This essay discusses race as a socially constructed concept.
The concept of “race” is gaining importance in English, especially American English discourse these days. Considerations over race are almost unavoidable in policy making and political speeches. But despite the prevalence of the concept of “race”, philosophers dispute whether “races” really exist, and a consistent single account, or even a dominant account, of such a concept appears never to have been reached.
Primarily, theorists have tried to provide analyses of “races” from two approaches—the biological approach and the social approach. Biological accounts of races include attempts to equate race with “subspecies”, “ecotypes”, “inbred lineages”, or “cladistic groups”, etc. Social accounts, by contrast, advocate that the concept of “race” is a social construction, which gains its meaning from objective social reality or historical conditions. Despite the great variety in biological accounts, most of them suffer from a semantic mismatch with our ordinary usage of “races” and embody inconsistency in the theories themselves. By contrast, analyses from the social perspective embrace greater coherence, even if few reliable empirical evidences can be found to support them. Generally, the idea that race is socially constructed, like the concept “gender”, is more promising than seeing it as some biological reality, like the concept “sex”.
But if race is a concept constructed by men, is there sufficient reason to retain it in social science studies? My own stance on this question is that race does need to be considered at present and for some period in the future, due to the social reality and historical stage we are in. Yet it should be expected that at some time in the future, the term “race” would lose its sense and we would take up a new measure, “ethnicity” for instance, to mark the distinctive physical and cultural features among social groups.
In the following parts, I will proceed by first elaborating on several biological accounts with reasons why they fail to work satisfactorily. Then I will explicate a possible way to define race as a social construct by mostly referring to Haslanger’s theories. In the closing part, I will voice my own opinions on how we should treat the concept “race” in the present context and in the ideal future context.
To get hold of how we might tell different races biologically, it is beneficial for us to consider how we tell one sex from the other. The most definite and essential way of doing so is probably by checking the existence of the Y chromosome in the chromosomal complement. The more usual and accessible means is to check the characters of individual organisms: a most decisive way is to check the reproductive organs; moreover, we can tell by looking at secondary sex characteristics, like the presence of Adam’s apple and the shape of the body, etc.
But deciding to what race an individual belongs can be hundreds of times more complicated than deciding sex. At the genetic or chromosomal level, no decisive evidence can be picked out to categorize people into a specific race. And at the biological character level, telling an organism’s race is perplexing, for the phenotypes we use to designate races are usually continuous and constantly contradict each other. We think of a creature with a uterus as a woman, and one with an Adam’s apple as a man. These two characters are not likely to occur on a single organism simultaneously. Yet when deciding individuals’ races, chances are good that typical phenotypes belonging to different racial groups are manifest in a single person.
Having realized that explaining races merely from a genetic perspective is hopeless, many theorists like Mayr and Andreasen tried to provide an account appealing to the evolutionary story. They argue that before an ancestral species evolves into two distinctive species incapable of producing fertile offspring with each other, the ancestral species undergoes a stage where there are subspecies existing within it. These subspecies have accumulated features unique to the group and distinct from other groups. In the animal world, where reproductive isolation exists owing to geographical or spatio-temporal structural reasons, such an account may be reasonably accepted. Yet with human beings, where genetic flow through interbreeding has not been rare since ancient times, introducing “subspecies” as a synonym for “races” can be awkward, as there is no way to demarcate groups in the absence of reproductive isolation. In the evolutionary view, other claims, trying to interpret races as ecotypes, for instance, basically meet the same problem as the subspecies theory.
In addition to the above attempts to analyze races biologically, there are defenses of biological racial realism using more anthropological methods. A representative definition is from Hardimon, asserting that the word “race” means a group distinguished from other individuals by distinctive physical features, whose members are linked by a common ancestry peculiar to the group, and that has the same geographic location as its origination. Though this sounds plausible, the account is suspicious from every aspect. First of all, people’s physical features, as we have discussed, are continuous and can vary greatly within a group. Secondly, the fact is that most acknowledged races share the same ancestry, which is not peculiar to the race. And thirdly, scarcely any races are monophyletic according to human history.
However sophisticated biological analyses of races may be, they more or less deviate from our ordinary usage of “races” and fail to tell the complete story of the origin of all races and how they can be demarcated from each other.
Given that “races” cannot be fully illustrated by natural science, it is reasonable to think that “race” is only a social construct. But scholars believing in social realism about race differ as well. Some argue that race is real in the social sense without any negative attitude toward the biological candidates, while others believe that race is real only in the social sense. Some think race is only politically real, some see it as politically and culturally real, and some advocate that it is real as a socio-historical construct. In this section I will mainly discuss race as a political socio-historical construct based on Haslanger’s theory.
Sally Haslanger provided a very unique insight into the concept of “race” by paralleling it with the concept of gender. In defining what gender is, she inserts the idea of social hierarchy into the definition. She argues that a woman is an individual with expected body features who is subordinate in social position and whose subordination is reinforced by the existing social gender structure. By the same token, a race is a group of people with certain body features who are assumed to have a certain ideology and occupy a certain position in a society, and that ideology and position a race shares are reinforced by the existing allocation of roles and social positions to different racial groups in the existing society. Haslanger does not explicate how that imbalanced racial allocation of social positions came into being. Yet arguably, it is a consequence of historical events. Especially in the United States, the disadvantaged situation coloured groups are generally in is the aftereffect of their ancestors’ unfortunate past.
Though the idea of hierarchy is built into the concept of gender and race, Haslanger argues that it is not necessary for individuals to be fitted into any gender or racial categories. One can always refuse to be gendered or raced. But the hierarchical notions of gender and race are of political importance, for policy makers need to use them in contexts concerning social equity. She further illustrated that the most ideal situation should be that hierarchical genders are replaced by non-hierarchical genders and races are replaced by some non-hierarchical notions like idealized ethnicities.
People may feel it is a sensible strategy to eliminate the notion of race because the prevalence of talking about it solidifies social stratification. My attitude toward the usage of the notion is that its elimination is an ideal situation, yet it will not be achieved by scientists’ and politicians’ suddenly abandoning the notion in practice. Instead, it should be achieved via paying closer attention to this concept in the present context.
The social reality is that racial bias and value-based stereotypes are still prevalent, though in a much more implicit way than before. The police are prone to see dark-skinned individuals rather than fair-skinned ones as potential criminals subconsciously. Teachers treat students with different skin tones and eye colours in different manners and attitudes without even realizing it. Under most circumstances, individuals do not intend to prioritize a certain race, but they just do. This instinctive bias is more or less based on the social fact that, in general, whites are better educated, in better material conditions, and commit less crime. Abstracted from these social facts is a folk generalization, which projects onto individuals when people are dealing with specific cases.
Stopping the use of the concept “race” is not the best policy to prevent racial prejudice, for folk generalization is ubiquitous and would proceed even if the word “race” were abandoned. The most effective way of achieving a non-hierarchical notion of social groups should be lifting the overall social position of disadvantaged races, promoting their degree of education and alleviating their poverty through preferential policies, as well as increasing the proportion of these races in the elite class—all of which call for an in-depth discussion of the concept “race” and the application of it in social and political practices.
