracts” in Ethereum should not be seen as something that should be “fulfilled” or “complied with”; rather, they are more like “autonomous agents” that live inside of the Ethereum execution environment,
Buterin’s “smart contract” terminology in the Whitepaper is fascinating because of the delicate way he balances between legal and alegal usage. It’s even more fascinating in light of earlier Ethereum Whitepaper versions.
Block 5: Ethereum Whitepaper (earlier version)
Here’s how the Ethereum Whitepaper described contract in an earlier version:
A contract is essentially an automated agent that lives on the Ethereum network, has an Ethereum address and balance, and can send and receive transactions. A contract is “activated” every time someone sends a transaction to it, at which point it runs its code, perhaps modifying its internal state or even sending some transactions, and then shuts down.
Echoing the delicate balance between legal & alegal postures in subsequent versions of the Whitepaper, Yellowpaper, (and Polkadot Paper, and so on) Buterin was clearly conscious about contract enforcement needs — and Ethereum’s potential inability to satisfy those needs.
Buterin clarified:
Note that financial contracts of any form do need to be fully collateralized; the Ethereum network controls no enforcement agency and cannot collect debt.
A creative CryptoLawyer, Buterin left broader questions of enforcement relatively open. But it’s important to observe how legal ambivalence and indeterminacy structure Buterin’s legal thought.
On the one hand, Buterin creates alegal constructs (the “Ethereum network controls no [law] enforcement agency”). On the other hand, the “financial contracts of any form” that are “fully collateralized” are clearly contemplated as binding, enforceable, legal forms — including potential integration with existing financial instruments and legacy institutions.
Block 4: Szabo (1997) & Miller (1997)
You’ll recall that Gavin Wood drew inspiration from early work on smart contracts by Szabo (1997) and Miller (1997). Here are the full references:
- Nick Szabo, Formalizing and Securing Relationships on Public Networks (1997).
- Mark S. Miller, Computer Security as the Future of Law (1997).
Here’s Szabo (1997):
Here’s Miller (1997):
Because Miller (1997) expressly relies on Szabo, we’ll conflate the two for purposes of this argument, and proceed to earlier blocks.²
Block 3: Szabo (1996)
Szabo wrote this in 1996, but this is a recent screenshot:
Block 2: Szabo (1994)
Block 1: Szabo’s Bit Gold (“a long time ago”)
Here’s Szabo writing on December 27, 2008, two months after the publication of the Bitcoin Whitepaper.
A long time ago I hit upon the idea of bit gold. The problem, in a nutshell, is that our money currently depends on trust in a third party for its value. As many inflationary and hyperinflationary episodes during the 20th century demonstrated, this is not an ideal state of affairs. Similarly, private bank note issue, while it had various advantages as well as disadvantages, similarly depended on a trusted third party.
Interesting. We’ll return to this bit of gold shortly.
Block 0: Szabo’s Legal Training (2003–2006)
To the extent there’s an earlier “Block 0” that can shed light on Szabo’s thinking, it’d be something like the contract law books that Szabo was expected to read in law school.
But Szabo doesn’t cite contract law casebooks or treatises, forcing readers to reconstruct where he might have picked up his views on contract. For instance:
Even though he does not cite contract law theory, we can still draw inferences about Szabo’s Block 0 from his exposure to a standard American law school curriculum during his time at George Washington University, School of Law (2003–2006).
(The only block more formative and consequential would be the block where Szabo learned language. The significance of this will become apparent shortly.)
Now that we’ve linked the chain of conceptual blocks, let’s audit the blockchain to see how mistakes in genesis blocks lead to costly forks down to the road.
7. What is a Contract?
Here is what Szabo was expected to learn in law school about contract as a legal form:
A contract is a legally enforceable agreement between two or more parties.
A contract is not, as Szabo suggests, a “set of promises agreed to in a ‘meeting of the minds.’” (Hereafter, ‘meeting of the minds’ = MOM).
Although the difference between these two definitions may not be immediately apparent to non-law people, the legal difference between these two positions is like night and day, oil and vinegar, 1s & 0s.
The definition up top is settled law, globally. Szabo’s definition, by contrast, is doctrinally, conceptually, legally, etc. — dead wrong.
(7)(A) Subjective Theory of K
Szabo’s definition of contract rests on what’s known as the subjective theory of contract formation (and also, contract interpretation, on which more, later). The subjective MOM theory suggests that we can somehow look inside the parties’ minds to decrypt mutual assent. And if the parties have MOM, and there’s some formal proof of their MOM, then voila, the parties have created a contract.
The subjective approach makes a lot of sense initially. It’s very intuitive, and it comports with how lay people generally understand contract formation processes.
The subjective approach has a big problem, however: it doesn’t work in real life. This is because parties can and often do lie about their mental intent after the fact. If MOM is the standard for contract formation (and interpretation), as Szabo suggests, then contract disputes quickly devolve to different versions of “he said, she said.” In evidentiary and dispute resolution efficiency terms, these he/she-said disputes are extremely costly.
To be fair, Szabo isn’t making up the subjective test from scratch. A few hundred years ago, MOM used to be the way English courts determined whether or not we had contracts (see, e.g., the classic case of the two ships Peerless).
But when folks saw that MOM was not reliable, we evolved to a far more efficient legal framework for determining whether or not there’s a contract: the so-called objective theory of contract formation — which is a general principle of contract law recognized globally since at least the middle of the 20th century.
Thus, from the moment he published his views through today, Szabo’s approach to contract has stood in conflict with existing law. We can call this the subjective SC problem (a subset of a broader SC self-enforcement problem). This is also one aspect of the SC legality problem (what Zamfir correctly identifies as a conflict between SzaboLaw and, ahem, Law).
(7)(B) Objective Theory of K
Under the objective theory of contract formation, to determine whether or not a contract exists, we look to the external manifestations made by the parties (their words &/or deeds) from the perspective of a neutral reasonable third party.
Because we learned that parties can easily lie about their contractual intent after the fact, we don’t pry open the parties’ minds and ask them whether they subjectively intended to enter into a given contract. Instead, we look to what a hypothetical reasonable person would say about the alleged contract parties’ words and actions.
If the imaginary reasonable person C would conclude “yeah, both A and B exchanged signs that suggest they wanted a contract,” then we may have one of the two elements of contract formation: mutual assent. (The other element is form, which in the Anglo-American tradition is known as consideration, and in civilian traditions is sometimes described as causa, etc.)
(For more contract formation basics, please see Contract Law Primer for Crypto/DLT.)
(7)(C) Third Party Involvement
The split between subjective and objective theories of contract formation becomes even more pronounced when we reintroduce the role of the third party whose involvement is indispensable for contract formation.
Among the many functions served by third parties (C, D, E, etc.) in the supposed contract relationship between A & B, third parties (*plural*) are the legal actors who determine whether or not a contract exists.
These third parties are present irrespective of whether we’re in 1820s London adjudicating a contract dispute by reference to the subjective theory of contract formation (SzaboLaw), or whether we’re crypto-arbitrating this dispute in 2020s Malta or Macau by reference to today’s more objective standard (Law).
The third party is always there in contract practice, as well as in the settled definition of contract the world over:
A contract is a legally enforceable agreement between two or more parties.
Third parties are the ones who make an agreement “legally enforceable” or not. This is because law en-force-ment is necessary to obtain remedies for breach of contract or other contractual disputes. And prior to dispatching legitimate force mechanisms to seize assets or throw breaching parties in jail, humans want multiple opportunities for third party review of whether there ever was a contract, whether it was breached, the extent of damages, if any, and so forth.
Importantly, third party involvement does not necessarily mean “courts” or “arbitrators” or “mediators” or other “neutrals.”
Third party involvement can be thought of as a set of baseline expectations that contracting parties bring to their contracting affairs.
Third party involvement answers basic operational questions of contract law, like: (1) who decides whether a K exists, and/or what it means, and/or whether it was breached; (2) what standard should be used to determine contract formation, interpretation, breach in a particular context; etc.
According to Szabo, parties can form (& enforce) contracts autonomously, without need for third parties. But that’s not true. We can call this the SC third party problem (also a subset of the broader SC self-enforcement problem).
(7)(D) Third Party + Objective Standards
As seen above, the objective theory of contract formation is simultaneously an invocation of third party mechanisms (e.g., the hypothetical reasonable third person who is examining the supposed contracting parties’ words and actions).
In operation, this often allows third party mechanisms to become minimized to the point of being imperceptible in day-to-day contracting affairs.
The possibility of third party involvement, in other words, is already folded into the objective theory of contract — which serves to channel the parties towards dispute resolution mechanisms that avoid actual third party involvement.
Think of the last time you received a refund for a purchase. Neither you nor the store probably invoked a “trusted third party” to iron out your differences; but the possibility of third party involvement is, in part, what channeled you and the store to resolve the dispute bilaterally.
Consciously or not, the reason you and the store likely felt comfortable reaching a casual contract settlement is because of the possibility of invoking other third party dispute resolution mechanisms if the new settlement failed to protect the parties’ reasonable expectations.
All this is well-studied and amply understood by contract theorists and economists. This is why nobody should be afraid of third party ghosts. They are in the ether all around us every single day as we go about our contracting lives. We know a fairly good deal about how these ghosts work, and they seem to work remarkably fine at micro- and macro- glocal scales.
Third party ghosts (C, D, E, etc.) lubricate, rather than stifle, commercial intercourse between the As & Bs of the world. Importantly, a major reason why existing third party mechanisms (*plural*) facilitate commercial exchange is because they are inherently decentralized. This is true irrespective of whether we’re talking about a consumer-grade (A ↔ B) online transaction or a complex multi-party instrument like, say, a letter of credit.
The objective theory of contract formation does not imply centralized control. Not even remotely.
Furthermore, even the subjective theory of contract formation was never about forcing parties into binary duels over whether or not there was MOM. Even in the ancient subjectivist times, contract disputes ultimately went to a third party (C) for the reason that (A) and (B) couldn’t sort out their MOM problems by themselves.
SCs that seek to lock parties into predetermined dispute resolution pathways (#CodeIsLaw) seem fixed and determinate, but this is an illusion. In reality, even in the most hardcore maximalist SC context on the most diehard immutabilist blockchain, contracting parties can always invoke off-chain remedial mechanisms to address what they perceive as an on-chain injustice.
This is true irrespective of the fact that the party or parties invoking the off-chain remedies now had agreed to fully on-chain remedies earlier. There is no way to block parties from invoking off-chain remedial mechanisms. We can call this the SC remedial problem.
(7)(E) Trust Minimization
Szabo’s contract theory asserts that SCs will “minimize” reliance on third party mechanisms. But despite being pressed further, Szabo cannot answer how SCs actually minimize reliance on third parties in the dispute resolution context.
Szabo has no theory of trust minimization, as such. Instead, what Szabo offers is hoped-for, fiat-like, trust minimization: “just switch to this SC, and you’ll see just how much more performance-incentivized you’ll become.” There’s no theoretical or technical engine for trusted third party minimization in Szaboian smart contract theory, nor can there be one, for the simple fact that third party involvement is inescapable in contract disputes.
It turns out that third party dispute resolution is not only inescapable; it is actually quite desirable. It has proven to be a highly efficient mechanism of channeling parties away from duels towards relatively more peaceful dispute resolution forums. (Please note, even if parties agree to randomize the dispute resolution mechanism and “opt-in” to coin-flip dispute resolution processes, either party is always free to “opt-out” of the coin-flip mechanism after the fact simply by arguing that the “opt-in” was invalid — SCs do nothing to change this reality).
The fact of the matter is that SCs are themselves third party mechanisms from the vantage point of SC contracting parties.
Much more can be said about how crypto efforts to eliminate trust in third parties has only introduced more need for trust in … third parties. But for now, it is enough to point out that an SC theory premised on the minimization of trust in third party mechanisms is prima facie incompatible with contract law and practice — smart or otherwise.
We can call this the SC trust problem.
8. Contract Enforcement Basics
Already, our audit of the genesis block reveals a lot of problems, such as: (1) subjective SC problem, (2) SC legality problem; (3) SC third party problem, (4) SC trust problem, and a broader (5) SC enforcement problem. These problems flow from a foundational doctrinal error — Szabo’s reliance on MOM as opposed to the objective theory of contract.
Yet it’s not just contract formation basics where Szabo is wrong. His anxiety about third-parties results in mistaken approaches to contract interpretation and contract enforcement, as well.
Here’s a quick chart, just to sharpen the contrast.
And here’s the SC enforcement picture, according to Szabo:
The problem with Szabo’s position here is that contract enforcement is literally the most important part of any contract theory, irrespective of whether we’re talking about “smart contracts” or contracts — by his own terms.
Wondering whether a ‘smart contract’ is legally enforceable is the whole point. It’s “100% of the law,” going back to Szabo’s fractional imaginaries of “9/10ths” & “99/100ths.” Nobody cares about “smart contracts” that are not enforceable. And en-force-ment = legal en-force-ment. There is no other kind. (The supporting argument is available here.)
Legal enforcement necessarily implies third party review; without it, the supposed legal system is nothing more than a system of self-help measures that quickly lose legitimacy and effectiveness.
Proof?
Humans had self-help mechanisms for most of our history. As we’ve evolved, we’ve learned to spigot and channel our disputes by invoking (even if it falls short of actually involving) third parties. The alternative is might-makes-right, which is patently unjust.
9. Doctrinal/Theoretical Basics
Please note, the (1) subjective SC problem, (2) SC legality problem; (3) SC third party problem, (4) SC trust problem, and the broader (5) SC enforcement problem are just a few of many fundamental mistakes in Szabo’s approach to contract.
We need to be crystal clear on this: Szabo’s SC “theory” is riddled with mistakes of a basic, foundational, terminological and conceptual nature.
Szabo isn’t just advancing unpopular, outdated, or “minority” doctrinal positions. Szabo’s writing on contract is not just wrong because it is contrary to accepted usage and established doctrine. Szabo’s SC theory is also flawed because:
- it it is objectively incoherent and, separately, because it is internally inconsistent (see above; see below).
- Szabo’s conception of SC radically restricts parties’ existing contract rights;
- an immutabilist/write-your-own-law SC paradigm preferences financial crypto instruments over general-purpose blockchain instruments that offer broader social utility.
Scale is everything. Compounded, the list of SC problems above is just the tip of the iceberg of legal problems that are traceable directly to the genesis blocks under audit here.
And again, a dApp buidl on sand won’t scale.
10. Good Faith & FRCP, Rule 11
We respect Szabo’s right to evangelize his unique flavor of subjective libertarian legal thought. But Szabo is not free to misrepresent the nature of fundamental socio-legal forms like contract.
As a U.S.-trained law person, Szabo owes duties of candor to the public at large. He can agitate for modification of existing law, like going back to, say, the subjective theory of “he-said/she-said” contract law. But he has to do so in good faith.
Good faith argumentation (legal or otherwise) means sticking to basics, which includes very careful usage of Legalese when advocating for change.
Szabo knows all this. Back in 2010, he stressed the importance of understanding basic legal concepts. Szabo even promised a series of blogs explaining legal basics. That effort never materialized, which is a big reason why we stepped in to offer clarity and a space for critical crypto legal thought.
Given that everyone agrees we need to get the fundamentals right, our own return to basics must capture a more fundamental layer of socio-legal exchange than even “property” or “contract” or “Law.” This means understanding how language works.
Szabo’s SC theory gets this wrong as well, leading to something we can call the SC language problem.
