By Gordon Rugg
If you’re a researcher, there’s a strong temptation to find a Grand Unified Theory for whatever you’re studying, whether you’re a geologist or a physicist or psychologist or from some other field.
That temptation is understandable. There’s the intellectual satisfaction of making sense of something that had previously been formless chaos; there’s the moral satisfaction of giving new insights into long-established problems; for the less lofty-minded, there’s the prospect of having a law or theory named after oneself.
Just because it’s understandable, however, doesn’t mean that it’s always a good idea. For every Grand Unified Theory that tidies up some part of the natural world, there’s at least one screwed-up bad idea that will waste other people’s time, and quite possibly increase chaos and unpleasantness.
This article explores some of the issues involved. As worked examples, I’ll start with an ancient stone map of Rome, and move on later to a Galloway dyke, illustrated below.
In a museum in Rome, there’s a collection of marble slabs like the one in the photo on the left above. This slab has an inscribed image that looks like a classical theatre; next to the image is the word “THEATRUM”. It doesn’t take much genius to guess that this might be part of a map. They’re from an ancient map of Rome, known as the Forma Urbis Romae, that used to cover a wall of the Temple of Peace.
Once you’ve realised that the slabs form a map, you can then start working out how they should fit together to form that map, with every slab having its own unique place. Only about ten per cent of the original map has been found, but in principle, whenever a new slab fragment is found, there should be a single location in the map where that slab belongs.
That’s the beauty of a grand unified theory: A good one will correctly predict where future discoveries will fit. For example, when Mendeleyev created the Periodic Table for chemistry, it predicted the existence of numerous elements that had not yet been discovered, and more importantly, it predicted what the properties of each element would be.
So far, so good.
However, not all great unifying discoveries are the same. The periodic table predicted a range of properties for each undiscovered element. Newton’s theory of gravitation, in contrast, only deals with one property, namely gravity, and only works up to a point (which is where Einstein’s work takes over). Hooke’s law deals only with tension in springs, but does describe all springs; would that count as a grand unified theory? Pushing the issue to an extreme point, all animals could be described as land, water or air living, or as mixtures of all three; would that count as a grand unified theory?
When you look at questions like this, you realise that although the idea of a Grand Unified Theory is a nice one, it isn’t actually much use. A more useful way of looking at theories is in terms of a range of attributes. Some examples are:
- Prediction versus description
- Range of applicability
- Universality (does it apply without exceptions, or it it just a rule of thumb?)
These are all issues that have been discussed in depth in the theory and philosophy of science, so I won’t go into them in more detail here.
What I’ll look at instead is ways in which the search for a Grand Unified Theory can cause problems.
As mentioned at the start of this article, there are strong professional temptations towards looking for Grand Unified Theories. On top of that, there are strong psychological biases that push human beings in the same direction.
One well known bias is pareidolia, i.e. seeing illusory patterns, such as faces in clouds. It’s very easy to see what looks like a pattern in a batch of data, such as recurrent clinical features in a set of patients, when in fact there is nothing more than a random cluster. This overlaps with confirmation bias, where we tend to remember the cases that fit into the pattern we created via pareidolia, and to forget the cases that don’t fit.
Another bias is towards seeing agency in actions. We tend to think that things happen because of deliberate actions by someone or something. This is a long-running problem in public understanding of how evolution works. There’s a recurrent tendency for people to talk in terms of e.g. giraffes evolving long necks in order to reach leaves in trees, as if the giraffes made a conscious decision to evolve that way.
When it goes wrong
One common outcome of the biases above is conspiracy theories. Here’s an example. You’ll sometimes see images on the Internet like the one below, described as human-made metal spheres, found in rocks millions of years old.
Moqui marbles: By Paul Heinrich – Own work, CC BY 3.0, https://commons.wikimedia.org/w/index.php?curid=3691420
In fact, they’re natural concretions, but because they can take such strikingly regular forms, they’re often assumed to be the product of intelligent design and of human manufacture. A wide range of conspiracy theories are based on the belief that regularities in outcomes are caused by intelligent planned action.
