Rights to Her Labor: Women Workers on Mexico’s Southeastern Railroads

Abstract

Across the twentieth century, hundreds of women worked as nurses, cooks, cleaners, and teachers on Mexico’s railroads. They have been overlooked in histories of the railroads and Mexican industrialization more broadly, their limited number perhaps suggesting that their work is not of analytical importance in understanding processes of economic development and class formation. On the contrary, these women’s work constituted many of the most coveted labor rights of the postrevolutionary railroad workforce, itself a symbolic vanguard of Mexico’s working class and one of the most important beneficiaries of the expansion of social and economic rights ushered in by the Mexican Revolution. The gendered division of labor characteristic of the railroads was neither accidental nor insignificant. Railroads used the feminization of the work of social reproduction to write off structural failures and predictable shortcomings in welfare provision as failures of femininity. Women became scapegoats for the consistent violation of workers’ rights through underfunding and understaffing. In tracing this process, the article models a historiographical and methodological intervention with broader relevance. It suggests that the social and labor rights that expanded around the world in the late-nineteenth and early-twentieth centuries must be studied not only from the vantage of legal or political history, but as themselves questions of social and labor history. Making these rights real depended on socially reproductive work that has often been marginal in accounts of industrialization and economic development. It is impossible to understand the political economy of social and economic rights without understanding women’s work.

In 1959, railroad worker Jesús Covarrubias was away from his worksite on the Ferrocarril del Sureste (FS), in southeastern Mexico, when he suffered an accident. His union challenged the railroad’s finding that this was not a work accident and thus did not merit compensation. Covarrubias had to leave his workplace, the union reasoned, because it was “uninhabited” and the company hadn’t provided “a dining service for its workers, nor an easy and risk-free way of obtaining food.” The railroad wouldn’t budge. Because Covarrubias was a state employee with no access to labor arbitration boards, that was the end of the matter.¹

As Mexico entered its mid-twentieth-century economic “miracle,” more of its workforce became wage-dependent. This partial proletarianization has been understood as, quintessentially, a process of dispossession, as population growth and the subordination of peasant producers to the needs of industry eroded market-independent subsistence.² But in postrevolutionary Mexico, for a privileged sliver of the working class, to be proletarian was to be possessed of new economic and social rights: limited working hours, compensation for workplace accidents, medical care, and education. Workers became derechohabientes—rights-holders—entitled to employer-provided benefits, often in the absence or limitation of public options.³ The 1917 Constitution and 1931 Federal Labor Law required that large employers provide Article 123 schools whenever work communities were located more than three kilometers from existing schools, as well as healthcare and compensation in cases of occupational injury.⁴ Strong unions negotiated further benefits, including medical care for dependents, while less powerful unions might be lucky to secure just employer reimbursement for workers’ medical costs, and much of the vast population of agricultural and informal workers made do with family attentions, scanty or nonexistent public services, or private care paid out of meager earnings.⁵ For rail and oil unions, industry welfare provision was preferable to Mexico’s slowly expanding social security institute, the Instituto Mexicano del Seguro Social (IMSS). Railroad workers weren’t fully incorporated into IMSS until 1982, and the state oil company, Pemex, operates its own medical service to this day. Across Mexico’s decades of state-led industrialization, its “welfare state” was more accurately a patchwork of private, public, and employer-provided welfare services, in which hundreds of thousands of industrial workers and dependents received medical care from employers, and tens of thousands of children were educated in employer-financed schools.⁶

The unequal alliance forged between the Mexican state and organized labor was crucial to maintaining the country’s (in)famous postrevolutionary, authoritarian stability.⁷ Part of that alliance depended on access to these social and economic rights, which were notionally universal but in practice tethered more to certain kinds of employment than to citizenship.⁸ If these unequally distributed rights were in part what made possible the stability of the postrevolutionary state and the nation’s rapid economic growth, what made possible this particular constellation of rights?

In the case of Mexico’s railroads, it was women’s labor. When Covarrubias made his complaint, he and the union doubtless imagined that the “dining service” in question would be provided by women cooks, like the many others serving railroad workers in the region. Indeed, across geographies, women’s un- or under-paid domestic and reproductive labor has transformed wages into subsistence, comfort, and care, reproducing the labor power subsequently deployed in productive work on both quotidian and generational timeframes.⁹ The terms on which such labor is provided, and its relationship to production, are central questions in feminist scholarship.¹⁰ Mexico’s heterogeneous welfare system presents one opportunity to examine the historical specificity and contingency of this articulation, as industrial concerns hired small armies of women cooks, cleaners, nurses, teachers, and laundresses to comply with contractual obligations to care for and reproduce their workforces. Workers’ rights to healthcare, education, and clean workplaces were guaranteed through the partial commodification and de-domestication of what had traditionally been “women’s work,” even as the Mexican state exalted an (often unattainable) male breadwinner ideal and cast suspicion on women working outside the home.¹¹ It is impossible to understand the political economy of Mexico’s celebrated social and economic rights without understanding women’s work.