What is to be expected when the concept of “race” is eliminated in the future world? I do not know what to say for sure. Yet I can describe an interesting situation where a civilization does not have such notions as “race”. Concepts like “race” or even “nation” were not introduced to China until the late 19th century. By that time, the concept closest to “race” was something similar to “ethnicity”, which is a culturally related concept. In the Tang Dynasty (600–900 AC), one could be accepted as ethnically Chinese without a national identity card, regardless of skin tone or eye colour, as long as one spoke Chinese, wore Chinese costumes, and conformed to Chinese etiquette. In fact, there were many westerners with fair skin, blue eyes, and blonde hair living the Chinese way in East Asia at that time. In such a society, one’s biological phenotypes or body features are of trivial importance and have barely anything to do with value-related issues such as fortune, reputation, and position. By describing this ancient Chinese society, I do not mean to say that a society without any concept of “race” or even “nation” is necessarily more advantageous. Nevertheless, it at least shows that the notion “race” is not as essential as many people assume it to be, and a society structured without such a notion is perfectly possible.
[1] Haslanger, S. (2000). “Gender and Race: (What) Are They? (What) Do We Want Them to Be?” Noûs.
[2] Kitcher, P. (2007). “Does ‘Race’ Have a Future?” Philosophy and Public Affairs.
[3] Spencer, Quayshawn. (2017). “Racial Realism I: Are Races Real?” Philosophy Compass.
[4] Spencer, Quayshawn. (2017). “Racial Realism II: Are Folk Races Real?” Philosophy Compass.
This essay discusses how the interpretive social sciences have some fundamental epistemic aim (e.g. “understanding”) that differs from fundamental aims of the natural sciences. I later realize that according to these philososphical terms, ‘mech interp’ is actually ‘mech explaination’.
The basic functions traditional natusral science serves are already recognized and agreed upon by scientists and society—to explain, to predict, and to manipulate. Yet starting from the 20th century, the emergence and development of the sciences of man make us return to the question of science demarcation and rethink what purpose science—especially social sciences such as politics, economics, and sociology—should serve.
Charles Taylor, in his thesis Interpretation and the Sciences of Man, argued that one aim of the social sciences is to provide an interpretation of social meanings connected to social practices in particular societies. After his paper was welcomed by many social science and philosophy students, Michael Martin spent a chapter in his book Readings on the Philosophy of Social Science criticizing Taylor’s theory for being too exclusive and even to some extent implausible. Stephen Grimm and Karsten Stueber both stressed the difference between the human sciences and the natural sciences, and noted that when it comes to the human sciences, psychological interpretations or understanding-as-taking-to-be-good is of essential importance.
To answer the question whether the social or human sciences have different aims from the natural sciences, my own stance is that the social sciences still share some aims—namely, explanation, prediction, and manipulation—with natural science, yet such aims may never be as successfully and precisely achieved as in the natural sciences. Apart from these aims, the social sciences possess unique aims like understanding due to the nature of the subjects they study. In this essay, I will first draw on different features of the social sciences and the natural sciences. After that, we can elaborate on how the aims of the two differ.
According to Taylor, hermeneutic sciences have three distinctive features. First of all, there needs to be an object or a field of objects of study. Such objects can be societies, institutions, actors, etc. Second, there needs to be a distinction between the meaning and the expression, in other words, a distinction between the interpreting and the interpreted. Finally, there also needs to be a subject. He also indicates that the last point is what demarcates hermeneutic and natural sciences. Taylor does not explicitly mention what sciences should be put into this category of hermeneutic science, but we can speculate that what he has in mind is all social science, for any social science more or less covers studies which satisfy all requirements. I admit that the above features are a good identification and description of social science, but using the name “hermeneutic” is too dictatorial.
The most striking difference between natural science and social science is what they study. In natural science, we talk about the motion of objects and the reaction between chemicals. These objects and chemicals have no desire, no ambition, and no free will. You may argue that in not all natural sciences do people study things without will: men are constantly occurring in research topics in biology and medicine. It is true though; in these studies, what the man in question thinks and desires is usually purposefully omitted. What we care about is not the man, but his heart, artery, tissues—you name it. In a word, in natural science, the studied thing is an object.
For an object, laws of nature can always be reliably applied and induction hardly ever fails. The moon came to be gibbous this month, last month, and every month before. And we can confidently say that the sun will rise tomorrow out of the belief that laws of gravity, motion, and the travelling of light are credible enough. Sam has jogged when the moon was gibbous this month, last month, and every month before ever since I knew him. But I cannot confidently say that Sam will go out jogging next time when the moon is gibbous. Sam could change his mind. He could get a serious disease which prevents him from jogging. I can never say for sure that Sam never failed to and will never fail to jog, for whether he jogs or not depends on his will.
Sam here in the example is a subject. And all social science studies incorporate investigations on subjects who have minds, and whose minds are shaped and influenced by their past experience and their social milieu. Social science is constructed upon these subjects it studies, and cannot desert the very basic premise that its subjects have will and desire as well as the power to make a change.
Based on this feature of the sciences of man, I agree that social science studies can’t avoid falling into the so-called “hermeneutical circle” to some degree. By “hermeneutical circle”, Taylor means that one cannot understand any part of the whole without referring to an understanding of the whole; on the other hand, the understanding of the whole depends on the understanding of its parts. In Martin’s piece of work, he rebuts that the hermeneutical circle is a feature possessed solely by social science by introducing Kuhn’s theory-laden argument. He says that in natural science, observations are usually influenced by established theories. For instance, if we ask someone to tell the suit of the card, and mix a black-coloured heart in it, the black-coloured heart is very likely to be identified as a heart or a spade. Yet on the other hand, the theory is established upon and checked by observations, just as the classification of the cards is first established upon observation. Martin claims that if the circularity of natural science can be overcome, the hermeneutical circle can be overcome in a similar manner, say by appealing to probability evaluation.
But this parallel between the circularity in natural science and the hermeneutical circle is inappropriate in the following two senses. First, as Taylor suggested, observations made in natural science are always presented in brute data, that is, empirical facts, yet the direct evidence for social science theories is scarcely presented in this form. In the proceeding of natural science, brute data serve as a breaking point of the circle: once data contradicting the theory are found, we can acutely sense that something goes wrong with the hypothesis. But with social science, there is barely any standard we can refer to in order to validate an interpretation, except the context from which that interpretation is generated. Secondly, since what is studied in natural science is objects and the causal structure of the natural world is rigid and fixed, applying the understanding-as-grasping-of-structure method to perform research is quite enough. Therefore, when there is a mismatch between the structure and the understanding, we can start to find fault with the theory with certainty. However, the research field of social science is constructed out of meanings, whose causal structure is always intricate and ever-changing, under which circumstance we have to employ the understanding-as-taking-to-be-good method. Such a method cannot be carried out without any reference to the context of a certain question.
I said in the last section that equating social science with hermeneutic science is dictatorial, because there should be aims other than interpretation for social science. It also explains, predicts, and manipulates.
The explanatory function of the social sciences has been appropriately illustrated by Martin’s case of Marla N. Powers’s menstrual taboo research. Indeed, when social scientists question and work on the origin of social phenomena and particular events, when they wonder why a certain occurrence is the way it is rather than another, they are constantly tracing along a causal chain or a causal net of an existent event. Such a causal net does not necessarily have to be woven entirely by meanings. An anthropological study can provide strong evidence for this. Around the North Pole, anthropologists found there is an Eskimo tribe which has a custom of taking the elderly out and leaving them in the tundra to be eaten alive by beasts. The theory behind this terrifying ritual is that under the extreme climate and low productivity, the whole tribe would suffer extinction if they do not do so in order to save resources for the young. In this case, I would not consider the anthropological theory an interpretation. For it to be an interpretation, the theory should be something like “the convention of the tribe makes the young Eskimos feel urged to do so”, or “certain myths and beliefs in the tribe make the young Eskimos feel that taking the elderly out is a duty”, etc.
Also, social science can serve purposes such as prediction and manipulation. How mainstream economic theories work these days is very strong evidence for this. Economists make hypotheses, build models, collect data, and make predictions—much the way natural scientists would do in their fields. But economists, especially econometricians, can only do so based on the hypothesis that several properties or features of men are held fixed. In other words, men are treated like objects in these practices. But conducting social science in this manner has a price to pay. Because the approximation in the first place is inaccurate, the error of the result can never shrink to the same degree as in the natural sciences. And to cut down the error and make as good an approximation as possible, these social scientists ultimately still have to resort to interpretation.
Interpretation contributes to human society in three facets, which makes it a desirable aim for the social sciences. First, it helps us to organize the memories of the past, including common memories of a society or an institution and the memories of an individual. Once actions and events of the past are understood through interpretation, they would be retained more vividly and the storage of heuristic resources would be more abundant for us. Second, they help us empathize and feel empathized with. This empathizing function is based on the first memory-organizing function. The more past and strange events and situations we are acquainted with, the better we can articulate our own feelings and comprehend the feelings of others. Experiences as such are not rare: at the time we finish appreciating a tragedy, we suddenly feel that some repressed feeling of ours is released. When people can empathize with others better, it naturally follows that they know more about how to act properly under certain circumstances. Lastly, interpretation can explain and even shape the common meaning shared by a society, and foresee or even trigger revolution.
To conclude, social sciences share certain similarities with natural sciences in terms of certain aims such as explanation, prediction, and manipulation. Yet the former differs from the latter in the sense that the features of their study content are very different, making interpretation a very important aim of the social sciences.
[1] Taylor, “Interpretation and the Sciences of Man”
[2] S. Grimm, “How Understanding People Differs from Understanding the Natural World,” Philosophical Issues
[3] Karsten Stueber, “Understanding vs. Explanation? How to Think about the Difference between the Human and Natural Sciences,” Inquiry
[4] Michael Martin, “Taylor on Interpretation and the Sciences of Man” in Readings in Philosophy of Social Science (ed. Martin)
Following Hempel’s D-N model of explanation, theories equating the explanation of an event to providing its causes sprang up. Among these theories, the idea that “to explain an event is to provide information about its causal history” is initiated by Lewis in his paper Causal Explanation. Admittedly, providing causal histories of a certain event is a very prominent characteristic most explanations share under many circumstances. However, interpreting all explanations of events as the provision of information about their causal histories can be far-fetched, and is prone to meet with potent counterexamples which are difficult, or even impossible, to resolve.
In this essay, I will elaborate on two types of situations where Lewisian causal explanation theory fails—namely, dispositional explanation and distinctively mathematical explanation. I will illustrate how Lewis is aware of the potential risks his theory is confronted with and how he attempted to defend himself in the following section. Then I will focus on dispositional and distinctively mathematical explanations respectively to show why Lewis’s defense is not successful.
According to Lewis, any event that needs explanation stands somewhere in the complex causal net. It has a complex causal history. The causal history of an event has a tree-like structure, and in the branches of the tree reside various other events. My tripping over might be caused by the threshold in front of me, and my friends’ giving me a call, and the surrounding crowd’s blocking my sight of the threshold. The existence of the threshold may be caused by the architect’s design and the craftsman’s work. For the causal history of my tripping over, I can incorporate everything mentioned above: the threshold, the call, the crowd, the architect’s design, the craftsman’s work, my going to the place, my birth… Or even the Big Bang!
But it is ridiculous to say that my tripping over can be explained by the Big Bang. Neither would Lewis say the Big Bang is explanatory. To be counted as an explanation, the information provided should be sufficient, correct, and up-to-date, promoting the overall understanding of an event’s causal history for the person informed. And explanations can be evaluated by standards such as whether they contribute to clarifying the causal history of the event, whether they are able to identify falsehood, or whether the causal process they involve is familiar, etc.
At first glance, this causal interpretation of explanation seems to work very well. After all, when we are asking why-questions, a satisfactory answer we get normally starts with “because”, and ends up with some strong, clear, and correct statements about the causes of the event in question. Yet Lewis admits that there are obstacles in the application of such theory, which are represented by three typical cases. The first case is that when we try to explain the traveling path of light, we appeal to Fermat’s law, which says that light travels in the path which takes up the least time. However, Fermat’s law appears to stand nowhere in the causal history of the path that the light takes. The second case noticed by Lewis is that we explain the stopping of the collapse of a star as “a more collapsed state would violate Pauli’s Exclusion Principle”. In this case, the explanation seems to provide no causes at all. The third case is that when we are explaining one’s immunity to smallpox, we may say that he has been vaccinated, and the possession of the antibody makes him immune. Yet there seems to be no causal link between the possession of the antibody and the immunity.
Briefly, Lewis tried to defend his theory from these counterexamples in two directions: to the first and second cases, he replies that information about causal history is indeed provided, and to the third case he replied that dispositions such as being immune to smallpox cannot be regarded as events. From Lewis’ perspective, Fermat’s principle itself is rejected as an explanation. It is explanatory because it implies explanatory information such as the speed of light, and how waves propagate in different media, which are causes of the light’s path. Secondly, explanatory information is also provided in the collapsing star case, just in a negative way; yet negative information about causal history, argues Lewis, is still explanatory information. In reply to the third case, it is said that being immune is a dispositional property, yet not an event to be explained.
But this dispositional objection to causal explanation theory is dismissed too hastily by simply asserting that dispositions are not events to be explained and the causal histories of dispositions are untraceable, because in many contexts, especially in non-physical sciences, we are always including dispositional properties among the causes of a particular event. Under certain circumstances, some dispositional properties are even dominant causes of the event in question. If there is no way to tell the causal history of a dispositional property, then the causal event caused by that property would become obscure, and there cannot be credible causal explanations for the causal event as well.
Consider the following study on marketing conducted by Lambrecht, Tucker, and Wiertz in 2015 and their explanation of the result of their study. In the research, they collected statistics on people who included hot search keywords in their Twitter posts. Some of them included those words in their posts on the very first day when they occurred on the top search list; some used them on the second day; the latest users studied posted the words on the fourth day. What the research group did was investigate how the frequency of clicking on ads on Twitter is related to individuals’ time of including top search keywords in their posts. Surprisingly, they found that the highest frequency of clicking on the ads appeared in the group where people included the words on the fourth day. The explanation of the result is that these people have a disposition to agree with other people’s opinions and believe what others say.
This explanation may not, in a strict sense, be a scientific and rigorous one, but still is convincing and has potent explanatory power. The argument can be reduced to the form as follows:
(a) A and B are strongly positively correlated events.
(b) We can think of a dispositional property d of the subject S, which is causally related to A and B.
(c) Therefore, we may reasonably speculate that A and B can both be explained by d.
In this particular case, A stands for including hot keywords in the Twitter post on the fourth day; B stands for clicking on ads more frequently; and d stands for the property of easily agreeing with others.
But put this explanation in the light of causal explanation theory, it is strange how the explanatory power of d should be granted. For d is a dispositional property, and has no explicit causal history. Consequently, adding d to the causal history of both A and B should yield no extra information about the history of both events.
One possible way out is to replace Lewis’ “causal history” with “causal connection”, as Marc Lange indicated in his paper What Makes a Scientific Explanation Distinctively Mathematical. If we grant that something is explanatory not only because it provides information about its causal history, but also because it provides information about its causal results, and anything causally related to it, then the explanatory power of dispositions can possibly be justified. Nevertheless, I don’t think making such an alteration is prudent. After all, it is even stranger to say that the driver’s drunkenness is because of the car accident.
Lange depicted a very detailed picture of what distinctive mathematical explanation is, and why such explanation is non-causal. Yet it appears that his ambition is far greater than merely rejecting the adequacy of causal explanation theories. He also aspires to show that no explanation is non-causal, except for distinctively mathematical ones. On this point I can hardly agree, though Lange’s project on the whole is indeed a very compelling one.
There is a uniform recipe for Lange’s versions of non-causal explanations. In the traditional philosophical debates between empiricists and rationalists, there is a long-lasting headache for the former: to explain the acquisition of knowledge in logic and pure mathematics like algebra and geometry. Logic and mathematics are very peculiar because they can be true without any reference to the empirical world. The so-called distinctive mathematical explanation can be regarded as an empirical event which has a mathematical counterpart, to be explained by a pure mathematical theorem. To name some, the mother cannot equally distribute 13 strawberries to her 3 children without cutting them because 13 is not divisible by 3. And if I cut the three angles of a triangular cardboard and put them together, two of the edges can form a line, because the sum of the three angles of a triangle is 180 degrees. The “pure reason” world of mathematics has no direct causal relation with the empirical world, which is the realm of causation. Consequently, explanations appealing to mathematics manage to escape the discussion of causation with ease.
In the same manner, I argue that explanations appealing to logic have this sort of advantage as well. For example, I can explain that when I am in Mansfield College, and Mansfield College is in Oxford, it can be concluded that I am in Oxford, because if A⊂B, and B⊂C, we can conclude that A⊂C. An explanation like this is also a non-causal one.
Now, I have examined the theory which interprets explanation as providing information about an event’s causal history and tried to attack it from certain aspects. I think the theory fails to deal with explanations which have a dispositional element in them, and cannot account for distinctively mathematical and logical explanations. Presumably, there are other corners of explanation where the theory in question fails to reach. Yet I think these two groups of counterexamples already have the power to prove that the causal history account of explanation can turn out to be unsatisfactory.
[1] Lewis, D. K. “Causal Explanation.” 1986.
[2] Lange, M. Because without Cause. Oxford University Press, 2016.
[3] Sober, E. “Equilibrium Explanation.” Philosophical Studies, 1983, 43(2): 201–210.
[4] Hempel, C. G., and Oppenheim, P. “Studies in the Logic of Explanation.” Philosophy of Science, 1948, 15(2): 135–175.
Induction, by its definition in the Oxford English Dictionary, is “a method of discovering general rules and principles from particular facts and examples”. In ordinary cases, those “particular facts and examples” are what we have experienced in the past, and we use them to make predictions about the future and the unobserved instances.
In an important sense, we cannot live without such a method of induction. We have bread fill our stomachs because we believe that it will nourish us. For it did so yesterday, last week, last year, and it did so to numerous ancestors of ours. In fact, a counter-inductive way of living can be ridiculous. D. H. Miller once depicted a hilarious picture of living a counter-inductive life in his article The Warrant of Induction:
“Counter-inductivists aren’t just odd: they’re mad. Imagine one. He won’t eat bread: he thinks it would poison him, because it never poisoned anyone before. He would eat cyanide, which he also expects to freeze in the oven and bake in the fridge; but not by swallowing it. He won’t use any language people so far understood, or breathe air, or drink water. And throughout his (brief) life he inconsistently defends his wholesale counter-inductivism that as it’s almost never worked yet, it will now.”
But ultimately, are there really any solid grounds for us to rely on inductions and trust the conclusions derived from such a method? In deductive arguments, the conclusions are already entailed in the premises, yet such is not the case with arguments which are inductive, making the validity of induction dubious. When we reason deductively, with the premises that “All Greek philosophers had beards” and “Socrates is a Greek philosopher” in mind, we can draw the conclusion that “Socrates had a beard” with certainty. Yet if we are to argue inductively, the argument shall be constructed as follows: “Pythagoras had a beard; Zeno had a beard; Thales had a beard; Aristotle had a beard… All Greek philosophers that we know of had a beard. Socrates is a Greek philosopher. Therefore, Socrates had a beard.” Though put in a similar form, the conclusion drawn from the inductive inference appears to be less certain than that drawn from the deductive inference. But why?
The problem of induction was first noticed and discussed by David Hume in his book An Enquiry Concerning Human Understanding. Putting it briefly, Hume attacked induction from two angles. Firstly, it is totally possible for unobserved cases to deviate from those observed regularities. Secondly, the reliability of projecting the past to the future can never be guaranteed. Following Hume, other vivid and potent examples where induction fails have been raised. Bertrand Russell once constructed an example where a peasant fed his chicken every day throughout its life, until one morning, he came and wrung its neck. From the chicken’s perspective, induction worked very well until that catastrophic morning. Additionally, Nelson Goodman composed an even more illustrative example: suppose there is a special colour called “grue”, which applies to things that are green if observed before time t and blue otherwise. Before time t arrives, all observed green emeralds are also grue, so induction appears to support equally well the claims that “all emeralds are green” and “all emeralds are grue”. But after time t, these two hypotheses diverge, for the latter predicts that emeralds first observed after t will be blue.