Another possible outcome of these biases involves medical diagnosis, particularly involving syndromes. It’s tempting to see clustering of diagnostic symptoms as the indicator of a single underlying disease. However, there’s a significant risk that the cluster is either purely a coincidence, or is the result of one symptom causing another.
One example that has been debated in some depth in the literature is Gerstmann Syndrome, which claims that an unusual set of symptoms are manifestations of a single underlying syndrome; sceptical researchers, however, have questioned whether the syndrome involves anything more than some patients happening to have that set of symptoms by simple chance.
Similar questions have been asked about conditions such as schizophrenia, where there have been repeated suggestions that the label “schizophrenia” is nothing more than a reification that brings together a semi-arbitrary loose collection of symptoms in a way that gives no significant useful insights into what is actually going on in the patient.
One obvious type of grand unified theory featuring agency in actions is religion. Religions typically include creation myths, which explain the creation of the world in terms of a deliberate planned action by an animate agent. They disagree about whether the plan is a complete master plan like the Forma Urbis Romae, or more like a Galloway dyke, where there’s an overall concept, but most of the detail isn’t planned in advance, and can take different forms depending on chance. This disagreement is also a major issue in philosophy, particularly the philosophy of ethics. If everything is pre-determined, then there’s no obvious way for people to have free will and to make meaningful choices, in which case, concepts such as blame and crime are meaningless.
There are similar debates within other disciplines. Physics is an obvious example, but there are plenty of others. In biology, there’s a long-running discussion about whether features such as the emergence of complex multicellular life are pretty much inevitable, or whether they’re the result of sheer chance. A classic example is what would have happened if the Chicxulub asteroid and/or the Deccan Traps hadn’t wiped out the dinosaurs; would the world now be dominated by intelligent dinosaurs filling much the same niche that humans now occupy? In sociology and history, there’s been a similar discussion about the way that societies have developed; how much of this has been driven by random chance, and how much by underlying system pressures that would cumulatively push history in a particular direction?
Those are big, overarching, systematic attempts at grand unified theories. At a less formal level, there are a lot of specific beliefs that offer a far-reaching explanation of some individual features of the world. For example, there’s the just world hypothesis; the belief that overall, justice is done, and that if someone undergoes a bad experience, it will be balanced out later on by something good happening (and vice versa). Another common belief is that humans have a soul which survives after the body dies. Beliefs such as these often fit together fairly well on a small scale; for instance, the just world belief often co-occurs with the soul belief, resulting in concepts such as karma, or judgment of the soul after death, to decide whether it will be rewarded or punished in its afterlife. They don’t usually fit together very well on a bigger scale, because they create at least as many new questions as they answer (for instance, the question of where souls come from), but at a day to day level, they can provide a comforting illusion of explanation.
Grand unified theories are a tempting goal. However, that doesn’t mean that they always work.
This is a lesson that many disciplines have learned the hard way. Artificial Intelligence (AI) in particular is an area where many attempts at grand unified theories have perished ignominiously. The usual pattern is that AI forces you to work through all the details of your theory, and that the details are where the theory unravels spectacularly. Usually, you end up discovering that the problem you’re tackling has to be handled by a messy, low-level, brute force approach involving a large number of specific facts, rather than by a clean, elegant principle.
That’s not the most inspiring of conclusions. However, there’s an interesting implication that arises from it. Often, a limited-scope principle provides powerful new ways of looking at long-established fields. This happened in evolutionary ecology, when John Maynard Smith introduced the use of game theory, which provided major new insights. The same happened in politics, where concepts such as zero-sum versus non-zero-sum games gave new ways of looking at decision making.
I’ll return in a later article to the topic of widespread individual beliefs about the world, since these have far-reaching implications for the human world – in particular, for politics and religion.
Notes and links
Sources for original banner images:
Roman map image: By Ulysses K. Vestal – http://www.theaterofpompey.com/auditorium/imagines/maps/modernfur1.shtml, Copyrighted free use, https://commons.wikimedia.org/w/index.php?curid=906511
Drystone wall image: By RobertSimons – Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=27057613
There’s more about the theory behind this article in my latest book: Blind Spot, by Gordon Rugg with Joseph D’Agnese
Overviews of the articles on this blog: