Change is a central aspect of the world. Various changes are viewed as 'evolution', without there being consensus about a common definition. Some discuss evolution as a fact, others as a theory. Some see evolution as a causal process. Some talk about a definition of evolution, while in fact what is defined is a measure, such as a 'change in gene frequencies'. Some see selection as part of evolution, while others don't (e.g. 'evolution' with 'selection').

To bring the many viewpoints closer together we suggest that it may be profitable to think of evolution as indicating a central 'pattern of change' that, in a range of forms, can be recognized in the world. We suggest to use the name of 'Big Evolution' for the comparative study of all 'patterns of evolution', both inside and outside biology.

If one sees evolution as a pattern, this makes it easy to prove or disprove the presence of this pattern in the world. Such an approach can solve a lot of confusion about whether or not evolution is a fact, or a theory, or a process, or still something else.

For references and detailed information about this topic:
A rethink of evolution, and the construction of a family of ‘patterns of evolution’ inside and outside biology
Gerard A.J.M. Jagers op Akkerhuis, Hendrik Pieter Spijkerboer, Hans-Peter Koelewijn 2014


When defining the concept of evolution, a range of topics may have a relevant impact on the outcome. In the following paragraphs we will discuss a selection of these topics. At the end we will indicate a direction for creating a definition of evolution.

1. What is a suitable context for the anchoring of concepts that are used in definitions of evolution?

In our view, a suitable context should define all concepts from the ground up. In this respect, it is relevant that the entire operator hierarchy we discuss in this website, can be constructed from the ground up. And as a next step, the operators can be used as the basis for defining other systems that consist of interacting operators.

We think that due to this special property, the operator approach has a high utility as a basis for analysing the organisation of any system. A demonstration of the utility of the operator hierarchy has been the organism concept. The question what is an organism has proven very difficult to answer. Based on its stringent context, the operator hierarchy allowed for the following solution: "An organism is an operator of a type which is equal to, or higher than the cell type (represented by the bacteria)". Various other practical applications of the operator hierarchy for defining concepts are offered in Jagers op Akkerhuis et al. 2014

2. Does a definition focus on a measure of evolution or on the concept of evolution?

There is a difference between a definition of a measure of evolution, and a definition of the concept of evolution. For example if one states that: "Evolution is the change in the inherited characteristics of biological populations over successive generations", one actually defines a measure of evolution. However practical such a measure may be, it aims at the quantification of certain results of evolution. Such a quantification of results does not indicate what evolution 'is' in a philosophical sense. If we want to distinguish a phenomenon in the world that represents evolution from other phenomena that do not represent evolution, we are in need of a philosophical definition. Suggestions for how one can arrive at a philosophical definition will be offered in the next paragraphs.

3. Do we view selection as a process or as a pattern?

Darwin indicated that: “This preservation of favourable variations and the rejection of injurious variations, I call natural selection” (Darwin 1859, pp 80-81). To be able to observe natural selection thus defined, it is not sufficient to focus on a single individual. An individual can survive or die, but it cannot show 'selection'. To observe selection, which implies that some variations are 'favourable', and others are 'injurous', one must observe properties of two or more individuals in the offspring generation. These individuals may differ from their 'parent(s)' and may differ from other individuals of their generation. To comply with selection, the differences between the individuals must cause them to be unequally successful in their struggle for survival and, when conditions allow, the production of offspring.

Clearly, it is not possible to observe selection in a situation where all the individuals are equally (un)successful, or when a single parent produces a single offspring, which produces a single offspring, etc. And it is neither possible to speak about selection if individuals are different, but these differences do not translate in unequal survival (a situation that applies to neutral mutations).

This shows that an observation of selection focuses on the pattern of differences in the survival succes of individuals. This pattern is an emergent aspect of all the underlying processes, but the pattern is not part of the process itself. Even though many people frequenly speak about selection as a process, the need to analyse differences implies that -from a logical point of view- selection is not a process but at pattern.

4. Can we view evolution as a process or not?

Whether or not evolution can be viewed as a process may depend on what is seen as 'evolution'. If one accepts as evolution any and all change in the shape or shapes of one or more objects, then it will be easy to indicate change which represent evolution. However, one can also decide to limit the definition of evolution to 'evolution with selection'. As the definition now includes selection, it is no longer obvious that 'evolution' relates to a process. The reason for this, as was discussed above, is that it may be more correct to view selection as a pattern, than as a process. For this reason, evolution with selection combines offspring production, which can be viewed as a process, with selection, which we see as a pattern. Because of the combination of process and pattern, we do not view the outcome as representing a process. Instead we refer to this combination as a special pattern of change: the 'pattern of evolution'.

