Rethinking and Revising Goode’s Contribution to Global Family Change

Abstract

This essay provides a comprehensive overview of William J. Goode’s contribution to the study of global family and social change. I begin by describing Goode’s theoretical perspectives and outlining his theses dating back to the 1960s. I then provide an assessment of where and why some of his predictions proved wrong and elaborate on what we have learnt on changes in families at the global level over the past half century. Lastly, I speculate on how Goode would rethink his arguments nowadays in light of fifty years of new evidence and scholarly developments – both theoretical and methodological. In so doing, I highlight shortcomings of current approaches and outline directions for future family research and theorizing.

Introduction

Half a century has passed since William J. Goode published World Revolution and Family Patterns (1963), one – or perhaps, the most – influential study on changes in families at the global level to date. His book was influential not only because it raised the status of family sociology as an established and well-respected field of study, but also because the breadth, scope, and boldness of its arguments had few scholarly precedents. Goode’s work provides a historical and cross-cultural study of family change around the world and traces linkages with processes of industrialization that were unfolding at the time. According to Cherlin (2012), his descriptions of patterns in the West, complemented by trends in India, China, Japan, Africa, and the Middle-East, make it the “most comprehensive study of the family to have appeared since Edward Westermarck’s 1891 treatise on the evolution of family life, the History of Human Marriage” (p. 577). This said, in light of the fact that many of Goode’s predictions did not realize and the world changed in radical – and often unexpected – ways, the book has often been criticized and deemed a relict of the past. Given how unique a contribution to social theory Goode’s work provided, I do not agree with this view, and I will outline the reasons in what follows.

This review is organized as follows. First, I will briefly describe Goode’s perspective. Second, I will give an assessment of where and why Goode’s predictions proved wrong. Third, and related to the above, I will provide an overview of what we have learnt on global family change over the past half century. Fourth, I will speculate on how Goode would rethink his argument(s) nowadays in light of what we have learnt from recent scholarly developments. Ultimately, the aim of this reflection piece is to critically rethink and revise Goode’s contribution to the study of family change by discussing and highlighting shortcomings against which better theory has been, is, and will be produced.

Goode’s perspective

Goode (1963) predicted that with the spread of industrialization other parts of the world would embrace the Western model of the family, namely the “conjugal” family form characterized by strong marital bonds and weaker bonds across generations and within lineages. In his words: ‘whether the economic system expands through industrialization […] extended kinship ties weaken, lineage patterns dissolve, and a trend toward some form of the conjugal system generally begins to appear – that is, the nuclear family becomes a more independent kinship unit.” (p. 6). Goode’s argument rests on the idea that the conjugal family form is the one that fits “best” with advanced forms of industrialization, as a job-based economy requires independence, extended education, and a more mobile labor force – features that somewhat undermine the traditional family form relying on the control of elders, patriarchy, and arranged marriage (Furstenberg, 2013). In short, Goode believed that young adults would rationally choose to form conjugal families because these were ideally suited to the industrial age (Cherlin, 2012). Evidence of rapid economic growth and the development of a modern economy had already moved beyond the West in the early post-War era to parts of Asia, just as Goode was completing his 1963 book, which contained a large array of extant data from 50 countries and analyzed the impact of family on societal change.

Theoretical and methodological weaknesses

As early as in the 1970s – soon after Goode completed his 1963 work – it became clear that most of Goode’s theoretical predictions were wrong. His view of family change proved inapplicable both theoretically and methodologically. Theoretically, Goode was writing at a time in which functionalism was the reigning sociological theory, and Davis (1959) and Parsons’ (1943) perspectives were shaping most scholars’ minds. Functionalism, which aligns with the view that all aspects of society – including the family – serve a specific function and are necessary for its survival, supported the idea that the conjugal family was well-suited to the needs of industrialized societies as partners could move freely, and husbands could follow the needs of the labor market, hence aiming for occupational and personal success. Functionalism also featured in Goode’s paper on the power of love (1959), which suggested that random mating – or free choice – is often ‘controlled’ by older kin as a way to preserve lineage patterns and the social stratification system, with this more likely to be the case among the upper strata because kin lines are of greater social importance to them (relative to lower strata). Relatedly, Goode’s argument was deeply rooted in a breadwinner-homemaker Beckerian view of the family, with the husband working for wages, and the wife caring for children at home, in a sort of stable equilibrium driven by within-couple skill complementarities. Lastly, in his claim that the non-Western world would eventually converge to patterns of family change observed in the West, Goode heavily drew on modernization theory, according to which underdeveloped societies would eventually transition from traditional to modern, aiming for a “higher” form of development (Thornton, 2005). Modernization theory, one of the major perspectives in sociology of national development and underdevelopment around the 1950s, seeks to identify the variables that contribute to social progress and development of societies and contends that societies develop in fairly predictable stages through which they become increasingly complex.

Despite Goode recognized that industrialization was occurring at different rates of speed and rejected a purely evolutionary theory of family change (Cherlin, 2012), he nevertheless accepted the argument that Western-style industrialization would spread around the world. Yet, disconfirming Parsonian, Beckerian, and modernization views, right after Goode published his work the family became more complex in several ways. After a historical “low,” divorce rates rose rapidly, fertility declined, the proportion of births occurring outside of marriage rose, family forms became more complex and diverse, women started entering the labor force, and the transition to adulthood became more protracted and complex. In short, Goode could not anticipate the unfolding of a Second Demographic Transition (SDT), at least as it was occurring in most Western societies (Lesthaeghe, 2010, 2014).

Methodologically, I can identify at least three weaknesses. First, Goode’s argument rests on a set of empirical propositions – some of them actually tested and some of them based on speculation – rather than on a coherent theoretical whole that predicts convergence in families worldwide. Although this is not a flaw per se, it becomes problematic when it leads to bold predictions such as the global convergence to the conjugal family form of the West. For global convergence to hold, each “component” of the family (e.g., fertility, marriage, inter-generational dynamics, intra-generational relationships, etc.) needs to become more similar across countries or regions of the world – an empirical investigation conducted in Pesando and the Global Family Change (GFC) team (in press). If only one component of family change becomes increasingly similar across societies – such as the decline in patriarchy – this is unlikely to lead to global convergence in family forms. Note that, with reference to this, there is ample scholarship that contends that not even in fertility there is evidence of global demographic convergence (e.g., Dorius, 2008; Wilson, 2001, 2011). Second is the assumption – not explicitly stated but subsumed in the text – that nation-states are the only appropriate units of analysis to study family change. Given the current globalization of production, the spread of technology and social media, and the increasing interconnectedness of countries through micro-, meso-, and macro-level networks, nation-states might be too small a unit to study changes in families. Similarly, in line with claims made by Cherlin (2010), Furstenberg (2014), and McLanahan (2004), there is extensive evidence of diverging family patterns across socio-economic groups within societies. In this case, nation-states might be too big a unit of analysis. Third, Goode views industrialization as sort of an exogenous force that shapes family change from “behind the curtain,” while scholars have come to increasingly recognize that the nuclear family could have been a consequence of industrialization as well as a cause (Ruggles, 2012). Indeed, it is worth mentioning that some of these limitations are a byproduct of the time in which Goode wrote his book.¹ At that time, nationally representative cross-national data and long time-series were very limited – almost non-existent for low- and middle-income countries (LMICs) – and methods to uncover causal relationships and account for reverse causation not widely adopted, at least by sociologists. In effect, it is not by chance that theories of family change were mostly produced by “reading history sideways” (Thornton, 2001, 2005).

Global family change over the past half century

So, where exactly did Goode’s predictions prove right (or wrong), and what have we learnt from the past half century of family change? In broad terms, we can claim that despite several mispredictions, Goode was correct in identifying at least two societal trends that seem to have been spreading – although not necessarily converged – across many societies and are rooted in the belief that structural factors do have an impact on families. First is the decline in patriarchy and the loosening of extended kinship ties related to the idea that children benefit less and less from remaining tied to their fathers. Ruggles (2012), who cautions against reaching hastened conclusions that rely on (improper) measures of multi-generational co-residence, claims himself that the changing configuration of the family was made possible by a shift in the balance of power within the family (namely, a defy of authority) created by the opening of economic opportunities. In his 2003 paper Goode states: ‘over a long historical period, very likely more than two centuries, the trend in family changes seems to be toward a weakening of family controls. Family law systems were gradually transformed […] in the same direction’ (p. 17). Second is the spread of Western ideologies related to couple autonomy, companionship and romantic love – which were already touched upon in his 1959 seminal paper on the theoretical importance of love. Recent studies show that these ideologies have spread broadly even in societies that have made little progress towards industrialization (Cherlin, 2012; Jensen & Oster, 2009; La Ferrara, Chong, & Duryea, 2012; Thornton, Pierotti, Young-DeMarco, & Watkins, 2014), as suggested by Goode. However, a comprehensive and well-structured assessment of the adequacy of Goode’s theoretical predictions requires distinguishing changes both by world region and by family domain.

Starting from a focus by region of the world, Goode’s predictions proved perhaps even less applicable to high-income Western societies than to non-Western societies. While he predicted that illegitimacy rates would drop in the West, births outside of marriage increased by about 40 percent over the 50-year period. While he assumed that little change would occur in age at marriage, the median age at marriage increased by seven to eight years for both men and women. While he predicted that fertility would remain at the current level of the time (around three children per woman), the total fertility rate in most Western societies dropped below the replacement level of 2.1, reaching lowest-low levels in some (mostly Southern) European countries (Kohler, Billari, & Ortega, 2002). While he did not mention any form of partnership before marriage, cohabitation as an alternative family form or as a precursor to marriage spread widely. While he predicted that women would have stayed home to do housework or care for children, they attained college degrees in higher proportions, gained greater economic independence, and took well-paying jobs. Furthermore, Goode did not predict any rise in the divorce rate, which in fact rose dramatically to the point that by 1980 about half of all American marriages were likely to end in divorce (Cherlin, 2012). On top of these failed predictions, single parenthood increased, together with same-sex couples, living-apart together (LAT), and transnational families – none of which Goode mentioned. Hence, if anything, the past 50 years of family change in Western societies have been consistent with a move away from the conjugal family form.

As far as non-Western societies are concerned, Goode’s predictions described properly some family changes observed across East Asia (mainly China and Japan) and the Middle-East and North Africa (MENA). Specifically, East Asia is the region that has experienced the most extensive economic development in tandem with the emergence of smaller families and declining parental control over individuals’ choice – which substantially reduced the prevalence of arranged marriages.² Overall, this region has seen the emergence of relationships where the emotional and romantic components take the main stage (Cherlin, 2012). Japan has instead followed more closely some of the SDT hallmarks of later marriage, increasing cohabitation, and greater union instability. Some MENA countries have also experienced changes that are consistent with Goode’s hypothesis, such as less hierarchical relationships between spouses (Engelen & Puschmann, 2012) and a decline in divorce rates in contexts where these have traditionally been very high, such as Egypt and Algeria (Tabutin & Schoumaker, 2005). Conversely, in India the conjugal family form has not emerged anywhere, and parents still play an important role in choosing their children’s marriage partners (Vogl, 2013), although children are given more say in the choice of their partner and the process has become more collaborative (Allendorf & Pandian, 2016; Fuller & Narasimhan, 2008). As in most other parts of Southeast Asia, although fertility has declined substantially, teenage marriage – often followed by teenage fertility – remains high. In sub-Saharan Africa (SSA) child marriage has declined (Koski, Clark, & Nandi, 2017), age at marriage has risen substantially (Grant & Furstenberg, 2007; National Research Council, 2005), yet fertility and polygamy remain high (Smith-Greenaway & Trinitapoli, 2014), and progress has been geographically uneven, particularly across Western Africa. Lastly, in Latin America – which Goode did not include in his seminal study – there is evidence of marked fertility decline, increasing cohabitation, and other SDT hallmarks (Esteve, Lesthaeghe, & Lopez-Gay, 2012; Esteve & Lesthaeghe, 2016), which do not align with any of Goode’s predictions.

Moving to specific family domains, I seek to devote a bit more space to areas such as (i) union instability and divorce, (ii) inter-generational relationships, and (iii) and transition to adulthood patterns. Beginning from the former (i), only in his follow-up work World Changes in Divorce Patterns (1993) Goode predicted a continuing rise in union dissolution, noting that at some point the weakening of the family would generate counterforces that would hamper this increase (Cherlin, 2017). In tandem with this, he hypothesized that as barriers to divorce fall because of changes in the law, social norms, and women’s independence, divorce would become more common among the poor than the wealthy, reversing the traditionally positive social class gradient. Recent work has confirmed that some of these late-coming predictions proved prescient, as the social class differential now exists, and the rate of formal divorce is negatively correlated with indicators of social class. However, as Cherlin notes (2017), the stable “high-divorce” societies described by Goode – such as Muslim nations in North Africa – have disappeared, and stable “low-divorce” societies – such as India and other areas in South Asia – have become very rare. The message Cherlin conveys is that we cannot focus on formal marriage and divorce anymore, but we ought to consider unions, union dissolutions, and re-partnering more broadly. In fact, if we only focus on formal divorce, we might be observing a partial international convergence to levels below the highest rates ever observed, although still moderately high (Cherlin, 2017). A recent study by Clark and Brauner-Otto in SSA (2015) aligns with this idea, hence disproving Goode’s hypothesis of rising divorce with industrialization.

As for intergenerational co-residence (ii), based on a long tradition of theories that postulate an inverse association between household complexity and economic development (Le Play, 1884), a sizable literature supports the view that co-residence of elderly with their children has declined across most of the world, in line with Goode’s predictions (1963). Goode actually heavily borrowed on Le Play’s classification of family systems as “joint” (patriarcale), “stem” (souche), or “nuclear” (instable). Although – as shown by Ruggles (2012) – conclusions in the area vary widely by type of measure chosen (individual versus household-level), perspective adopted (younger versus older-generation), and demographic adjustments (population composition, availability of kin, etc.), there is consistent evidence that intergenerational co-residence in North America and Northwest Europe began to drop in the 19^th century, with changes accelerating in the 20^th century and coinciding with declining authority of family patriarchs (Ruggles, 2012). While some of these patterns have also been observed in high-income East Asian societies, evidence from LMICs is more blurred. A recent study by Ruggles and Heggeness (2008) covering 15 developing countries found no clear trends in co-residence over the past decades, and a slight increase in intergenerational families headed by the older generation – the family configuration most similar to Le Play’s “stem” and “joint” family system. In these societies, evidence of stable family forms is more consistent with the importance of cultural norms, the costs and benefits of co-residence for each generation, and emerging housing concerns.

Lastly, focusing on transition to adulthood patterns (iii), the type of transition that Goode described for the Anglo-speaking countries and Western Europe was a swift and seamless passage into adulthood, characterized by completing education, entering the labor force, leaving the parental home, and marrying – all happening pretty much simultaneously (Furstenberg, 2013). Yet the transition to adulthood became more protracted, less uniform, and less orderly throughout the West, with young adults initiating sexual activity, completing education, and leaving the parental home at varying ages, based on a combination of economic (e.g., extension of education and differential labor market prospects for men and women), cultural (e.g., growing gender equality), institutional (e.g., increased availability of contraception or abortion laws), and ideological (e.g., open attitudes and sexual freedom) factors between and within countries. These factors led to a series of “packages” of family practices that can be categorized as follows: “early and orderly”, “early and disorderly”, “late and disorderly”, and “late and orderly” (Furstenberg, 2013). As noted by Furstenberg, there thus appears to be no general convergence to a single pattern of transition to adulthood within or across Western nations, and even less so across LMICs that are currently experiencing massive increases in schooling that translate into longer educational trajectories, which in turn shape other later-life transitions (Grant & Furstenberg, 2007). Interestingly, Furstenberg (2013) notes peculiar transition to adulthood patterns in some Asian countries, where marriage has been delayed yet not accompanied by increases in premarital sex, thus qualifying young adults’ transitions as “late and orderly.”

Rethinking Goode’s perspective today

Certainly, Goode would heavily rethink his thesis today. First of all, he would abandon his prediction of a global convergence to the conjugal family form of the West. The focus on convergence did in fact disappear in his follow-up book (1993) and subsequent papers (2003), yet claims such as ‘family structures and family law have been moving in the same direction all over the world over the past two centuries’ (2003, p. 15) remain.³ Perhaps he would not agree with the opposite claim made by Therborn that there are seven rooted family systems in the world which are ‘on the whole not converging and in some respect rather diverging; they will also characterize the world in the foreseeable future’ (2014, p. 3). This latter claim heavily stresses the indelibility of family norms and values, which Goode – and some other contemporary scholars such as Ruggles (2012) – would not agree with. After all, if there is a portion of Goode’s argument that proved more applicable to current low- and high-income societies is the independent role of norms and the permeability of values and ideals. Yet he would have realized that the convergence claim was more of a product of the economic, societal, and scholarly context of the time. Second, in light of the dramatic gender revolution that unsettled the seeming inevitability of the conjugal family, he would be more cognizant of the predominant role women played in the process of family change. Although he acknowledged the “disadvantaged” role of wives within the household, at the time Goode assured that few women would move into prestigious occupations because they would still be required to care for children and engage in housework (Cherlin, 2012) – as also emerging from his edited book The Other Half: Roads to Women’s Equality (Epstein & Goode, 1971), which in multiple passages hints at the authors’ acceptance of ordinary assumptions about sexuality, sex differences, and social arrangements (Gilbert, 1972). Although women still do the majority of housework all over the world (Ferrant, Pesando, & Nowacka, 2014), we know that nowadays in most advanced societies women make up the majority of physicians. Third, in light of economic and socio-cultural dynamics that are driving family patterns apart between classes through increasing educational homogamy, income inequality, and diverging paths of prosperity and insecurity within societies, Goode would revisit his study focus and devote more space to within-country and within-region family dynamics. Fourth, in light of the massive “deindustrialization” that has occurred in the West, Goode would perhaps trace a tighter connection between family change and globalization, rather than industrialization, in a spirit closer to Therborn (2004). In sum, Goode nowadays would keep on stressing that some family forms, behaviors, and ideals spread across contexts in tandem with socio-economic and structural transformations, yet he would not take the 1950s Western family as the sole endpoint of global social change.

To conclude, I would like to stress that despite all the criticism Goode’s influence persists nowadays and keeps influencing how scholars think about family systems change by providing a benchmark against which better theory is produced – as demonstrated throughout this essay. Perhaps Goode’s biggest mistake was to not realize that he was writing at a time of relative “uniformity” that followed a period of complexity and anticipated another period of even greater (yet different) complexity (Cherlin, 2012). Nevertheless, the effort he undertook in making the field of family sociology both global and comparative was massive and remarkable given the limited tools of the time. This said, I agree with Cherlin (2012) that ‘sometimes a flawed yet ambitious intellectual exercise can be more enlightening than a successful but simple one’ as ‘it can provide a rich historical record of social change’ and ‘challenge critics to come up with something better’ (p. 603). The aim of this essay was to stress once again the theoretical relevance of such ambitious exercise, and intends to stimulate scholars to keep producing theories and test competing explanations of why and how family systems change on a global scale.