11. Language > Law
The foundation of all human relations — whether crypto, contract, and/or Law — is language. And language is inherently imprecise, indeterminate, and mutable.
To solve the supposedly vexing problem of contractual ambiguity, Szaboian SCs necessitate the creation of a universally objective contract “markup” language. Then parties will use this universally objective language to enter into self-enforcing contracts, where the likelihood of breach is minimized, and there’s no need to even think about third parties.
Except, like the mythical Esperanto, this objective contract language is easy to conjure up in one’s mind, but it’s really difficult, if not altogether impossible, to get other people to wrap their hands and tongues around.
The pull towards objectivity and clarity is totally human. But so is making mistakes.
As we go about building increasingly complex social structures, we strive to create more objective dialects for particular applications. Thus, we get things like the Uniform Building Code, Uniform Commercial Code, Napoleonic Code, Bankruptcy Codes, and so on. CryptoLegal equivalents include projects like OpenLaw, Szabo’s Formal Language for Analyzing Contracts (2002), Ethereum’s Solidity, and so on.
But no matter how smart the architects of a particular field, humans have not been able to transcend linguistic and conceptual ambiguity.
Exhibit A is litigation arising out of ultra-formalistic, hyper-objective, and ultra-rigid contracting spheres, like construction disputes, or financial disputes. If it was possible to create a universal lingua mercatoria, and then translate that language into universal lex mercatoria, humans would have done that a long time ago (given the extraordinarily high utility, productivity, and welfare gains this would theoretically unleash).
Humans keep trying to build universally objective systems of meaning, and law. Humans keep failing at this, because anything resembling a mandatory universal lex mercatoria is typically highly coercive and grossly inefficient. Szabo’s stab at A Formal Language for Analyzing Contracts failed because it also turned out to be highly coercive, centralized, and grossly inefficient.
As other teams continue to embark on the project of universal contract scripts, we should remember that the Napoleonic Code wasn’t just a project of enlightened thought, freely adopted by the nations of Europa and her “foreign domains” (e.g., colonies). Legal harmonization and universalization projects typically go hand-in-hand with hegemonic ambitions.
By contrast, the most effective socio-legal harmonization projects (i.e., consensus protocols) are those that incentivize adherence to a roughly shared vocabulary of rights and obligations, one that gives adherents maximum flexibility and individual freedom of action — in a non-coercive manner.
Socio-legal consensus-building protocols (like Law, as Zamfir correctly points out) require a shared commitment to consistency, predictability, and uniformity in baseline usage of Legalese (mindful that 100% objectively “correct” usage is simply impossible, for anyone, at any time).
12. (Bad) CryptoLegalese = (Bad) CryptoLaw
Szabo’s freestyle approach to law and to language renders his SC theory internally inconsistent and irreparably unstable. Here’s why:
All over the world, law students learn that Law, first and foremost, hinges on correct usage of Legalese. One’s stock as a law person hinges on this. The strength of one’s arguments hinge on this.
Correct usage of Legalese = 10/10ths of the Law.
The broader crypto community also knows the importance of getting the basics right. Again, whether built with C++ or UCC, a house buidl on sand will not stand.
With that in mind, it’s crucial to dissect and assess Szabo’s 1994 definition of “smart contract” to see why the mix-and-match is so conceptually unstable.
When one’s arguments in support of a self-contained, universal, objectivist, and immutable contract programming language and self-enforcement system suffer from patent and latent ambiguities as glaring as these, those faulty arguments serve as the strongest refutation of the feasibility of the underlying aims.
13. One, Too, … Many
By this point, we hope you’ve realized that the mistakes are not isolated or cherry-picked instances of conceptual chop-chop. When it comes to contract theory and the enforceability of a “smart contract,” the conceptual problems are found up and down the chain.
But as Zamfir and others have pointed out, the mistakes also go far beyond “smart contract” theory. These are major errors about the foundational nature, limits, and functions of contract — with profound theoretical, conceptual, and practical effect for CryptoLaw, generally.
Unfortunately, these are now mistakes that judges, juries, legislators, regulators, arbitrators, and others will spend thousands of hours decrypting — and trying to correct. And we’ve only scratched the surface.
As a closing example, we’ll revisit Block 4 in the SC blockchain above. Here it is again, Szabo (1997):
To casual readers, this may not raise red flags. But to analysts trained in formal logic and/or lawgic, the three sentences are a quintessential illustration of a conceptual dead-end:
- We’ve already covered the problems with MOM, so from the first sentence, good luck figuring out how “minds” get “formalized,” or what formal lawgic we’re supposedly tapping into.
- The second sentence is refuted by Szabo, who says that property, not contract, is the basic building block of a market economy (see Szabo’s Power Point slide above). (Lest anyone think this is “just semantics,” please recall that the property-contract-property nexus is one of the biggest theoretical battlefields in all of blockchain.)
- “[…] encoded into common law” is a patent contradiction in terms, like saying “CRISPR-ed into C++.”
When freestyle Legalese gets crystallized by the dogma of immutability, we get faulty SC conceptual blocks. In the real world, broken Legalese leads to greater uncertainty and higher overall transaction costs. The only people who benefit from more uncertainty along these lines are lawyers like Szabo.
14. Who’s to Blame?
To be fair, the problems with “smart contract” and CryptoLaw are not attributable solely to Szabo. The actual genesis blocks extend far below Szabo, and the genesis blocks of the most important blockchains to come extend far above Szabo.
In the context of today’s immutability battle, Vlad Zamfir has done crypto a service by broadening the aperture to include many others who at one point or another agitated for variations of the position that crypto is alegal and/or can/should operate outside of channels of state/legal control.
Yet even though the critiques here apply more broadly as well, Szabo is clearly a big actor in this drama. Therefore, it makes sense to continue the critique, especially with a view towards other facets of CryptoLaw, such as crypto legal personality, crypto property, jurisdiction, and the unanswered questions of legality, trust, and subjectivity.
The main task ahead is to magnify this audit. CleanApp will continue doing its part, but we cannot do this difficult work alone. We need your help. From now on, every word, every line of code, every bit of SC syntax needs to be inspected because everything that’s built on top of faulty code & bad assumptions will crumble. This is Law School 101, Risk Mitigation 101 & InfoSec 101.
It’s also Szabo 101 —
A K buidl on sand will not stand.
15. Conclusion
As politically costly as this has been in crypto circles, our intervention has actually been a direct response to Szabo’s own call for critique.
In 2002, when he outlined A Formal Language for Analyzing Contracts, Szabo said he welcomed participation:
Your suggestions for changing or adding more kinds of contractual terms to our language are quite welcome.
So that’s exactly what we’ve tried to do, and will continue doing.
In Why’s Szabo Afraid of “Smart Contract” Critiques? we showed how Szabo’s attempt to rewrite contract law radically restricts contract rights and opens many new threat vectors that can be exploited against developers, unwitting populations, and individual Liberty. We have shown how a “self-enforcement” conception of “smart contract” is actually doing far more harm than good for blockchain development.
Here’s Szabo again:
I’m probably making other assumptions I haven’t discovered yet — if you find any please let me know!
We found many. Answering Szabo’s call, we’re letting everyone know.
— — — — — — — — — — —
¹ The Ethereum Whitepaper that was published in 2013 has undergone internal revisions since. The Whitepaper has also been complemented by a series of subsequent technical papers (including Yellowpaper (2014), cited above, Mauve Paper (2016), Polkadot Paper (2016)). Each of these papers pushes “smart contract” usage in new directions, including Gavin Wood’s Polkadot Paper suggestion to move towards the development of a “contract language.” These more narrow uses are somewhat outside of the scope of the present analysis, but certainly demand much closer scrutiny.
² Some developers today emphasize a potential alternative Miller-ite approach to “smart contracts,” while noting that Miller & Szabo are most likely complementary frameworks.