5. How does 'evolution' as a 'pattern of change', relate to other, non-evolutionary patterns?

Processes in the universe lead to various patterns of change. Some patterns involve the change of a single system (e.g. the changing course of a river). In other examples, generations occur, while no selection can be observed (e.g. a reproducing organism with one or more offspring which all survive). In still other situations things show generations and a pattern of selection can be observed (e.g. an organism of which the offspring vary in their survival).

For historical reasons, each of the above three processes is referred to as 'evolution'. This ambiguity easily leads to misconceptions.

To prevent such misconceptions, we suggest to use a general context in which different patterns of change can be given different names. As such a general context we use the thermodynamic viewpoint of 'flow patterns'. An example of a flow pattern is a river with estuary. In a general way, changes of flow patterns in estuaries and other flow systems have been descibed by Bejan as 'morphing'. By using flow patterns and their morphing, it becomes possible to identify different 'patterns of change', and to refer to some of these as morphing, and some as evolution. Meanwhile the general context of morphing links evolution to flow patterns and thermodynamics.

For example, the change of the flow of a river, or the formation of a snowflake, can be viewed as 'plain morphing'. But morphing must not always be 'plain'. Morphing patterns can be more complicated. For example, it is possible to view the genealogy that includes many generations of organisms as a flow pattern. One can decide to refer to changes of such a pattern over generations as 'generational morphing'. In fact, still more complicated flow patterns exist which combine generational morphing and selection (as a pattern!). These we view as patterns of evolution. By suggesting the names of 'plain morphing', 'generational morphing' and 'evolution' we contribute to a solution to the problem that at present the concept of 'evolution' indiscriminatingly refers to a broad range of very different patterns of change.

6. Is it possible to agree on a least complicated (almost trivial) pattern of evolution?

If we want to describe evolution by means of a pattern, we will have to be precise about the kind of pattern. We will, for this reason, start here with the construction of such a pattern, and we do this in the form of a graph. In this graph, there are entites which form the nodes, while processes are the connections (the ‘lattices’ of the graph). We will try to construct a least complicated, reduced graph. However trivial such a graph may seem, in our veiw it would have the advantage of low ambiguity and high clarity: every aspect of the graph is well defined and can be discussed in a precise way.

To construct a least complicated, yet comprehensive graph for evolution, we
minimum complexity evolution graph.jpg
A least complicated graph of a pattern of evolution. A pattern of evolution is shown in which a single original leads to a next generation with two different objects.
must minimise all aspects. Accordingly, we start with a single original (the ‘parent’). We are well aware that this choice differs from the choices made by those which approach evolution by using populations. The reason why we do not start with a population , is that it is difficult to define the population, as a concept, unambiguously. Moreover, evolution may also in reality start with a single object, for example an asexual bacterium, and such situations must be covered as well by the graph. In addition, the graph is based on a minimum complexity of all entities involved, leading to a single generation, only two second generation objects (the ‘offspring’), a minimum amount of structural change during the production of offspring (only one of the offspring shows a change in a single property, namely B1 and B2), and a minimal potential (either yes or no) for the second generation objects to give rise to a next generation. It is furthermore relevant that the graph also uses a single difference measure, which is based on the occasions where structural differences of the offspring lead to differences in the capacity to produce next generation organisms (which excludes effects of neutral variations). The difference measure is used to observe 'selection’. The minimal graph is shown in the figure on the right. Original A may be formed de novo, or may be derived from an earlier object.

7. Big Evolution: from a graph, to a family of graphs

As we see it, the least complicated graph for evolution can serve as a reference. Accordingly, we view as evolution any pattern of change that can be mapped in a one-to-one way with the least compicated graph. From this basis, it is possible to imagine that the concept of evolution is linked to a broader family of patterns of evolution. We named the science which deals with the exploration of this family of patterns Big Evolution.

8. Is it possible to prove or debunk evolution?

The answer to this question depends in our view on how one defines 'evolution'. In my perception, the concept of evolution refers to a range of patterns of evolution that are studied by Big Evolution, as indicating a range of patterns of change. Whether or not evolution can be proven thus depends on the definition of the concept, and on the measuring of changes in the world that fit to the pattern described by the concept. We focus first on the definition of the concept of evolution. A concept cannot be proven by performing measurements. The simple reason is that the concept of evolution is the result of a definition. The concept is the outcome of an agreement about a number of criteria which together are indicated as the definition of evolution. Once there is agreement about the definition, however, it is perfectly possible to prove that phenomena in the world exist which match the criteria of the definition. And if this is the case, this confirms that such phenomena represent evolution. In this sense, the 'existence' of evolution can be proven or debunked.