Such understanding remains limited for the post-1940 period despite pathbreaking scholarship on earlier decades.¹² An eclectic array of sources has shed light on the experiences of women working in and out of the home; on the contradictions they faced as women, workers, wives, and mothers; and on the “uneven and incomplete” commodification of reproductive work in the late-nineteenth and early-twentieth centuries.¹³ This scholarship has also illuminated how export economies relied upon and reconfigured reproductive labor and gender identities.¹⁴ Mexico’s railroads offer an intriguing variation on this problematic: the creation of constitutional and contractual rights to care that were satisfied by the partial incorporation of an existing gendered division of labor into industrial concerns and labor unions. If early labor histories often focused on industrial workplaces, and more recent work has considered the domestic and reproductive labor performed in homes, streets, markets, and state institutions, the emergence of Mexico’s fragmented, employer-dependent welfare apparatus suggests an opportunity to bridge these literatures. What happened when productive and reproductive labors were legally and economically sutured together on balance sheets and payrolls? John Womack’s question—how has “industrial work in Latin America … taken gender’s conjugation, or changed its declension”?—takes on new life in the context of what Ann Farnsworth-Alvear calls “Latin American experiments with ‘welfare capitalism.’”¹⁵ Answering it requires decentering the normative male railroad worker to draw out the women who have remained in his shadow, thereby obtaining a fuller understanding of working-class masculinity, the heterogeneous ways women’s labor subsidized development, and the contradictory ramifications of industrialization on this thing called “women’s work.”¹⁶

One challenge is excavating work practices from a discourse of industrial modernity, emergent in the 1940s, that foreclosed the political openings of Mexico’s 1910 revolution and relegated women to the domestic sphere, even as their labor remained essential to family and national economies alike.¹⁷ A uniquely well-preserved set of railroad personnel files for three railroads in southeastern Mexico permits examination of women’s everyday working conditions as they made real the rights of their (male) coworkers, and as “women’s work” was transformed—and preserved—by its inclusion within railroad companies.¹⁸

A few words about these companies. Railroads were foundational to Mexico’s nineteenth-century inauguration into capitalist economic growth.¹⁹ Initially financed by foreign capital, the railroads were “Mexicanized” in 1908 as Ferrocarriles Nacionales de México (FNM), with the government as majority shareholder. In the 1930s, Lázaro Cárdenas’s administration expropriated FNM and proposed a dramatic expansion of the rail system, financed by public–private partnerships.²⁰ But private capital was not forthcoming, and from 1930 to 1961, the government built only two lines.²¹ One of these was FS, which connected the still-private Ferrocarriles Unidos de Yucatán (FUY) to the rest of the system. In 1968, FS and FUY merged as the parastatal Ferrocarriles Unidos del Sureste (FUS), and workers moved from the Ministry of Public Works (Secretaría de Comunicaciones y Obras Públicas, SCOP, later the Secretaría de Comunicaciones y Transportes, SCT) union to the national railworkers’ union, the Sindicato de Trabajadores Ferrocarrileros de la República Mexicana (STFRM).²²

Capital was not the only recalcitrant factor hampering government ambitions. Mexico’s geography, long lamented as a hindrance to national integration, posed formidable technical problems to engineers and fatal problems to laborers.²³ Mexico’s staggeringly high rates of industrial accidents were exacerbated by the southeast’s swampy environment, which was propitious for flooding and disease.²⁴ Railroads were ill-equipped to provide food and services to remote workers, relying instead on workers’ wives—meaning hundreds of women and children were exposed to the unenviable conditions of spartan railroad camps.

Railroads were compelled, however, to shoulder some burdens of social reproduction. FUY completed construction of its hospital in 1919.²⁵ FS opened its Coatzacoalcos hospital in 1936.²⁶ Aid stations, schools, and the Campeche hospital followed. Railroads’ abundant archival production offers an excellent vantage from which to examine the inner workings of these “fringe benefits,” which were in fact complex services constituted largely by women’s labor.²⁷ Two-thirds of teachers employed by these railroads were women. Sixty-six of sixty-eight nurses hired after 1940 were women. All camp and hospital cooks, cleaners, and laundresses were women. When workers claimed welfare rights, they were claiming rights, most often, to a woman’s labor.

As Covarrubias’s story suggests, the changing valorization of women’s work within the rubric of industrial employment was fraught. Women workers were relegated to the bottom of the wage hierarchy, with far fewer opportunities for advancement than their male counterparts. But to be paid a wage at all, much less to have union representation and to be included, albeit ambivalently, within one of the country’s most important industrial workforces, was a far cry from the unpaid, or miserably paid and unprotected, domestic labor that remained the lot of most women in railroad communities. Succinctly put, the changing relationship of the work of social reproduction to the labor market opened conflicts over the content of “women’s work” that were shaped by a bevy of competing forces: its status as a right or benefit of (men) workers; its proximity to (often unpaid) domestic labor; and women workers’ inclusion within labor regulations and legal protections from which domestic work—paid or not—was constitutively excluded.²⁸

The effect of these pressures was contradictory, both exploiting and transforming the gendered division of labor underpinning development.²⁹ Railroad employment provided hundreds of women with steady wages and access to healthcare, housing, and, sometimes, pensions. It allowed them to attenuate, or even extricate themselves from, dependence on a male breadwinner. But this was no straightforward triumph of economic inclusion. Women gained labor rights, but other workers gained rights to their labor. Both companies and fellow workers sought to entrench a flexible and self-sacrificing femininity as a form of labor discipline and mechanism of extraction. Women received lower pay than similarly skilled male workers. The nature of their work was indeterminate and not amenable to reductions in time and task flexibility, meaning they struggled to claim their own rights.³⁰ The proximity of women’s commodified reproductive labor to the unpaid work of railroad wives and daughters made them vulnerable to informalization and unemployment should companies succeed in shedding some of the costs of the workforce’s social reproduction. Indeed, as companies sought to shuffle these costs back onto workers, their families, and the state at every turn, women workers faced intensifying precarity, culminating, in 1975, with the unwinding of decades of coincidence between “women’s work” and “railroad work” through the forced retirement, layoff, or casualization of teachers, cooks, cleaners, and nurses. What follows, then, is a story of railroad workers’ labor rights, which is also, necessarily, a story of women workers.

Teaching

Schools were crucibles for many of the tensions that characterized women’s railroad work. Article 123 schools were, in the first instance, constitutional entitlements underwritten and administered by children’s parents’ employers—a volatile combination.³¹ They were also workplaces in which teachers had the same employer, labor rights, and contractual entitlements as the rest of the railroad workforce. But was teaching really “railroad work”? Should teachers be entitled to the same benefits as railroad workers? As workers sought to secure quality education for their children, and companies sought to minimize spending on social services and avoid interference by state inspectors, they came down on different sides of these questions at different moments. Three sets of conflicts—over teacher allocation, teacher performance, and the nature of teaching labor—illuminate how companies and workers tried to resolve these tensions through the feminization of teaching and, eventually, its wholesale ejection from the category of “railroad work.”³²

The first set of conflicts began before teachers set foot in the classroom. Railroads waited for work crews to demand teachers and sought to provide the minimum required by law. Workers in the La Unión, Chiapas camp demanded a teacher given their distance from existing schools. The daughter of a railroad worker killed on the job volunteered as long as she was provided a house.³³ When fifty-four children in Tenosique were “suffering” the lack of a teacher, Libertad Mora was dispatched—an arrangement agreeable to her, as she had for months been petitioning to move to Tenosique to care for her ill mother.³⁴ In fall 1957, FS fired teacher Margarita Garduza for “budget reasons.”³⁵ The union protested, and an FS inspector agreed, observing that the sixty-four students in attendance were “excessive” for one teacher. Garduza was rehired.³⁶

It was thus possible for teacher and family interests to align. More commonly, however, company cost-cutting pitted teachers and families against each other. Though teachers, like work crews, were at the mercy of fluctuating labor demand along the railroad’s perilous geography, they were denied the priority in camp housing afforded to crews. The only semi-urban railroad school was in Allende, where competition for placements was fierce. The rest were located in towns and camps scattered between Coatzacoalcos and Campeche. Teachers found themselves living in school buildings, paying high rents for private housing, or occupying the most run-down and unhygienic company houses.³⁷

