Description

We seem addicted to explanations, especially when it affects our own lives or our understanding of the world. We are constantly asking why questions and there always seem to be multiple explanations possible. In this chapter, I am going to discuss alternative explanations, meaning that there are potentially alternative ways to explain the phenomena you want to explain.

In philosophy and science, alternative explanations are often sought in order to arrive at the best possible explanation of the phenomenon. We practice this because we want to explain the world around us in the best possible way. In science, the goal of gaining a better understanding of the world is often achieved. For example, astronomers who can explain why solar eclipses occur when they do, economists who can explain how inflation works or physicists who can explain how magnetic fields operate. And so, there are other successful explanations within science.

Covering Law Model 

Before delving deeper into the concept of alternative explanations, I will first outline what constitutes a (scientific) explanation.

In the philosophy of science, one is concerned to determine what a scientific explanation means. One prominent German-American philosopher who had an explanation for this is Carl Hempel. Hempel presented the idea of the covering law model. In this model, Hempel argued that scientific explanations arise from responses to “explanation-seeking why-questions”. For example, why does the earth revolve around the sun or why is baldness genetically determined? These questions require satisfactory explanations. According to Hempel, a scientific explanation is sufficient when you could determine the essential features that such an answer must have, then we would know what a scientific explanation is (Okasha, 2002).

The covering law model has the same structure as that of an argument, namely with premises and a conclusion. Thus, the premises explain why the conclusion is true. For example, someone wants to know why sugar dissolves in water. This is an inquiry that seeks an explanation for why. According to Hempel, in order to respond to it, we must build an argument why its conclusion is “sugar dissolves in water” and why its premises explain why this conclusion is accurate. The work of giving a description of a scientific explanation then shifts to precisely describing the relationship that must exist between a set of premises and a conclusion in order for the former to be considered an explanation of the latter (Okasha, 2002, p. 37). Hempel set himself that challenge.

Hempel provided a threefold solution to the issue. First, the argument must be deductive, meaning that the premises must follow from the conclusion. Second, each premise must be true. Thirdly, there should be at least one general law in the premises. A general law is also referred to as a law of nature – for example, the law of gravitation.

However, the covering law model encounters a number of counterexamples even though it accurately describes the structure of many real scientific explanations. One example is the problem of irrelevance. Assume a little child is in a maternity unit at a medical facility. A man named John is the only person in the room who is not pregnant, and the youngster observes this and asks the doctor why. John has been regularly using birth control tablets for the past few years, the doctor responds. Regular users of birth control tablets never get pregnant. Consequently, John is not pregnant.

Suppose that the doctor’s assertion that John suffers from mental illness and uses birth control pills, which he believes to be beneficial, is accurate. The child’s response to the doctor is obviously not helpful despite this. The fact that John is a man and men cannot become pregnant is the apparent explanation for why he has not gotten pregnant. Although this is consistent with Hempel’s model, we cannot say that this is a correct statement when we want to conclude that John is not pregnant because of the birth control pills. Having alternative explanations is therefore desirable in scientific research (Okasha, 2002, p. 43).

Inference to the Best Explanation

For as we saw in John’s example, there are explanations that are sometimes irrelevant to the phenomenon that needs to be explained, even though it may be deductively correct. Explanations may fail to meet the correct explanation in various ways. For this reason, it is always important in science to look further into other possible explanations for the phenomenon. Generally, what should be strived for is the best explanation. Investigating alternatives can help us arrive at this best explanation (Baggini, 2010 p. 72).

Leading philosopher of science Peter Lipton elaborated on this method in his work and called it inference to the best explanation (IBE). According to Lipton, to arrive at the best explanation, the most “loved hypotheses” must be considered. The loveliness in this sense here means that the hypotheses with the most explanatory power, is the best. In other words, loveliness is the explanatory quality, whereas explanatory quality is the measure of how sufficient a potential explanation is, of how much understanding it provides and if it is or if it were an actual explanation (Lipton, 2001, p. 119). The logical reasoning for finding IBE goes as follows:

• There is a collection of data, facts or observations (D)
• There is a hypothesis (H) that explains (D)
• There is no other hypothesis that explains D as well as (H) does
• Therefore (H) is probably true (Lipton, pp. 442-443).

To measure explanatory loveliness, Lipton argues that size, precision, mechanism, unification and simplicity are good candidates for explanatory virtues. “Better explanations explain more types of phenomena, explain them more precisely, provide more information about underlying mechanisms, unify apparently disparate phenomena or simplify our overall picture of the world” (Lipton, 2001, p. 106).

