On November 28, 2023, Lee Cronin and James Tour had a debate regarding the origin of life research (Harvard Cambridge Faculty Roundtable). Tour argued that Origin of Life research is a scam that has made absolutely no progress in understanding how non-living things could give rise to living things via mindless natural mechanisms, while Cronin argued that real progress is being made.

Lee Cronin is the Regius Chair of Chemistry in the School of Chemistry at the University of Glasgow. He was elected to the Fellowship of the Royal Society of Edinburgh, the Royal Society of Chemistry, and appointed to the Regius Chair of Chemistry in 2013. His research interests include: Artificial Life, Chemical Evolution, Chemputation, Digital Chemistry, and Assembly Theory.

James Mitchell Tour is an American chemist and nanotechnologist. He is a professor of chemistry, materials science, and nanoengineering at Rice University in Houston, Texas. His research interests include: Chemistry, Materials, Electronics, Nanotechnology, Graphene, and the Origin of Life.

Assembly Theory:

At the debate, Tour argued that the assembly of the building blocks required for living things was too complex for any known mindless natural mechanisms to achieve, and that Origin of Life researchers, like Cronin, had made no meaningful progress.  Cronin countered with his Assembly Theory, arguing that Assembly Theory had made meaningful progress in being able to detect living things (when compared to non-living things) with a very high degree of accuracy, and could be used to understand the Origin of Life as well.

What is Assembly Theory?

So, what is Assembly Theory? Evidently, it can be confusing to understand since Tour himself initially had a hard time understanding it, as did a number of other scientists.  Computer scientist and logic expert, Hector Zenil, for example, argues strongly that Cronin’s Assembly Theory is nothing more than a “dictionary-based compression algorithm” that is commonly used by computer algorithms to compress files (Link, Link, Link) – specifically the LZ77 compression algorithm.  Basically, Assembly Theory is simply determining the “complexity” of a system in very much the same way as has historically already been done in computer science – by “Combining the power of Shannon Information/Entropy and Kolmogorov-Chaitin complexity by way of Solomonoff-Levin Probability” (Link).  The problem is that none of these measures of “complexity” is a measure of meaningful or functional complexity.

Consider, for example, the following sequences:

• “In the beginning God created the heavens and the Earth”
• “ax epm iwmxseigm uio wazirlf hgt pohmsat qwv hrg ouirm”

Both of these sequences have 54 characters each. Since both are composed of the same 26-letters of the English alphabet, plus a space, the amount of uncertainty eliminated by each character is fairly similar. The probability of producing each of these two sequences at random is also similar. Therefore, both sequences have a fairly equal amount of Shannon information – even though one is “meaningful” while the other is not. This means that the mathematical formula for Shannon information is, in fact, quite simple. It is: I = -log2p where p is the probability of a particular sequence, out of all the potential options, being realized. Obviously, then, given that all other things are the same, the Shannon “information” of a sequence will increase as the length of the sequence increases – regardless of “meaning” or “function” or a complete lack thereof. Combined with Kolmogorov/Chaitin complexity (KCC) of a sequence (a measure of its compressibility relative to a particular compression mechanism, such as a Universal Turing Machine or UTM), what you end up with is “Assembly Theory”. The claim of Cronin that Assembly Theory is actually fundamentally different than Shannon Entropy and Kolmogorov Complexity is effectively challenged by Zenil (Link).

Zenil holds two PhDs, one in Computer Science and one in Logic and Epistemology from Paris 1 Sorbonne/Ecole National Superiore, and Lille 1 in France. He is the Associate Professor/Senior Lecturer in Healthcare and Biomedical Engineering at King’s College London (KCL), where he leads and conducts research-develop technology at the intersection of information theory, complexity science, causal universal and beneficial AI at the Departments of Biomedical Computing and Digital Twins and the King’s Institute for AI.

See also Zenil’s fascinating blog on “The 8 Fallacies of Assembly Theory“

Only A Compression Algorithm?

The Element of Selection:

So, is it really true that Assembly Theory is nothing more than a compression algorithm?  Well, Onsi Fakhouri, an astrophysics researcher and tech executive, has an interesting take on this. In his discussion with James Tour, he noted that while Assembly Theory is a compression algorithm, it also adds the additional element of “selection” – the need for a “selector” to explain a given system or arrangement of building blocks (be it a living thing, an arrangement of poker cards, or Lego blocks).