Teachers complained of the unhealthful conditions, inadequate housing, and inaccessibility of the camps where they were sent. Especially for teachers who were finishing secondary school or teacher training, unpredictable relocations to remote camps felt unjust and counterproductive.³⁸ That FS paid a premium for the inconvenience of its schools was satisfactory to many, but for some, constant separation from family took a toll: “It is indispensable for me [to be stationed] nearer to my family to save my home without risking my job,” Zoila Uco wrote.³⁹ Uco supported her children, mother, and former railroad worker husband, but the size of her family and her mother’s ill health meant they could not live in a camp. “If it is impossible for me to be a teacher in Tenosique [with her family], commission me as a cleaner in the office or Hospital, as I am one of those who badly need this wage for my children’s bread,” she pleaded.⁴⁰

If families and companies were the prime movers in struggles over teacher allocation, once a teacher had been installed, companies took a backseat. Railroads had little interest in monitoring teacher quality, hiring teachers based on convenience rather than credentials.⁴¹ Families, on the other hand, considered education a right—and guarantee of skilled, high-paying union jobs. Education was a source of pride as well as social mobility, and it was a widely legible language of claims-making: “as you know, Mr. Manager, we cannot leave our children a more useful patrimony than preparation through study.”⁴²

In the close-knit context of railroad camps, teachers and families lived cheek-by-jowl and workplace hierarchies permeated community life. Under these circumstances, the association between motherhood, nurturance, and children that had initially opened the teaching profession to women was a potent way to question their fitness as teachers.⁴³ Enmeshed in small company communities, dependent on the goodwill of local union representatives and crew leaders, and often married to fellow workers, women teachers were exposed to constant policing of their after-hours behavior. Workers legitimated assessments of teacher performance through gendered surveillance of teachers’ personal affairs. In a complaint against Adela Moctezuma, the head of the parents’ association and crew captain, Anastasio de los Ríos, wrote that the parents were “very unhappy” with her work. While he referenced her treatment of the children briefly, he focused on her frequent visits to Tenosique, “without permission from anyone” to “have intimate disputes with her husband,” a railroad mechanic. When de los Ríos and three laborers were questioned further, the complaints shifted: de los Ríos said his daughter had passed her exams but “it turns out the girl knows nothing and I am upset because the girl knows nothing.” The others had similar concerns: One said Moctezuma had failed his son without letting him sit the exam, “which I don’t think is right.” Another said Moctezuma passed his son without examination, “and he doesn’t even know how to write his name.”⁴⁴ The teacher’s deficits as a wife and woman were inseparable from her deficits as a mentor. The more teachers exposed themselves to community scrutiny, the more vulnerable they were to accusations of impropriety, however ill-founded.

Zoila Uco’s case is again instructive. In 1959, she had recently married laborer Apolinario Hernández. Together with the stationmaster (also head of the parents’ association), Uco organized a Mother’s Day celebration. One worker later called it a “fiestecita,” and according to him, “Everything started out fine.” But when the stationmaster got drunk, things took a turn. He asked Uco to dance with him, but she, “being a married woman, with her husband present,” refused. This angered the stationmaster, who reported her to municipal authorities and possibly threatened her with a pistol and knife. Only then did an FS investigator intervene.⁴⁵ Ultimately, the railroad did not discipline her, but neither did it do anything to address the undercurrent of sexual violence she faced. Indeed, while Uco’s experience of male sexual entitlement was enough to call her conduct into question, a public incident in which a (married, father to at least nine) male teacher lost $3,500 during a night with a sex worker saw him resume his duties without comment.⁴⁶

Male teachers were not wholly exempt from gendered assessments of their performance. The notion that a male teacher should be “a righteous man with a vigorous and pure character” to “forge men who are also capable of serving [their fatherland] with bravery and loyalty” was widespread.⁴⁷ But men, more than women, could leverage their (self-professed) conformity with masculine ideals as sources of exoneration, even honor. They were also far less subject to scrutiny, particularly when off the clock.

A final set of conflicts made the underlying question in all of these cases—was teaching really railroad work?—explicit. If it was not, what did that mean for teachers’ relationship with railroad labor regulations? If it was, how to square the particular demands of teaching with work rules designed for a railroad—especially robust protections in cases of illness and injury, and the presumptive interchangeability of workers of the same category?

Because teachers’ work was not a liability issue for the company, there was no system for providing substitutes, as there was for nurses. Teacher absences in schools with only one or two teachers could cause chaos. Though joint state-railroad schools were better staffed, state-funded teachers would not cover railroad teachers’ classes.⁴⁸ Adding fuel to this fire was the fact that teachers tended to have larger families and heftier domestic responsibilities than nurses. Families, children, and staff thus experienced teachers’ rights to sick, maternity, and unpaid leaves as burdens. One solution was to separate teaching work from railroad work to justify the curtailment of teachers’ labor rights.⁴⁹ The director of a joint state-railroad school complained to FS about María del Carmen Davila’s absences, which were permitted under railroad policy: “[T]he work of a teacher is not like that of a railroad worker, because our work material is the children, the children of your own workmates who patriotically fulfill their sacred duty by sending them to school.”⁵⁰ While a railroad worker might be justified in taking advantage of his contractual benefits, Davila was not. She may have had labor rights, but her labor was also a right to which other workers, and children, were entitled.

Jointly funded schools witnessed a second instantiation of this conflict. FS insisted that its teachers were not obligated to take orders from anyone but company supervisors, and teachers were to follow railroad labor regulations—which covered, among other things, generous sick leave. This was consonant with federal law, which stipulated that Article 123 teachers were employees of the companies that paid them. But different disciplinary regimes and entitlements within one workplace were a headache that incentivized states to assume more educational responsibilities. In the long term, the railroad succeeded in extricating itself from schooling. But in the short term, it faced the contradiction of insisting railroad teachers were subject to railroad policy while attempting to deny them the same labor rights as other workers. In 1965, Adela Moctezuma and another teacher requested to switch places. Labor regulations allowed employees of the same job category to do so, on the assumption it wouldn’t matter if two manual laborers wanted to swap brigades. However, neither the railroad nor parents wanted to change teachers in the middle of the school year. The manager, against all precedent, stated that teachers did not fall under the labor regulations as their work was not “exclusively [related to] the railroad.”⁵¹ He denied their request.

Over the course of the 1960s, culminating in 1971, FS/FUS divested itself of its educational role. On the whole, this was doubtless a positive development: Children’s right to education would no longer hinge so directly on their parents’ employment. But any reconfiguration of the social organization of care brought conflict, especially for those who felt their sacrifices were overlooked, or worse, had disadvantaged them. Public schools required, as a matter of course, credentials that railroad schools did not. Because better-trained teachers were given more urban placements, teachers without credentials could get locked into working in camps distant from teacher-training programs for years, at the cost of a chance at a public-school job.⁵²

It is impossible to follow teachers’ post-railroad careers systematically. Some landed jobs in public schools. Others, even with credentials, did not. This could be a bitter end to decades of work. One of the longest-serving teachers, Libertad Mora, started her railroad career as a laundress and cook before becoming a teacher in 1950. Twenty years later, she was within five years of retirement and wanted a pension instead of severance: “It is unjust that the company where I left my youth not recognize this and remove me from my position without a second thought.” Thirty-seven and divorced, she worried that no one else would hire her, leaving her children with no one to support them.⁵³ In the denial of Mora’s pension was the final answer to the question that had dogged teachers and railroad communities since the 1940s: Teaching was no longer railroad work, even if it once was. Or at least, the railroad wasn’t going to pay for it.

Nursing

Through the 1940s and 1950s, railroad nursing was not the professionalized occupation that it would become by the late 1960s. Like teaching, it was strongly associated with domestic and women’s work.⁵⁴ Railroad nurses had eclectic backgrounds: some had completed secondary school and nursing training while others had not finished primary school, differences reflected in the large spread of nursing wages. Nurses’ responsibilities were likewise heterogeneous: stripping and making beds; cleaning surgical instruments; administering injections; monitoring vitals; updating medical files; providing medications; verifying patient documentation. Some nurses, through skill, luck, or both, garnered additional responsibilities and a bit of extra income. In other cases, companies took advantage of, but did not compensate, specialized skills.⁵⁵

Unlike teaching, nursing’s relationship to the railroad was never questioned. Nurses always belonged to the union (teachers lost representation when the STFRM took over from the SCT union in 1968). Railroad medicine was an area of international concern, and albeit for different reasons, unions, doctors, and companies acknowledged nurses’ importance.⁵⁶ None of this meant, however, that their employment was free from tensions similar to those affecting teachers, though conflicts over their work manifested differently. If for teachers, the problem of defining their work hinged on whether it was “railroad work” in the first place, for nurses, disputes often turned on the assertion of boundaries between what a nurse could and could not be required to do, and why.

