Seth Bauer
Modern philosophy tends to view human beings as either entirely free or completely determined by natural causes. Perhaps the truth is somewhere in between. When it comes to a person’s ideas, it is likely that they are, to some extent, determined by environmental forces, most notably those surrounding their childhood. Based on the results of past social studies, there is good reason to believe that a home without a father has a negative effect on the child’s emotional development (McLanahan 2013). Conversely, the intact nuclear family appears to be a net benefit to the child’s emotional development in comparison to a fractured or non-traditional family. Up until now, no one appears to have attempted to trace the effect of fatherlessness on the intellectual life. My research question is whether the absence of a father during childhood leads to the development of a life philosophy incompatible with Christianity. For my research, I compiled a dataset of the one hundred most influential philosophers in history and compared the relationships between fatherlessness, marital relations, and philosophy. My results show that although there is not a statistically significant direct relationship between fatherlessness and ideology, the indirect relationship between the two is strong. It appears that fatherlessness influences one’s approach to marital relations which, in turn, influences philosophy.
Literature Review
Past research has studied the effect of absent fathers on the emotional development of children (McLanahan 2013), the long-term psychological effects in their adult years (2013), and the likelihood that the child will get married and have a healthy (traditional) marital relationship (Lang 2001). In all three cases, there appears to be a negative effect on individuals when their father is absent during their childhood.
McLanahan (2013) summarizes a large number of empirical studies performed using six different models.[1] She finds strong evidence that a father’s absence affects children’s social-emotional development, increases risky behavior in adolescents, and suggests that psychological harm can persist throughout the course of one’s life. Kevin Lang and Jay L. Zagorsky add a further consideration in an article entitled “Does Growing up with a Parent Absent Really Hurt?” (2001). This study examines the relationship between “years lived with mother,” “years lived with father,” and six outcome variables.[2] The results indicate that a father’s death (absence) greatly reduces the probability that his son will be married. I expected to find this same relationship within my own data set. According to Gary D. Sandefur and Thomas Wells, family structure also has a measurable effect on one’s likelihood to pursue further education, but it is not the most critical factor (1999).[3] Since the sample for my own study consists of philosophers who attained a high degree of intellectual achievement, this is not a perinate relationship in my study since this variable is naturally controlled. However, it is important to keep in mind that the intellectual tenacity of philosophers with absent fathers may be exceptional in itself.
Theory
Prior studies have studied the short-term and long-term psychological effects of children who grow up with absent fathers. However, none have attempted to make a connection between this childhood experience and the formulation of a consistent system of philosophy. At the very least, I expected to find a relationship between an absent father and the likelihood that the philosopher would get married, which would be consistent with the results of Lang and Zagorsky’s study. Furthermore without their father to provide them with an example of manhood, I expected to see that these philosophers are more likely to break from traditional norms due to adverse psychological effects consistent with McLanahan’s study. My initial theory was that just as there are short-term and long-term psychological effects on fatherless children, so also would it have an intellectual effect on one’s life philosophy. The reasoning here is that ideas are formed and made sense of by a human mind inseparably linked to one’s psychology. If this is indeed the case, we would expect to see a relationship between fatherlessness and marital choice as well as a direct relationship between fatherlessness and particular philosophies.
My secondary theory was that perhaps the intellectual effect on a philosopher’s ideas do not come directly from the experience of not having a present father, but are rather derived from it indirectly through the intermediate variable of their approach to marital life. According to this theory, three variables are at work in a causal chain: fatherlessness => practice (approach to marital relations) => ideas. Fatherlessness is clearly an independent variable because it happens prior to all of the other variables, but it is harder to make the case that one’s activity affects ideas. What if it is actually ideas that influence a philosopher’s marital approach? If this were the case, the relationship would look like this: fatherlessness => practice (approach to marital relations) <= ideas. Before testing this theory, this objection needs to be resolved theoretically.
A possible response to this objection is that in this study we are dealing with a fully formed philosophical system. According to Lawrence Kohlberg, an individual begins life with an attitude that is naturally self-serving, and this motivates one’s initial actions (Barron 58). Gradually, the individuals develop beyond this stage as they compile a set of moral principles which they will use to justify their past and future actions (Barron 58). Because our perception of the world is not entirely clear and is influenced by strong psychological forces, people have a tendency to form beliefs that justify actions they have already done or decisions they have already made (King, 218). In 2000, Dolores Albarracín published an article entitled “The Cognitive Impact of Past Behavior: Influences on Beliefs, Attitudes, and Future Behavioral Decisions,” where she writes that the “the causal impact of past behavior on later decisions was clearly established under the conditions we investigated” (20). This study suggests that people do not make decisions in the abstract, and neither do their ideas come into being without arising from a previous encounter with the world. Although there is undoubtably interplay between the two variables, it would seem that philosophers’ approach to marital relations is less influenced by their ideas as it is itself a source of influence on their philosophy.
Data and Methodology
For my own study, I compiled an original data set on the one hundred most influential philosophers in history, using a publication by Encyclopedia Britannica as my starting point. After removing all people from the list who were still alive in order to mitigate selection bias, I cross-referenced this work with Peter Kreeft’s book on the same topic, which I used to fill in a few missing 20th century philosophers. My database ended up with ninety-nine philosophers, and the sample naturally controls for intellectual ability, since it is fair to assume that all of these philosophers were highly intelligent. The independent variable in my dataset is “father_absent,” and my dependent variables are “married,” “sexual_ethics,” “grouped_ideology,” and “metaphysics_epistemology_scale.” I also entertain the possibility that “grouped_ideology” could be an independent variable for “sexual_ethics” and test this relationship later on in my study. Because my sample size was relatively small, I expected statistical significance to be difficult to achieve, although the Lamba measures of association would be quite high if there was indeed a relationship between variables.
Definitions
“Father_absent” is coded 0 for not absent and 1 for absent during childhood. “Married” indicates whether or not the philosopher was married, with a code of 0 for not married, 1 for married, and 2 for religious celibacy. “Sexual_ethics” indicates one’s approach to marital relations. A score of 0 indicates that a philosopher had a non-traditional view of marriage. This is determined by the philosopher having known sexual relations outside of marriage, marrying another woman while his wife is still alive, and/or explicitly approving of this activity. A score of 1 indicates that the Philosopher had a traditional approach to and view of marriage. The “grouped_ideologies” category is a combination of the variables “Nihilist,” “Materialist,” and “Utilitarian,” meaning that a philosopher need only ascribe to one of these views to be included. For the purposes of this study, a “Nihilist” philosophy claims that life is purposeless, and a “Materialist” philosophy contends that nothing exists except physical, scientifically testable things, and a “Utilitarian” philosophy suggests that the only standard of right and wrong is achieving the greatest good for the greatest number. These views depart from the traditional Christian belief that there is purpose and meaning to life, a spiritual world exists, and there are moral principles other than achieving the greatest good for the greatest number.
Analysis
After running crosstabulations on the relevant variables in a variety of variable combinations, the results looked promising. The relationship between “father_absent” and “married” was highly significant at <.001, confirming the previous studies of Lang and Zagorsky’s study (Fig 1). The levels of association for this relationship are quite high, with Lamba levels at around 34%. Furthermore, the similar relationship between “father_absent” and “sexual_ethics” was also statistically significant at <.001 as was expected (Fig 2.), “father_absent” during childhood is an excellent predictor of a philosopher’s either traditional or non-traditional approach to marriage and celibacy with Lamba levels of 28-30%. This result appeared to confirm the study by McLanahan.
Yet when testing my initial hypothesis, I found that the direct relationship between “father_absent” and the variable “grouped_ ideology” (whether one was a utilitarian, materialist, or a nihilist) was not statistically significant (Fig 3.). Lack of statistical significance does not mean that no relationship exists, but rather that there is not a great enough chance that these results would not be found by chance in the general population.
I then tested my second theory that proposed an indirect relationship between the same two variables, with “sexual_ethics” as an intermediary variable.. With a strong causal relationship already established between “father_absent” and “sexual_ethics,” I ran a crosstabulation on the effect of the variable “sexual_ethics” on “grouped_ideology.” The results were highly statistically significant at <.001 with Lamba measures of association of .172 for “grouped_ideology” and .082 for “sexual_ethics,” showing that a non-traditional approach to marriage (“sexual_ethics”) can explain Nihilism, Materialism, and Utilitarianism more than twice as well (17.2%) than vice versa (Fig. 4). According to these results, a philosopher with a non-traditional approach to marriage and celibacy has a 78% chance of being a Nihilist, Materialist, or Utilitarian, while a philosopher with a traditional approach has a 60% chance of being one of those three ideologies. These results support the theory that there is an indirect link between fatherhood and philosophy.
Earlier, I proposed that, in the causal chain of “father_absent,” “sexual_ethics,” and “grouped_ideology,” ideology might itself be an independent variable affecting “sexual_ethics,” which would undermine the causal direction suggested by my second theory. To confirm that grouped_ideology was indeed a dependent variable, I used “grouped_ideology” as a control variable for the relationship between “father_absent” and “sexual_relations” using (Fig 5.). The results remained statistically significant at <.001 and the symmetric Lamba levels remained at .30, providing strong evidence that fatherlessness rather than ideas affect one’s approach to marital relations. In the context of this paper’s broader argument, this crosstabulation provides substantial evidence that the causal chain is indeed: fatherlessness => practice (approach to marital relations) => ideas (combined three categories).
Conclusion
The results of my study show that there is a statistically significant indirect relationship between fatherlessness and particular philosophies, linked by one’s approach to marital relations. I confirm the findings of McLanahan, Lang, and Zagorsky’s studies within my own dataset, which makes it more likely that the findings of my sample provide meaningful insight into the general population. According to these results, it would seem that ideas are not quite as free as we might think and that they are significantly influenced by two major factors: (1) one’s childhood environment (in the form of fatherlessness), and (2) one’s past actions (namely one’s approach to marital relations). Philosophy, the currency of ideas, is arguably a much more formidable force than raw political power. We would do well to better understand the impact marital choices and the intact nuclear family has upon society at large and most especially upon ideas and ideology.
| (1.1) Was the subject married? * Father absent during childhood? Crosstabulation | |||||
| Father absent during childhood? | Total | ||||
| No | Yes | ||||
| Was the subject married? | Never Married | Count | 15 | 8 | 23 |
| % within Father absent during childhood? | 24.6% | 42.1% | 23.2% | ||
| Married | Count | 38 | 11 | 49 | |
| % within Father absent during childhood? | 62.3% | 57.9% | 49.5% | ||
| Celibate | Count | 4 | 0 | 6 | |
| % within Father absent during childhood? | 6.6% | 0.0% | 6.1% | ||
| Total | Count | 61 | 19 | 99 | |
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | ||
| (1.2) Chi-Square Tests | |||
| Value | df | Asymptotic Significance (2-sided) | |
| Pearson Chi-Square | 72.815a | 6 | <.001 |
| Likelihood Ratio | 74.533 | 6 | <.001 |
| N of Valid Cases | 99 | ||
| a. 7 cells (58.3%) have expected count less than 5. The minimum expected count is 1.15. | |||
| (1.3) Symmetric Measures | |||
| Value | Approximate Significance | ||
| Nominal by Nominal | Phi | .858 | <.001 |
| Cramer’s V | .606 | <.001 | |
| N of Valid Cases | 99 | ||
| (1.4) Directional Measures | ||||||
| Value | Asymptotic Standard Errora | Approximate Tb | Approximate Significance | |||
| Nominal by Nominal | Lambda | Symmetric | .341 | .076 | 3.782 | <.001 |
| Was the subject married? Dependent | .340 | .067 | 4.530 | <.001 | ||
| Father absent during childhood? Dependent | .342 | .098 | 2.960 | .003 | ||
| Goodman and Kruskal tau | Was the subject married? Dependent | .288 | .040 | <.001c | ||
| Father absent during childhood? Dependent | .323 | .067 | <.001c | |||
| a. Not assuming the null hypothesis. | ||||||
| b. Using the asymptotic standard error assuming the null hypothesis. | ||||||
| c. Based on chi-square approximation | ||||||
| (2.1) Traditional approach to marriage and celibacy * Father absent during childhood? Crosstabulation | |||||
| Father absent during childhood? | Total | ||||
| No | Yes | ||||
| Traditional approach to marriage and celibacy | Non-traditional | Count | 16 | 9 | 25 |
| % within Father absent during childhood? | 26.2% | 47.4% | 25.3% | ||
| Traditional | Count | 39 | 9 | 50 | |
| % within Father absent during childhood? | 63.9% | 47.4% | 50.5% | ||
| Total | Count | 61 | 19 | 99 | |
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | ||
| (2.2) Chi-Square Tests | |||
| Value | df | Asymptotic Significance (2-sided) | |
| Pearson Chi-Square | 58.124a | 4 | <.001 |
| Likelihood Ratio | 54.021 | 4 | <.001 |
| N of Valid Cases | 99 | ||
| a. 4 cells (44.4%) have expected count less than 5. The minimum expected count is 4.61. | |||
| (2.3) Symmetric Measures | |||
| Value | Approximate Significance | ||
| Nominal by Nominal | Phi | .766 | <.001 |
| Cramer’s V | .542 | <.001 | |
| N of Valid Cases | 99 | ||
| (2.4) Directional Measures | ||||||
| Value | Asymptotic Standard Errora | Approximate Tb | Approximate Significance | |||
| Nominal by Nominal | Lambda | Symmetric | .299 | .093 | 2.785 | .005 |
| Traditional approach to marriage and celibacy Dependent | .306 | .103 | 2.545 | .011 | ||
| Father absent during childhood? Dependent | .289 | .106 | 2.357 | .018 | ||
| Goodman and Kruskal tau | Traditional approach to marriage and celibacy Dependent | .265 | .055 | <.001c | ||
| Father absent during childhood? Dependent | .264 | .071 | <.001c | |||
| a. Not assuming the null hypothesis. | ||||||
| b. Using the asymptotic standard error assuming the null hypothesis. | ||||||
| c. Based on chi-square approximation | ||||||
Figure 2. Crosstabulations on the effect a father’s absence during childhood has on one’s future approach to sexual ethics. Once again, the approximate significance is <.001. The value of the measures of association is also quite high.
| (3.1) Was the subject a Materialist, Utilitarian, or a Nihilist? * Father absent during childhood? Crosstabulation | |||||
| Father absent during childhood? | Total | ||||
| No | Yes | ||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | No | Count | 40 | 13 | 70 |
| % within Father absent during childhood? | 65.6% | 68.4% | 70.7% | ||
| Yes | Count | 21 | 6 | 29 | |
| % within Father absent during childhood? | 34.4% | 31.6% | 29.3% | ||
| Total | Count | 61 | 19 | 99 | |
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | ||
| (3.2) Chi-Square Tests | |||
| Value | df | Asymptotic Significance (2-sided) | |
| Pearson Chi-Square | 4.055a | 2 | .132 |
| Likelihood Ratio | 4.709 | 2 | .095 |
| N of Valid Cases | 99 | ||
| a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 5.57. | |||
| (3.3) Symmetric Measures | |||
| Value | Approximate Significance | ||
| Nominal by Nominal | Phi | .202 | .132 |
| Cramer’s V | .202 | .132 | |
| N of Valid Cases | 99 | ||
Figure 3. Crosstabulation results for the direct effect of a father’s absence during childhood upon the variable “grouped_ideology” (whether one was a utilitarian, materialist, or a nihilist). The direct relationship is not statistically significant.
| (4.1) Was the subject a Materialist, Utilitarian, or a Nihilist? * Traditional approach to marriage and celibacy Crosstabulation | |||||
| Traditional approach to marriage and celibacy | Total | ||||
| Non-traditional | Traditional | ||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | No | Count | 10 | 39 | 70 |
| % within Traditional approach to marriage and celibacy | 40.0% | 78.0% | 70.7% | ||
| Yes | Count | 15 | 11 | 29 | |
| % within Traditional approach to marriage and celibacy | 60.0% | 22.0% | 29.3% | ||
| Total | Count | 25 | 50 | 99 | |
| % within Traditional approach to marriage and celibacy | 100.0% | 100.0% | 100.0% | ||
| (4.2) Chi-Square Tests | |||
| Value | df | Asymptotic Significance (2-sided) | |
| Pearson Chi-Square | 15.933a | 2 | <.001 |
| Likelihood Ratio | 15.315 | 2 | <.001 |
| N of Valid Cases | 99 | ||
| a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 7.03. | |||
| (4.3) Symmetric Measures | |||
| Value | Approximate Significance | ||
| Nominal by Nominal | Phi | .401 | <.001 |
| Cramer’s V | .401 | <.001 | |
| N of Valid Cases | 99 | ||
| (4.4) Directional Measures | ||||||
| Value | Asymptotic Standard Errora | Approximate Tb | Approximate Significance | |||
| Nominal by Nominal | Lambda | Symmetric | .115 | .110 | 1.005 | .315 |
| Was the subject a Materialist, Utilitarian, or a Nihilist? Dependent | .172 | .157 | 1.005 | .315 | ||
| Traditional approach to marriage and celibacy Dependent | .082 | .100 | .787 | .431 | ||
| Goodman and Kruskal tau | Was the subject a Materialist, Utilitarian, or a Nihilist? Dependent | .161 | .079 | <.001c | ||
| Traditional approach to marriage and celibacy Dependent | .070 | .037 | .001c | |||
| a. Not assuming the null hypothesis. | ||||||
| b. Using the asymptotic standard error assuming the null hypothesis. | ||||||
| c. Based on chi-square approximation | ||||||
Figure 4. Crosstabulation results for the effect of “sexual_ethics” (a traditional approach to marriage and celibacy) upon “grouped_ideology.” Note that the results become statistically significant once again.
| (5.1) Traditional approach to marriage and celibacy * Father absent during childhood? * Was the subject a Materialist, Utilitarian, or a Nihilist? Crosstabulation | ||||||||||||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | Father absent during childhood? | Total | ||||||||||||
| No | Yes | |||||||||||||
| No | Traditional approach to marriage and celibacy | Non-traditional | Count | 7 | 3 | 10 | ||||||||
| % within Father absent during childhood? | 17.5% | 23.1% | 14.3% | |||||||||||
| Traditional | Count | 28 | 9 | 39 | ||||||||||
| % within Father absent during childhood? | 70.0% | 69.2% | 55.7% | |||||||||||
| Total | Count | 40 | 13 | 70 | ||||||||||
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | |||||||||||
| Yes | Traditional approach to marriage and celibacy | Non-traditional | Count | 9 | 6 | 15 | ||||||||
| % within Father absent during childhood? | 42.9% | 100.0% | 51.7% | |||||||||||
| Traditional | Count | 11 | 0 | 11 | ||||||||||
| % within Father absent during childhood? | 52.4% | 0.0% | 37.9% | |||||||||||
| Total | Count | 21 | 6 | 29 | ||||||||||
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | |||||||||||
| Total | Traditional approach to marriage and celibacy | Non-traditional | Count | 16 | 9 | 25 | ||||||||
| % within Father absent during childhood? | 26.2% | 47.4% | 25.3% | |||||||||||
| Traditional | Count | 39 | 9 | 50 | ||||||||||
| % within Father absent during childhood? | 63.9% | 47.4% | 50.5% | |||||||||||
| Total | Count | 61 | 19 | 99 | ||||||||||
| % within Father absent during childhood? | 100.0% | 100.0% | 100.0% | |||||||||||
| (5.2) Chi-Square Tests | ||||||||||||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | Value | df | Asymptotic Significance (2-sided) | |||||||||||
| No | Pearson Chi-Square | 36.664b | 4 | <.001 | ||||||||||
| Likelihood Ratio | 37.078 | 4 | <.001 | |||||||||||
| N of Valid Cases | 70 | |||||||||||||
| Yes | Pearson Chi-Square | 25.041c | 4 | <.001 | ||||||||||
| Likelihood Ratio | 19.150 | 4 | <.001 | |||||||||||
| N of Valid Cases | 29 | |||||||||||||
| Total | Pearson Chi-Square | 58.124a | 4 | <.001 | ||||||||||
| Likelihood Ratio | 54.021 | 4 | <.001 | |||||||||||
| N of Valid Cases | 99 | |||||||||||||
| a. 4 cells (44.4%) have expected count less than 5. The minimum expected count is 4.61. | ||||||||||||||
| b. 3 cells (33.3%) have expected count less than 5. The minimum expected count is 1.86. | ||||||||||||||
| c. 7 cells (77.8%) have expected count less than 5. The minimum expected count is .21. | ||||||||||||||
| (5.3) Symmetric Measures | ||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | Value | Approximate Significance | ||
| No | Nominal by Nominal | Phi | .724 | <.001 |
| Cramer’s V | .512 | <.001 | ||
| N of Valid Cases | 70 | |||
| Yes | Nominal by Nominal | Phi | .929 | <.001 |
| Cramer’s V | .657 | <.001 | ||
| N of Valid Cases | 29 | |||
| Total | Nominal by Nominal | Phi | .766 | <.001 |
| Cramer’s V | .542 | <.001 | ||
| N of Valid Cases | 99 | |||
| (5.4) Directional Measures | |||||||
| Was the subject a Materialist, Utilitarian, or a Nihilist? | Value | Asymptotic Standard Errora | Approximate Tb | Approximate Significance | |||
| No | Nominal by Nominal | Lambda | Symmetric | .377 | .105 | 2.983 | .003 |
| Traditional approach to marriage and celibacy Dependent | .419 | .101 | 3.404 | <.001 | |||
| Father absent during childhood? Dependent | .333 | .122 | 2.320 | .020 | |||
| Goodman and Kruskal tau | Traditional approach to marriage and celibacy Dependent | .307 | .080 | <.001c | |||
| Father absent during childhood? Dependent | .262 | .081 | <.001c | ||||
| Yes | Nominal by Nominal | Lambda | Symmetric | .227 | .214 | .943 | .346 |
| Traditional approach to marriage and celibacy Dependent | .286 | .283 | .864 | .388 | |||
| Father absent during childhood? Dependent | .125 | .203 | .581 | .561 | |||
| Goodman and Kruskal tau | Traditional approach to marriage and celibacy Dependent | .324 | .055 | .001c | |||
| Father absent during childhood? Dependent | .313 | .114 | .002c | ||||
| Total | Nominal by Nominal | Lambda | Symmetric | .299 | .093 | 2.785 | .005 |
| Traditional approach to marriage and celibacy Dependent | .306 | .103 | 2.545 | .011 | |||
| Father absent during childhood? Dependent | .289 | .106 | 2.357 | .018 | |||
| Goodman and Kruskal tau | Traditional approach to marriage and celibacy Dependent | .265 | .055 | <.001c | |||
| Father absent during childhood? Dependent | .264 | .071 | <.001c | ||||
| a. Not assuming the null hypothesis. | |||||||
| b. Using the asymptotic standard error assuming the null hypothesis. | |||||||
| c. Based on chi-square approximation | |||||||
Figure 5. Crosstabulation results for the effect of “father_absent” upon “sexual_ethics” while controlling for “grouped_ideology.” The results remain statistically significant, suggesting that sexual ethics causally proceeds ideology rather than being caused by it.
[1]These models include OLS (ordinary least squares or logistic regression), LDV (lagged dependent variable), GCM (growth curve model), IFE (individual fixed effects), SFE (sibling fixed effects), and natural experiment.
[2]Namely, cognitive ability, education, marital status, earnings, household income and household wealth.
[3] Since my sample is made up the world’s most influential (and presumably intelligent) philosophers, I have intentionally chosen to avoid pursuing this particular relationship in my study. However, it is important to keep in mind that the intellectual ability of philosophers with absent fathers may be expectational in this way as well.