Personally, I hold the opinion that the problem of induction is inescapable. Despite various attempts to account for the problem of induction made by many genius philosophers, there’s no overwhelming argument which solves the problem convincingly. But the inescapability of the problem of induction brings about another serious question: there can be no reason for us to believe in scientific speculations and theories. Given that all inductive inferences are unreliable, are scientific inferences such as the law of gravitation any different from the chicken’s inference of being fed tomorrow? Our intuition refuses such an equation, but telling the difference between the two, in other words, telling a good inductive argument from a bad one, is no easy task. The main purpose of this essay is to try to work out the possible criteria to distinguish good and sound inductive arguments from poor and unreliable ones. Before looking into that, I will first list out some inductive arguments for the convenience of further consideration.
(i) The sun will rise tomorrow.
(ii) The peasant has fed the chicken every day throughout its life, so he will feed it tomorrow.
(iii) All crows are black.
(iv) All emeralds are green.
(v) All metals are conductive.
(vi) All objects on my desk are conductive.
(vii) The pompous buildings found in the desert were constructed by human beings.
(viii) Stonehenge was built by human beings.
(ix) Having been travelling for 24 hours, he must be very tired now.
(x) The mechanic at the bike store told me that my brake was about to fail.
Since the problem of induction is insoluble, and hence induction itself is not something that can differentiate between scientific arguments and random arguments like “the peasant is going to feed the chicken tomorrow” and ascertain the dependability of science, we need to look elsewhere for the affirmation of valid inductive arguments. In fact, Hume tried to justify our belief in propositions like “the sun will rise tomorrow” and “the stone will fall if you let go of it” by saying that nature has its uniformity. Though our preference for induction is no more than animal instinct, following inductive conclusions cannot do us any wrong as it conforms to the uniformity of nature. Yet resorting to the uniformity of nature can be ambiguous and indecisive. For example, are propositions like (iii) and (iv) conforming to the uniformity of nature? Before we actually find a crow that is not black, we can well agree that (iii) obeys laws of nature, which is totally another story once we find a crow that’s of another colour.
Hume’s problem of induction has inspired many of his successors to develop various systems in response to the difficulties the validity of science faces, among whom Karl Popper is quite an eminent one. Popper advocated that we should abandon the inductive explanation of knowledge altogether and invented the concept of falsifiability. According to the theory of falsifiability, good scientific arguments are ones which are falsifiable, yet have not been falsified. Being falsifiable means that a theory can be contradicted by evidence. For example, “the sun rises every day” is falsifiable, for it can be contradicted if the sun does not rise on a certain day. Yet “the soul exists” is not falsifiable, for we can neither prove nor disprove it by evidence. Falsifiability draws a line between science and pseudo-science, and in a sense replaces induction in justifying the validity of scientific theories. If we examine the ten inductive arguments listed above, we can see that all of them are falsifiable. As a matter of fact, all inductive inferences, which are drawn from daily experiences, are falsifiable.
But being falsifiable is not the only condition that restricts a theory or an argument to be a scientific one. To be identified as scientific, we also require an argument to be unfalsified. For instance, if we have not found a crow which is not black up to now, we can grant (iii) as an inductively justified statement. Goodman’s grue emerald example, however, shows a difficulty for this approach. Before time t, as the proposition “all emeralds are green” is falsifiable and not falsified, we can grant (iv) as an inductively justified statement. But the very same observations also support the proposition that “all emeralds are grue”. The problem, then, is that falsifiability by itself does not tell us why “all emeralds are green” should count as the better inductive inference. It may appear a bit weird to allow such a conflict in scientific statements to happen. But in reality, science itself is conditional, and old theories do need to be replaced by more illustrative and powerful new ones when the context of discussion changes.
But the falsification theory still encounters serious problems, for it cannot successfully tell a bad inductive argument from a good one until it has been falsified. The argument “all objects on my desk are conductive” is no weaker an argument than “all metals are conductive” until we find, say, an eraser on my desk. But it is odd to say that the validity of an argument depends on the discovery of counterexamples. Rather, whether an argument is a valid one should be determined by its internal structure.
The greatest problem of falsifiability analysis lies in the fact that it checks every argument, or hypothesis, separately without fitting them into the whole picture of the system of a certain science. It is impossible to justify the validity of a single argument without looking at it from a global perspective.
For any inductive argument, we need to resort to deductive reasoning to assure its validity. Consider the first argument, that “the sun will rise tomorrow”. We may say so out of two considerations. Firstly, the sun rose today, yesterday, the day before yesterday, and so forth. Therefore, we form the belief that the sun will rise tomorrow. And secondly, we may hold the belief that it is so because of the self-rotation of the earth and the law of gravitation, etc. What guarantees the validity of the inductive argument is not the first consideration but the second one, which deductively links it to other established scientific arguments. The object of the first consideration can be any inductive argument listed above, yet not every one of them can be considered in the second way. For instance, the proposition that “the peasant will feed the chicken tomorrow” cannot be concluded deductively from other established arguments, and hence cannot be seen as a good inductive argument. Other arguments may be, or may not be, good inductive arguments, depending on whether a deductive explanation can be given about the fact stated by the argument. For instance, if the proposition “all emeralds are green” is merely inferred from the fact that all emeralds observed up to now are green, it is not enough to say that such an argument is a good and reliable one. Yet if such a proposition can be further demonstrated by arguments about the crystal structure of the emerald and the way it interacts with light, etc., then we can regard argument (iv) as a good inductive argument. I refer to this way of determining whether an argument is a good one as the method of multiple verification.
Such a method may receive objections, saying that we cannot guarantee the validity of “the sun will rise tomorrow” by resorting to the law of gravitation, as scientific laws as such are themselves purely inductive ones when they are first established. As a matter of fact, if we seriously ponder over inductive and deductive inferences, we would surprisingly find that all deductive arguments, once given any content, incorporate premises which are inductive conclusions. For example, when we construct the argument that “all human beings die”, “Socrates is a human being”, and “therefore, Socrates dies”, the first premise is one generated by induction. But admitting that most fundamental scientific laws are initially constructed from pure inductive inferences does not impair the effectiveness of the method of multiple verification, because after deductive conclusions are drawn from those initial propositions, initial arguments purely derived from induction come to be deductively linked to their conclusions as well. At the same time, when conclusions are getting strengthened in empirical experiments, the validity of premises gets enhanced as well. For example, if we consider F=ma as a theory drawn from purely inductive inference, and we build a gadget of mass m, with a force F applied to it, whose acceleration is measurable. We repeatedly apply the same force to the gadget and constantly find its acceleration to be a in numerous experiments. In these experiments, the validity of “the gadget of mass m moves with acceleration a when a force F is applied to it” is verified, because we can draw the conclusion inductively from our observations as well as deductively from “F=ma”. Meanwhile, the validity of “F=ma” gets reinforced when the conclusions deductively linked to it get repeatedly confirmed by empirical experiences.
Lastly, we are able to examine the validity of other arguments listed in Section I using the method of multiple verification. “All objects on my desk are conductive” is clearly not a good inductive argument, for it cannot be verified deductively. “Everyone who has travelled for 24 hours is very tired” may be a relatively good inductive argument, for it is possible for us to find some biological evidence supporting it. Arguments (vii) and (viii) can be seen as derived from the general inductive argument that “all organized constructions are fruits of human activity”, yet the validity of each case should be measured by whether sufficient evidence can be found in support of such inferences. The last argument, that “the mechanic at the bike store told me that my brake was about to fail”, is tricky. The mechanic in question may get the conclusion inductively from his previous experience of fixing bikes, and there might be some coherent theoretical explanations of the relationship between the mechanic’s observation and the failure of the brake. But if those theoretical explanations are not recognized by the mechanic, we still say that the conclusion that “the brake is about to fail” is no more valid than “the peasant is going to feed the chicken tomorrow”.
[1] Hume, David. An Enquiry Concerning Human Understanding. Charles River Editors, 2018. (Sec IV, Sec V)
[2] Russell, Bertrand. The Problems of Philosophy. OUP Oxford, 2001. (Chap 6)
[3] Strawson, P. F. Introduction to Logical Theory. Routledge, 2011. (Chap 9)
[4] Mellor, D. H. “Inaugural Lecture: The Warrant of Induction.” 1988.
[5] Goodman, N. The New Riddle of Induction. 1955.
[6] Curd, M., and Cover, J. A. Philosophy of Science: The Central Issues. New York: W. W. Norton, 1998. (pp. 409–411, pp. 426–432)
[7] Popper, K. R. “The Problem of Induction.” 1953.
Is there good reason to believe possible worlds other than the actual one exist? In this essay, I will argue that although possible worlds are a useful tool for understanding modality, there is no good reason to believe that possible worlds other than the actual one exist as concrete entities.
Logic and semantics are not all about categorical propositions, which depict things going on in the actual world. In our daily life, as well as in scientific and philosophical discussions, we make modal statements about non-existent events all the time. For instance, when we say a glass is fragile, we are not making a statement about its existing properties such as its color or shape; rather, we are making a postulation that it will break into pieces if dropped or hit. Likewise, when a glass is broken and we make a counterfactual statement that “if I had been more careful, the glass wouldn’t have been broken”, we are also not making a statement about what actually took place. In science, all dispositional properties of matter, such as “solubility”, “flammability”, and “tendency to decay”, are modal. And in philosophy, topics such as determinism, anti-realism, and supervenience can hardly be illustrated without appealing to modality.
To put it bluntly, modality can be regarded as ideas presented with modal quantifiers such as “necessarily” or “contingently”. “Trump is necessarily human”, “Trump was contingently the president of America”, “Necessarily, 2+2=4”, and “Contingently, the sun rises every day” are all statements put in the form of language which are true. Naturally, we say that Trump is necessarily human, because being human is his essence, and that he was contingently the president because history could have turned out otherwise—say, he did not run for the 2016 election, or that he quit due to some accident. Also, we say that “2+2=4” is necessarily true because it could not be otherwise; but “the sun rises every day” is only contingently true because we can think of a situation where the sun does not rise on some particular day. With the first two propositions, modal quantifiers are used to modify the object “Trump”, and with the latter two, quantifiers are used to modify the whole sentence. We refer to propositions resembling the former as de re modality, and ones resembling the latter as de dicto modality.
A fact can be true in de re modality, while false in the de dicto sense. Consider the two propositions constructed by Theodore Sider, or two propositions about emeralds:
(a, de dicto) Necessarily, the number of the planets is odd.
(b, de re) The number of the planets is such that it is necessarily odd.
(c, de dicto) Necessarily, emeralds are green.
(d, de re) Emeralds are necessarily green.
While the de re instances are presumably true in the above cases, it is difficult to admit the de dicto statements to be true in both cases, for there can be a situation where the number of the planets is even and some emeralds are not green. Statements as such are very different from truths such as “1+1=3” and “No squares are round”.
Yet telling what is necessarily true and what is only contingently true in de dicto modality is very tricky. A traditional and most widely acknowledged way to make such a distinction is to appeal to possible worlds, which is a concept that originated with Leibniz. Leibniz’s biconditionals state that:
(i) A proposition is necessary iff it is true in all possible worlds.
(ii) A proposition is possible iff it is true in some possible world.
From this perspective, we cannot accept (a) and (c) to be true, for there can be possible worlds where the number of the planets is even and some emeralds are red, blue, or yellow.
But what are possible worlds really? Philosophers’ explanations differ. Roughly, there are two parties: the concretist (or realist, as David Lewis referred to it in his book) party holding that possible worlds are as real as the actual world; and the abstractionist party presenting possible worlds as abstract entities. Interpreting possible worlds as real and concrete existence is indeed an ingenious invention and successfully dodged many problems posed to traditional theories. But like most philosophers, I hold the opinion that accepting the existence of concrete possible worlds causes greater trouble than advantages. The remaining part of this essay will be devoted first to clarifying Lewisian possible worlds and the reason to believe in them, and then discussing the trouble of believing in non-actual possible worlds.
The concretist idea of possible worlds is best demonstrated and defended in David Lewis’s remarkable work On the Plurality of Worlds. Lewis held the opinion that there are infinitely many possible worlds which are as real as the actual world in which we are living. In Stalnaker’s paper Possible Worlds, published in 1976, he concluded Lewis’s hyper-realist description of possible worlds in four points:
(i) Possible worlds exist.
Possible worlds do exist and they are as real as actual ones, even if they are not actual. This first statement is already peculiar enough, because when we are using modality in normal life—for instance, when we say that “the glass wouldn’t have broken if I had been more careful”, the situation which we have in mind is one that did not actually take place. In some terminology, we would regard it as ‘non-existent’. Lewis, nevertheless, grants the existence of such a situation.
(ii) Other possible worlds are things of the same sort as the actual world—“I and my surroundings”.
Here, “surroundings” can be interpreted as things that are spatiotemporally related to me, and the actual world is the sum of all spatiotemporally related beings. In the same manner, other possible worlds are ones in which all things are real and spatiotemporally related. On the other hand, two spatiotemporally unrelated beings are in two different possible worlds.
(iii) The indexical analysis of the adjective “actual” is the correct analysis.
Lewis argues that the actual world is different from other possible worlds only in the sense that it is the world where we reside. Beings in other possible worlds should consider the very world that they are in as actual, while ours as merely possible.
(iv) Possible worlds cannot be reduced to something more basic.
The abstractionist philosophers usually adopt the opinion that possible worlds are collections of smaller entities. For example, linguistic ersatzists think possible worlds are sets of non-contradictory sentences and combinatorialists see possible worlds as sets of spacetime points. But Lewis argues that possible worlds cannot be reduced to any sort of units. It is not reducible, just like the actual world cannot be reduced to more fundamental units.