Detecting the Existence of a Selector:

The detection of a need for a “selector” is detected in a very interesting manner.  Take a sequence of 1000 randomly generated letters.  Such a sequence likely has a high level of Shannon information or “entropy” and Kolmogorov complexity. However, so do most other randomly generated 1000-letter sequences. What makes a particular sequence “special” in a given environment?  Well, it would become very interesting if exactly the same sequence were realized over and over again (i.e., a high “copy number”).  This would indicate some kind of preferential “selection” taking place – a “selector” that prefers this particular sequence over all other sequences of similar size with otherwise equal Kolmogorov complexity.  And, the validity of this conclusion has been successfully tested and demonstrated by Cronin (Link). In fact, Cronin uses his Assembly Theory to detect the likely presence of living things, among living and nonliving options, blindly fed into a mass spectrometer, with a very high degree of accuracy – which is pretty cool.

A Selector Can Produce Complexity:

The basic argument is that the selection of any random thing can end up producing noticeable “complexity” above the background noise, which is true. Everyone agrees. However, the type of complexity is important here when it comes to explaining the existence of a living thing. A given selector in the environment, like a specific temperature acting on a material like water, may produce something more “complex” and even “specified”, like a particular type of snowflake. The same thing is true for a selector like gravity acting on gases comprised of randomly arranged atoms and cosmic dust to produce stars, planets, solar systems, and galaxies. However, that’s not the same type of complexity as a living thing or a complex machine where multiple different types of parts are needed to interact in a very specific way to produce a very specific type of function (i.e., specified functional complexity at different levels having different minimum part requirements).  Assembly Theory doesn’t apply here since Assembly Theory isn’t concerned about the “right type” of selector that selects the “right thing” (Link).

Selecting the “Right Thing”:

In short, in order to explain the origin of life, or any other complex machine requiring multiple different interacting parts working in collective harmony to produce a given function, the selector itself must be even more functionally complex than what it is producing or creating.  Take a complex automobile, for example.  It could be produced entirely by robots in a factory. However, the robots themselves have to have access to be even more functionally complex than the automobiles that they are assembling, to including having access to all of the structurally complex information necessary to correctly assemble the automobiles so that they will actually function, in addition to their own structure and functional complexity.

Existence is the Proof?

Cronin’s argument that the ability of Assembly Theory to detect the need for a “selection mechanism” or a “selector”, even one as complex as would be needed to explain a living thing, proves that this living thing was the product of natural mechanisms.  Cronin argues as follows:

“There are three things here:  The environment, noise, and a thing. This thing, the object, it has several options. It can decay back into the noise and therefore have no consequences, just noise again, or it can go above the threshold and act on the resource, the noise, and pump that noise into more stuff.  And, as it keeps going, you know what happens in the end?  And this is how the first molecules and the first molecular networks complexified into a gigantic mess – because the mess wasn’t a mess. The mess was moving towards existence because the mess. Jim [Tour] doesn’t look at this over long enough of a time scale… everything on planet Earth is a product of selection. The contingency emerges with selection and the environment. Existence is the proof.” (Link)

Fakhouri counters with, “Existence is the poof??  No, Lee. It’s not”. (Link)  The argument that life exists, therefore life evolved by random chance, is a non sequitur – a very obvious circular argument. The best that Assembly Theory can do is detect that high-level selection has happened, but the very existence of life does not explain the mechanism of how life came to be – the actual nature of the “selector”.  The bottom line is that Cronin is just assuming what he wants to prove.  He hasn’t actually demonstrated anything regarding how any natural mindless mechanism could reasonably explain the origin of life…

Fakhouri asks the key question.  “What is the driving force behind selection?” (Link)

“If only we knew of a mechanism, something that routinely steps through combinatorial space to assemble functional objects. You know, functional objects like talks, and papers, and equations, and ideas like Assembly Theory. You know, a mechanism like, but nah – we all know that we’re not allowed to go there.”

Of course, we are “not allowed to go there” because the only known “selector” that is actually capable of producing complex machines, to include the type of machines necessary to produce a living thing, is an intelligent creator at the level of a God or God-like intelligence.

 

 

YouTube Video of Cronin vs. Tour: