Programmable Legal Entities and the Decline of Primeval Conservatism in Governance


  1. The super-progressive Maltese regulation on distributed ledger technology (DLT), among other things, presents programmable legal entities (PLE).
  2. The new type of legal entities should eventually grow to be way more complex than ordinary corporate structures.
  3. That super-complexity will induce new quality: meta-PLEs (large interconnected clusters of PLE-based organisations) which may significantly reduce conformism and conservatism in management. We’ve run a computer simulation that supports this conclusion.

Super-progressive Blockchain-related Regulation of Malta — Among Other Things — Presents Software-based Legal Entities, a First in History

Disclaimer: the author is not a legal advisor, his understanding of legal implications of the new regulations may be wrong. The purpose of this text is to share a hypothesis of a very broad nature, far beyond particular current legal definitions.

Malta’s legislative move toward understanding and regulating technologies included under the umbrella terms of “blockchain” and “crypto” turns out surprisingly logical. There are three sequential bills (the last one is delayed) that construct the natural progression of an attentive and unprejudiced stream of thought.

First comes the Digital Innovation Authority Bill that defines who’s in charge. Its very existence is a huge step forward: the fact that it has to be a completely new authority for some reason isn’t obvious to lawmakers in other countries. And it’s not only about being innovative — tense and unnatural fastening of what is clearly a new phenomenon to old institutions and definitions is dangerous.

Is bitcoin a commodity? Is it a currency?” or “Are ICOs security offerings?” are not only inappropriate, misleading questions; they are inherently harmful narratives that waste precious time for the entire civilisation. ICO founders play hide and seek with security authorities, while the overwhelming majority should simply be prosecuted for banal thefts, with no association whatsoever to any respectful backgrounds, such as securities markets.

On the other side, Bitcoin — as both the movement and as the technology — would probably be better off if it was completely free from any centralised influence, weirdly regulated exchanges and, even more so, futures trading — all of which receive access to the field through false legislative definitions. Had it been free without a retrograde institutionalisation, Bitcoin would have probably already formed many sustainable closed-loop economies throughout the world.

Admitting by law that crypto is nothing like anything we have had before and that the newly created authority has to regulate it is truly a great progress.

Second comes the Technology Arrangement and Service Providers Bill that regulates new types of legal entities (distributed platforms and smart contracts) and a new type of software administrators.

It is important to understand the core of this development. The whole smart contract idea has been the talk of the town for a few years now, but the phrase remains an oxymoron. Smart contracts are not contracts because no one is forced to follow them. Some limited classes of automatically performing deeds (“Send me some Ether, I will send you the doc.pdf”) can not build a business fabric. And you probably can’t make smart contracts legal contracts, even if you try to build up some regulation. At the end of the day, a conflict will always require a human judge to find a resolution.

But in making a software bot a legally acting body, a legal entity, is a completely different story. A company’s voters are now accounts filled with tokens, not humans directly. Humans will remain behind some of the accounts, but most of the simple managerial work will eventually be automated. Such “legal bots” don’t have to be perfect. They can make mistakes, just as normal companies do. Normal contracts supported by normal judges will keep things from becoming chaotic, just as they do today.

Although being #2 in the overall logic, the Technology Arrangement and Service Providers Bill has broader implications and attracts much more attention than the first one. New types of legal entities don’t emerge often; in fact, it probably only happens once per millenia or so, on average, so this bill is a really big deal.

The third piece, Virtual Currencies Bill, is set to regulate the tokenised value flow. This bill will without a doubt get the most attention because it directly deals with assets and cryptocurrencies. Brokers, exchanges, wallet providers, advisors, wealth managers, and other human actors fall under this piece of regulation.

However, I suppose it’s not human actors who will create the most far-reaching consequences of this new regulation.

Legally capable DAOs (distributed autonomous organizations) are, on one hand, only a part of what the Maltese legislators are bringing in today; on the other hand, this phenomenon may become disproportionately more influential than cryptocurrencies, blockchain or the entire global finance itself.

Eventually, DAOs development may lead to a restructuring of the most shameful aspect of today’s civilization: human management systems (hierarchies, democracy, etc.), which still fundamentally differ little from those used by primates or wolves, being heavily influenced by instincts and hormonal background.

Let’s take a closer look at what it could technically mean when some “Technology Arrangements” become real legal entities.

Programmable Legal Entities will Become Way More Complex than Incumbent Corporate Structures

Here’s what an ordinary human-managed organisation structure may look like; it rarely gets more complex than this:

In incumbent structures, every new functional node or a specific organisational “patch” is associated with high fixed costs and an additional possible loss of stability.

Unlike that, human-less or semi-automated software-based legal entities may develop to a much higher degree of complexity because a DAO can recursively co-own and co-manage other DAOs with ease and next to zero maintenance costs. Such structures will keep getting more and more complex with time as administrators will add patches to solve immediate problems, being naturally too lazy or short of time to reorganise and simplify. Since distributed apps are often focused on compatibility, DAOs will be connected to multiple services such as Colony, Aragon, district0x, etc. Possible inefficiencies and conflicts between layers will be solved with even more bridges and fixes.

Will that be a problem? Most probably not. Unlike structures that require humans, hardware and software can be build up into towers of patches and add-ons, and the system may still function great. Here’s an artifact at the Mercedes-Benz museum in Stuttgart.

A typical modern car is still essentially nothing more but a few thousand non-principal improvements, add-ons, and patches to this one hundred years concept. Despite having countless number of details, modern cars run just fine — in fact, substantially better than the original platform.

Complex Meta-DAOs can Defeat Strong Enemies of Reasonable Management

Quantity and complexity of components often eventually turn into a new system quality. A transition from micro to macro may be gradual and shockless. The world will be more and more reliant on the use of software entities to control the behavior of real-world resources. Most tasks share a similar structure: multiple parties come together to coordinate their activities in order to achieve a common goal, so a lot of research will be invested into the evolution of competing entity mechanics and new processes of coordination. A new type of logic will emerge to approximate reasoning, information granulation, linguistic computing, and other management related disciplines.

But will it lead to loss of control by the original beneficiaries? Yes and no — it depends what we mean by “control”. The vague definition is “to make things happen the way I want”; however, the problem is that people make most “decisions” only to activate (satisfy) neurons of the brain related to pleasure.

The strong enemy of reasonable management in particular and the progress in general is a primeval mix of conformism and conservatism. Conformism is biologically justified: even if you understand that you are being manipulated, you are unknowingly obeying. Conservatism (backwardness) is also a scary imminence being innate for a disproportionately large fraction of the population.

Sensitivity and vulnerability of “conservatives-in-mind” is a consequence of a prevailing negative attitude. “The ugly” always gets the highest priority in a conservative brain because any deviation from “the norm” is a signal about a problem that needs to be fixed. This thinking mechanism seems to have been anchored by evolution over the long 2.5 million years of the Pleistocene, when maintaining the old order of things, circular defense, and constant vigilance gave more advantages in the struggle to survive than, for example, ingenuity. The Pleistocene ended a mere 12 thousand years ago. About the same time humans domesticated dogs — live alarms — but the genes responsible for the ability to alert at each rustle have not disappeared.

Here’s our main assumption: DAOs can do exactly what dogs have done for us 12 thousand years ago— take care of the conservative part of management.

But how?

And how do we know that problem preventing and solving will be a much more popular application for DAOs than discovering new frontiers? And, regardless of preconceived evaluation by humans, why would DAOs be better at defence rather than offence?

Of course, no one knows a thing for sure as of yet. At this point in time, it can only be computer simulated; no actual experiments are possible.

To simulate this, my colleagues and I had to rethink the DAO concept, split the phenomenon into atoms, and come up with a handy definition of an elemental (“atomic”) DAO. We then considered a simplest possible example of a supply chain in which atomic DAOs operate (the slides only show in desktop browsers).

As far as a DAO definition is concerned, one of the first descriptions was given on the Ethereum website. Stripped of its context, that definition of DAO appears to be a needless election extension. Situated amid the dynamics that inspired it — token-based business calculus — DAO becomes a full-scale attempt to undermine many traditional inefficiencies of a centralised oligopoly economy.

The voting aspect itself is not directly relevant: typical platform-native DAO smart contracts are too primitive for practical applications, while the growing number of third-party projects targeting to deploy smarter voting still have a great distance to catch up with the well-developed voting theory and deeply analyzed practice.

We should rather consider a DAO as a bot, capable of performing a limited number of pre-arranged business activities. For example, think of it as a forklift driver with his machinery, instructions, the sea port navigation map, and some simple paperwork. A whole DAO, as it is broadly understood, is equal to many such congruently acting atomic DAOs. This atomic DAO unit reflects the core business processes better and is a handier object to study.

An atomic DAO can be considered a simple neuroplasticity-enabled mechanism with tokens used as neurotransmitters.

First of all, of course, an atomic DAO operates solely on tokens — that’s the whole point of DAOs. Tokens are like batons used in relay races. Every business transaction can be symbolised, verified, and done with the use of a specific, authentic token. Passing tokens normally does not involve a third party, but it may if there’s a specific need (escrow, for example, if joint-escrow is not an acceptable option).

Tokens carry value, access keys, signals (i.e. powered by IoT sensors, prediction markets, curated lists, or specialized “oracles”), and proof-of-goods (certificates). Tokens can be composite: for example, a certificate token can have payment coupons attached to allow for transportation. Tokens allow conditioned sending: for example, tokens sent from A to B arrive if B and/or A has sufficient balance of another token C.

If tokens can be negative (and they can), then the simple summation is a sufficient and reliable option for a main engine: DAO activation comes after enough inputs. To make it realistic, “dripping” should be applied — some value constantly dissipates (think fixed costs, for example). Binary locks are also a necessary element: some things can’t happen without signals, some signals provoke events that normally don’t happen.

To be able to optimise processes, DAOs should use weight adjustments when accepting tokens from other DAOs: they should perform a [delayed] comparison of the latest parameters with past outcomes and some sort of “industry standards”. By the way, voting, as it is understood in the sense of a “big DAO”, is simply a forceful readjustment of those weights for a class of atomic DAOs.

What are industry standards? The thing is no single DAO can “see” the big picture. There is no final judge and/or judgement of what is good and what is bad, and for what particular DAO. There can be no feedback loop from the “last layer”, as it is often implemented in neural networks; however, in each sector there must be some industry standards so that each DAO can estimate its relative effectiveness. Of course, a human administrator may play the role of an almighty god, but in a system complex enough, that would quickly prove inefficient.

So, we’ve run a number of simple simulations and our best bet is that if applied to real life systems, DAOs will mostly adopt behaviour schemes of a continually working pattern until a change is really needed, which would then be a prevailing norm.

In other words, there are no efficient incentive schemes that need to deploy DAO-bot workaholism. On the contrary, DAOs should be allowed to fail in a comparatively easy way, including the mechanisms to die off from inactivity.

The simulation has the normal set of reasonable scarcity demands (land for crops, coastline for ports, population for sales). It doesn’t consider price-related market (demand and supply) forces and focuses solely on logistics. The important difference from a regular setup is the ease with which a new business element can be installed or killed.

As a result, we could see some reasonable bio- and business-realism throughout the entire experiment: demand deficit vectors tend to frequently flip, system actors tend to have properties of laziness, mortality, habituation.


The problem of poor efficiency in supply chains and general high friction in business is not in the lack of trust between two business parties as most of the blockchain crowd preaches.

The problem is that we can’t trust ourselves to make many classes of management decisions being overloaded with outdated genes. Neither can we trust the Internet as the communication channel since it has brought an unavoidable third party — a cloud host — to each and every transaction. These third parties turn into monopolies, in spaces where there hasn’t even been room for a monopoly before; they end up creating monstrous problems in many spheres of life not directly related to business.

Software-based legal entities as described in the new Maltese Technology Arrangement and Service Providers Bill and set up as “atomic” DAOs acting as neurons with crypto-tokens acting as neurotransmitters are the real, usable means to:

  1. arrange practical, true peer-to-peer interaction for businesses. It is the first time since the broad adoption of the Internet that we can anticipate breaking free from unavoidable intermediaries (which tend to become monopolies).
  2. free human management from the entire class of conservative tasks and allow people to concentrate at inventing and forward-thinking. It may even develop to reshape the global politics.