But the points listed above are not exhaustive. To explain Lewisian possible worlds more completely, we also need to add that:
(v) There are no trans-world identities, but only counterparts.
In other words, in the Lewisian picture, possible worlds are absolutely isolated and individuals cannot travel from one world to another. When we are saying that “Trump could have lost the 2016 election”, what we refer to is neither Trump in the actual world, nor Trump himself in some possible world, but Trump’s counterpart in some possible world, who shares most characteristics with Trump, but differs from the actual Trump in a certain way.
The greatest strength of the concretist world theory is that it avoids the circularity most possibilia discussions confront. To be more specific, we introduce possible worlds in the first place to explain what is necessary in de dicto modality. Yet if we want to tell, in an abstractionist way, what has to be in possible worlds, the answer should be “what is necessarily true in the de dicto sense”. But concrete possible worlds seem to have resolved the problem. As possible worlds are real existences, what is in them is not to be explained by de dicto modality.
Nevertheless, given that concrete world theories manage to avoid this circularity, the problems they face tend to offset their advantage. One very important obstacle for them is that it is impossible for us to know what is happening in possible worlds. For one thing, real possible worlds are isolated from one another. We can turn to nothing but our imagination to tell what possible worlds are like. But being conceivable is by no means a justification for proper knowledge about what really exists. If we do want to grant conceivability as the justification of what exists in a possible world, and say that things exist in some possible worlds as long as we can think of them, we will again trap ourselves in the circularity problem other world theories share. For another thing, the number of possible worlds is infinite according to Lewis’s explanation. And to ascertain a proposition as necessary, we need to exhaustively examine every possible world, which is a task implausible to complete.
Additionally, according to the second character of the Lewisian possible worlds, it is impossible for there to exist two spatiotemporally related worlds, for there is not a single possible world which contains two spatiotemporally unrelated worlds. One possible way to reply to this objection, as Phillip Bricker suggested in Concrete Possible Worlds, is to allow plural analysis of worlds. But such permission to analyze two possible worlds together would lead to greater confusion. If we allow two possible worlds to be analyzed together, we can consequently allow three or more, and eventually expand to all possible worlds. On top of that, it would be possible for Trump to have won the 2016 election, and also have lost the 2016 election, if these two events happen to take place in the two possible worlds in question.
A third problem of Lewisian worlds is that the theory fails to give a representation of impossible things in modality. Since something is impossible, there is no way for it to exist in any real world. As Lewis himself indicated, there is no such thing as an impossible world. Therefore, to represent the impossibility of a proposition, we need, again, to exhaustively examine every possible world.
These are not all the problems Lewisian possible worlds have, but they are ones most likely to overthrow the achievements concretists may achieve in analyzing modality. There are other objections less tightly related to modality, yet equally important. For instance, if what exists in possible worlds are merely our counterparts, then it appears that we have no reason to care about our counterparts’ wellbeing. But under normal circumstances, when we are using modality, we do care about what will possibly happen. Also, Lewisian possible worlds seem to say that fictional beings such as talking donkeys, dragons breathing fire, and golden mountains do exist, which is difficult to accept.
The conclusion is that we probably cannot avoid adopting a circular analysis after all. Probably what is true de dicto necessarily is a priori. Things like true logical and mathematical propositions are true in themselves, and the introduction of possible worlds can never justify such truth. What we can do at best is to use possible worlds to help us think more clearly and systematically. For this purpose, the abstractionist version of possible worlds, or “paradise on the cheap”, is enough. Believing in the existence of real possible worlds comes at too high a price.
[1] Melia, J. Modality. Routledge, 2014.
[2] Lewis, D. “On the Plurality of Worlds.” Central Works of Philosophy, Volume 5: The Twentieth Century: Quine and After, 2005, 5: 246.
[3] Stalnaker, R. C. “Possible Worlds.” Noûs, 1976: 65–75.
[4] Lycan, W. G. “Possible Worlds and Possibilia.” 1998.
[5] Sider, T. “Reductive Theories of Modality.” The Oxford Handbook of Metaphysics, 2003.
[6] Mackie, P., and Jago, M. “Transworld Identity.” 2006.
[7] Menzel, C. “Possible Worlds.” 2013.
[8] Bricker, P. “Concrete Possible Worlds.” 2008.
This essay reviews Leibniz’s argument on the actual world being the best of all possible worlds.
One of the most impactful projects undertaken by Leibniz is delineating and demonstrating the actual world as the best of all possible worlds. Philosophers like Moore once argued that this metaphysical picture in Leibniz’s philosophy is valuable intrinsically, as it illuminates us on the essences or natures of individual substances, including human beings. But that should not be considered the sole contribution of this project to his philosophy as a whole. For instance, it also bolsters his considerations in ethics.
Leibniz’s discussion of the best world is concentratedly reflected in his work Theodicy: Essays on the Goodness of God, the Freedom of Man and the Origin of Evil. The term “theodicy” here stands for “divine justice”. This question of the justice of God can be traced back to Epicurus. Epicurus wondered: if God is infinitely powerful and just, then why would there be any evil in this world? To admit that God wills the existence of evil is to say that God is malicious; while to admit that God does not will evil but is incapable of taking it away is to say that God is feeble, neither of which is coherent with how we normally perceive God to be. In Leibniz’s time, a request for solutions to Epicurus’ question became particularly pressing and urgent in Christianity, for on the one hand, the Renaissance and the Scientific Revolution gave rise to groups of intelligent pagans who advocated polytheism, and on the other hand, Islamic nations challenged Christianity in Europe with their monotheistic religion. On top of the emerging new beliefs, people’s moral beliefs started to shake as well. Having witnessed the innocent suffering and evildoers fleeing from punishment, people naturally started to wonder whether being a successful criminal is a better strategy than being virtuous. In this context, Leibniz’s aim can be seen as unifying various theological considerations and reinforcing Christian philosophy, which guides its believers to act virtuously.
Epicurus’ question on evil can be separated into two parts: if evil exists, one has to either admit that God is not omniscient, or that God is not just and benevolent. These two aspects of the question are referred to by Murray as “the Holiness Problem” and “the Underachiever Problem” in the SEP entry “Leibniz on the Problem of Evil”. The main body of this essay will recount Leibniz’s project of reconciling the existence of evil and the perfection of God by focusing on how Leibniz deals with the two aspects of the problem. Then, in the closing part, I will briefly mention several problems which I find compelling and which may be difficult for Leibniz to answer.
It is argued that the holiness of God is stained by the existence of evil because he does not will the best and therefore is not morally noble. As a matter of fact, accepting that God is not morally just or benevolent is how Spinoza answers Epicurus’ question. God in Spinoza’s sense is not some entity that can be applied moral predicates, but nature itself. Then, if God does not necessarily will the best, the existence of evils is compatible with the idea of a Spinozan God. But the denial of God’s benevolence will lead to moral skepticism, which is a result not acceptable to Leibniz and many other Christian philosophers.
To understand whether God wills the good, it is essential to understand what goodness is to God. It is little disputed that Leibniz interprets what God takes to be good as “harmony”. But if we further enquire what “harmony” is, opinions differ. For irrational entities, it is argued that the greatest harmony is either (a) the maximization of the essences of substances, or (b) the simplest laws governing the greatest varieties of phenomena. While for rational beings, harmony is seen as the maximization of total happiness or virtue. It seems that these three interpretations can be linked to three different kinds of perfection distinguished by Leibniz, with metaphysical and physical perfection incorporated by (a) and (b), while moral perfection is incorporated by (c).
Let us consider metaphysical goodness and physical goodness first. To Leibniz, every existence but God is metaphysically imperfect. The metaphysically perfect God uses monads, or continuous immaterial atoms, to create the world, which are in place such that each one of them is in harmony with every other actualized monad. The “harmony” here, according to Moore, is a monad’s cohering with other monads, or two monads’ coexisting in a world with the same content. But this coherence in monads does not constitute enough reason for God to create. Additionally, God wills the actual world because of different degrees of perfection in actual substances. Hence, to create the best world, God should create as many substances as possible. Meanwhile, just as geometers and architects apply the most ingenious rules in accomplishing their goals, God uses the simplest hypotheses to build the best of all possible worlds. Leibniz barely provided any further illustration of “simplicity”, yet Rutherford once indicated that “simplicity” can be considered as degrees of unification. That is, the more phenomena can be illustrated by a single natural law, the more harmony there is.
In another light, the greatest harmony can be understood as the greatest moral perfection in rational beings. A point worth noticing is that Leibniz does not regard happiness as the highest or sole aim for God when creating the world. He said in the Theodicy: “The happiness of rational creatures is one of the aims (God) has in view; but it is not his whole aim, nor even his ultimate aim.” Neither is virtue the bare aim for God. Nevertheless, the greatest moral perfection for rational beings and the greatest happiness and virtue form a subtle balance with each other such that neither can be achieved without the achievement of the other. Happiness can only persist in the presence of virtue, which means acting in accordance with wisdom, and altogether they constitute a state where moral perfection resides.
In accordance with the three perfections mentioned above are three different types of evils—metaphysical, physical, and moral. The first type does not bother Leibniz much, as it is only absent in God himself. But physical evils (sufferings) and moral evils (sins) do require our closer attention, as they may be absent in some possible worlds. The presence of these evils in the actual world, Leibniz argues, is still not sufficient for us to accuse God of having a malicious will, for God does not cause them to exist, but only permits them to be there. Additionally, Leibniz maintains that evils are not even existences, but merely deprivation or lack of being.
Now that the doubt on God’s willingness to create the best world is cleared, we just need to show that the actual world is better than any other possible world, and also better than nothingness, to answer Epicurus’ question.
Reasons why one may reject the actual world as the best one are plain to conceive. First of all, human beings suffer from seemingly dispensable catastrophes, regardless of their qualities and virtues—earthquakes, floods, droughts—all of which may be disposed of in other possible worlds, and human suffering thereby alleviated. Moreover, quarrels, fights, and betrayals prevail in the world in almost any time period. If God is benevolent, why not create every rational being as wise and virtuous, and eliminate disputes from the world?
One approach to dismissing such objections is to say that sufferings and sins are by-products of goodness. If one has never tasted bitterness, he will never know that sweetness is good. Likewise, goodness is only manifest when sufferings are known to rational beings. In other words, sufferings prepare us for greater happiness, and the happiness gained outweighs the suffering human beings encounter overall, which makes the existing world more desirable than nothing. Additionally, there are moral reasons to assume this world to be perfect, where sins are punished and good deeds are rewarded for a spirit by God in some unperceivable ways, for in the mundane world, crimes go unpunished and sacrifices go neglected at times. A denial of the best and just world would devalue what is morally good.
But still, some sufferings are so unintelligible that we cannot conceive them as the slightest good—for instance, a flood that inundates hundreds and thousands of people. Leibniz would respond to this by saying that such suffering is not harmonious except in the limited picture that people as limited beings may perceive, but appears harmonious to God, who is an omniscient infinite being. As Leibniz described in On the Ultimate Origination of Things:
We know little but a small part of the eternity which extends without measure, for how short is the memory of several thousand years which history gives us… Look at a very beautiful picture, and cover it up except for some small part. What will it look like but some confused combination of colours, without delight, without art; indeed, the more closely we examine it the more it will look that way. But as soon as the covering is removed, and you see the whole surface from an appropriate place, you will understand that what looked like accidental splotches on the canvas were made with consummate skill by the creator of the work. (AG 153)
God and God’s will in this sense remain incomprehensible and unspeakable to human beings, but this attempt to place finite human intelligence before infinite divine intelligence leaves huge ground for Christian theology to argue.
In general, to answer Epicurus’ question and reinforce Christianity, Leibniz provided a relatively coherent metaphysical picture. He asserts that the actual world which God created is the best of all possible worlds—most perfect in both physical and moral senses according to God’s will. The presence of evils in this actual world does not prevent it from being the best one, for evils are by-products of goodness, and even prepare us for the better. Even if some evils appear unintelligible as good, it is only because we are not able to comprehend them due to the finitude of human nature.
Leibniz’s whole project of demonstrating the world as the best of all possible worlds brings us to a stage at which we can neither verify nor falsify, but only believe. Although hardly any fatal fault can be found in his demonstration, Leibniz’s project may sit in an awkward place in the Christian tradition. To illustrate, Christianity depicts heaven as a place of eternal bliss and permanent happiness. If Leibniz accepts the existence of heaven, then would it not be implausible for this earthly world to be created? On the other hand, if he rejects the idea of heaven, then more arduous work would be needed to depict the unperceived afterlife or future world. Also, his project can only persuade us that it is possible for this world, where evils are present, to be the best of all possible worlds; yet it has not given any hint as to why it is impossible for a world devoid of evils to be better than the existing world. After all, human painters may fail to construct pictures harmonious both on the whole and in the part due to their limited capacity. But since God is omniscient, omnipresent, and omnipotent, why can’t he create a world that is not only harmonious on the whole, but also harmonious when viewed from every part of it?
[1] Leibniz, Gottfried Wilhelm, Roger Ariew, and Daniel Garber. Philosophical Essays. Indianapolis; Cambridge: Hackett, 1989.
[2] Murray, Michael. “Leibniz on the Problem of Evil.” Stanford Encyclopedia of Philosophy, 2013.
[3] Jolley, Nicholas. Leibniz. London: Routledge, 2005.
[4] Wilson, Catherine. Leibniz’s Metaphysics: A Historical and Comparative Study. Princeton: Princeton UP, 1989, ch. viii.
[5] Rutherford, Donald. Leibniz and the Rational Order of Nature. Cambridge: Cambridge UP, 1995, chapters 1–3.
[6] Lodge, Paul. “Theodicy, Metaphysics, and Metaphilosophy in Leibniz.” Philosophical Topics 43.1 (2015): 27–52.