Scope refers to the extent of the explanatory power of the hypotheses. Thus, the hypothesis will explain a wider range of data as opposed to the alternative hypotheses. The following is an example:

• (H1): The main causes of high crime rates are economic issues like unemployment and poverty.
• (H2): The cause of high crime rates is due to a combination of factors, such as economic, social and cultural factors.

It is clear that (H1) has a narrower scope than (H2), therefore (H2) can extent to a wider range of data that is related to crime rates and may be favored in the analysis because of its broader scope.

Precision refers to the quality of an explanation in to what extent the explanation offers a specific, clear and testable predictions about what we should observe in order for an explanation to be correct. In other words, it should outline what we predict or expect to see if the explanation is true.

Mechanisms provide a deeper understanding of the data that is given. It can give an underlying process that explains the data. For example, when you want to explain the high crime rates, you have to explain the causal relation between crime rates, social status and economic factors.

Unification is the process of uniting and combining data into a coherent and unified framework. The difference between unification and mechanism is that unification focuses on the broader framework and connects the data and mechanisms explains in detail what the causal process is that lead to the specific data.

Simplicity refers to parsimony, in which parsimony means that the simpler explanations has to be favored. Simpler means here that fewer assumptions has to be made and fewer complex ad hoc elements in the explanation are favored.

Application

Significant questioning of a highly influential theory in the migration world, the “migration hump.”

The “migration hump” is a migration theory from 1971 (Zelinsky, 1971) that suggests that people in a developing country are more likely to migrate from that country when that country is actually developing more. The logic behind this theory holds that when a country develops, per capita income will increase. Therefore, a higher per capita income will make it easier to migrate. On the other hand, when the country develops even further and can call itself a high-income country (according to the World Bank, this entails a GNP of 3,956 and 12,235 US dollars in 2016), only then will migration from this country decrease significantly again. Indeed, at this stage, there will be less reason to leave the country. This theory thus concludes that income strongly influences migration, with poor countries having the lowest migration flow, middle-income countries having the highest migration flow and high-income countries also having a lower migration flow (Vermeulen, 2020).

The theory has also come under significant examination among policymakers as the European Union invests more and more money into its immigration policies, which are intended to deter people from coming to Europe. Assuming that the theory is true, it is irrational to think that increasing development aid can stop migration from underdeveloped nations. Since a country will advance up the curve if more development aid leads to higher development, more people will migrate, as is shown by the Migration hump curve published by De Correspondent. 

While this theory may be hypothetically correct, several migration theorists have commented on it recently, finding flaws when the theory is used in the long term and that it would not consider different migration trends. It would lack explanatory and predictive power, while these are important elements for a theory. First (1), new theory argues that multiple factors, rather than just income, come into play when it comes to migration from developing countries. Consider political factors, such as political regimes or war, or the geographical location of the country which may or may not be close to a western/northern country. Demographic transition whereas high birth rates and a decline in infant mortality in developing countries can cause high population growth. As a result, there are more young people than there are jobs for them, which can lead to rising youth unemployment. Since young people are more likely to emigrate than older people, this may increase the likelihood of emigration. Similarly, policy and legal frameworks are also a factor –  think of migration policies or integration processes that can differ between countries. In addition, (2) argues that the migration hump is a push-pull model through which migrants are seen as passive agents guided by capitalist systems, in which they assume that migrants are seeking for maximized income and utility (De Haas, 2021). Finally (3), new theories suggest that the migration hump indicates a correlation but not a causal relationship.

Analysis 

In this section, I will analyze the ways in which the new explanations provide a better explanation for migration theories. I will carry out this analysis using the above criteria; scope, precision, mechanism, unification and simplicity.

In the migration hump theory, a significantly narrower scope is used. Namely, the focus here is on migrants’ income. Focusing on income is often associated with neoclassical migration theory, in which economists look at economic reasons for migrants to migrate: reasons such as better wages or more employment opportunities. This is a good way of explaining labour-migration, but consequently, only partly says something about migration as a whole. As new theories argue, there are more factors at play when it comes to migration. Either way, the scope needs to be broadened. Therefore, a broader scope is better according to the criteria of inference to the best explanation.

The migration hump theory develops a theory based on a certain assumption, namely that when a low-income country develops more and becomes a middle-income country, people will possess more money to be able to subsequently migrate. If this assumption is true, we can assume that people from middle-income countries migrate more. While this assumption may be true, it does not necessarily mean that we can make all predictions for migration based on it. Capturing human behaviour is a complex matter and difficult to predict, especially when it comes to migration, which may therefore involve multiple factors. Thereby, this is a strong correlation found and not a causal relationship. New research wants to get closer to causality and in order to accomplish that, you must actually avoid drawing comparisons across countries. In addition, there are a plethora of other factors that could contribute to increased or decreased migration. You need to compare yourself to a country. In the loop of time. When Senegal becomes more independent, will it become more or less divided? When Afghanistan develops, will more or fewer people migrate to Europe? This is what you will discover.

The criterion-point mechanism plays a significant role when we talk about migration theories. In migration-hump theory, reasoning is based on a push-pull model. Push factors are reasons to leave the country, for instance because there are fewer economic prospects. Pull factors are reasons to move to a particular country, for instance because the economic prospects are better there. This model assumes that migration is based on an automatic flow of migration, in which the migrant plays a passive role in the process.

Neoclassical migration theorists explain migration by arguing that people are guided by the pursuit of maximum income and utilities. This mechanism turns out to be partly inadequate and misrepresents what migration processes actually look like. It tells us more about how migrants are pushed and pulled by global macro forces or as victims of capitalism only to be forced to seek a more prosperous life. This model does not justify the idea that people can also migrate based on free will and not be governed by push and pull factors. De Haas (2021) comes up with an alternative, the aspirations-capabilities framework, a mechanism in which people have the ability to freely choose where to live. This is a distinctly different mechanism to approach migration theory.

• Philosophical exercise: Migration theories can thus be reasoned from different mechanisms and placed against each other. In this case, the push-pull model is contrasted with the aspirations-capabilities framework. Which mechanism functions here as a better explanation for describing a migration process and why?

Migration hump theory links better economic prospects to migration, in which the driver for migrants is the pursuit of wealth. Although the theory seems to take only one factor into account, there are other links that Zelinsky makes. For instance, he links it to the Demographic Transition Model to get a deeper understanding of migration theory. The Demographic Transition Model explains how demographic characteristics shift from pre-industrial and agrarian to industrial and post-industrial eras of society. According to De Haas, on the other hand, the framework is not broad enough. Migration is a complex entity, making it difficult to place it in a particular framework. On the other hand, it may also not be possible to have a framework that is comprehensive when talking about a social science.

Migration hump theory does meet the criteria of simplicity, because by concentrating on one important variable, development, it simplifies the complicated interaction of factors that affect migration. One could even argue that the theory is oversimplified. The question, however, is whether a social science should want to meet this criterion. As I described earlier, migration is a complex phenomenon that precisely does not require a simple explanation. This ensures that the other factors are left out of consideration, in addition, simplicity ensures in this case that the theory has less predictability. Indeed, we cannot make predictions about people’s migration behaviours on the basis of this perspective alone. All we can say is something about economic motives.

Based on this analysis, we can conclude that the migration hump theory does not have much explanatory power and there are alternative explanations that can better elucidate the complex phenomenon. Alternative explanations can thus help us arrive at better explanations for the phenomenon. Based on Lipton’s criteria, migration hump theory does not score high, the only thing where the theory scores points is simplicity. In contrast, simplicity has been shown to be precisely not applicable when it comes to explaining a complex social phenomenon like migration. Lipton’s criteria for seeking the inference to the best explanation can therefore be very helpful in migration theory.

• Philosophical exercise: Since simplicity (in most cases) is not a good criteria point in social science, come up with another criteria point that would fit here and be in line with Lipton’s inference to the best explanation and describe how the migration hump theory would score on this.

References

Baggini, J. (2010). Philosophical Toolkit:  A Compendium of Philosophical Concepts and Methods. Wiley-Blackwell.

De Haas, H. (2021). A theory of migration: the aspirations-capabilities framework. Springerlink.

Liption, P. (2004). Inference to the Best Explanation. Routledge.

Lipton, P. (2001). What good is an explanation? Springer Verlag. (pp. 43-59)

Okasha, S. (2002). Philosophy of science: A very short introduction. Oxford University press.

Vermeulen, M. (2020). Een van de belangrijkste theorieen over migratie is ontkracht (en daarmee een hoop van mijn stukken). De Correspondent. 

Zelinsky, Z. (1971) “The Hypothesis of the Mobility Transition.” Geographical Review 6(2). (pp. 219-249).