This problem was exacerbated by the fact that railroads tried to shave down medical expenses by minimizing staffing costs. Across decades, doctors and union representatives requested more nurses in response to acutely felt need.⁵⁷ In many cases, the company was compelled to acknowledge that staffing levels were inadequate and to hire more nurses. But because it hadn’t created enough spots on payroll, the railroad resorted to hiring these nurses under whatever categories had vacancies: cleaner, cost checker, and assistant sleeping car supplier.⁵⁸ The apparent technicality of mismatched duties and titles could have serious ramifications even when there was no wage differential between titular job and work performed. As Hermenegilda Vázquez wrote, not only was she listed as doing “work I have never done,” but “I find myself in a position of inferiority relative to the other women with my specialty … I cannot appear on the escalafón [list of employees by seniority], nor gain rights to a promotion.”⁵⁹

Another group of women, hired in the 1930s or 1940s as cleaners or cooks, became nurses later in their careers. While apparently similar to the preceding pattern, their trajectories throw into relief a different logic. Daría Guzmán was born in the first half of the 1910s and completed two or three years of primary school.⁶⁰ FS hired her as a laundress in 1942.⁶¹ A few months later, she was promoted to cook.⁶² In 1954, FS planned to lay her off, but the union intervened and suggested she take the place of a recently deceased cleaner.⁶³ The company agreed. With a pay cut from $8.30 to $7.70, Guzmán was demoted but stayed in work. Four years later, she was promoted to assistant nurse, a position she held until retiring in 1969.⁶⁴ The 1958 promotion stated that it was “in accordance with the specific labors she has been performing.”⁶⁵

Evident again is the need for nursing labor, but also the fungibility of these feminized job categories: laundress, cook, cleaner, nurse. While supervisors were aware of the different kinds of work that adhered to each job category, in practice these distinctions became slippery. So it was that another cleaner-cum-nurse, Agripina Anzures vda. de Rodríguez, faced a complaint from a new supervisor that she was falling behind. In the subsequent investigation, the union representative, unusually, supported Anzures, saying it was “impossible” for her to do the work the supervisor demanded within the allotted time: while she “had the category of Assistant Nurse, at the same time she performs the work of an orderly and also that of a laundress … and has performed this work for many years … without the head of the Aid Station needing to report her.” For her part, Anzures called the accusations “childish:” “I have fifteen years of service and know my obligations to the letter.”⁶⁶

As Anzures’s insistence shows, the amorphousness of feminized job categories was related to the problem of asserting boundaries on a day-to-day basis at work. Even when nurses had titles as such, the question of what their job entailed was subject to dispute. Nurses, unlike most skilled railroad workers, did not have a contractually determined job manual.⁶⁷ These manuals were valuable because they reduced time and task flexibility by setting limits on what companies could require of workers. Simply put, they defined the work. Companies, too, had an interest in establishing detailed work responsibilities in order to minimize the risk of catastrophic accidents.⁶⁸ Importantly, not all medical employees lacked manuals. In the 1967 contract between STFRM and FUY, doctors had specific obligations.⁶⁹ No such provision existed for nurses. Contract language regarding hospital and aid station staffing was vague. The company committed only to “having in the Hospitals and Aid Stations the staff that may be necessary 24 hours of the day … and [to establishing] shift schedules as service requires, so that patients do not lack proper medical care.”⁷⁰

Some theoretical coordinates may be helpful. Social reproduction is profoundly indeterminate and thus demands a capacious, flexible workforce; yet, instead of compensating this flexibility with higher wages, employers often rely on gender-, race-, or class-segregated labor markets to obtain it on the cheap.⁷¹ Because nurses’ labor was an entitlement of other workers, it was not necessarily in union or company interests to specify nurses’ obligations in writing, lest they use such documents to resist expansionary pressures on their labor. Moreover, while it may be challenging to specify in advance what a conductor must do to complete a run (hence the efficacy of working-to-rule), it is easier than the impossible task of specifying, a priori, exactly what a nurse must do to keep a sick child alive.

This problem was anything but theoretical. On Sunday, October 1, 1959, around 7:20 p.m., Estanislao Olán and his wife brought their sick toddler, Teodoro, to the Coatzacoalcos hospital.⁷² By 7:55, Teodoro had died. Olán filed a heartbroken, irate complaint against the nurse on duty, Teodora Zamora. When he and his son arrived at the hospital, he wrote:

[Zamora] stated that there was no doctor on call and that I could find him in his house, which was unsuccessful, and while I looked for him in many places and did not find him, my son having passed away hours later without receiving proper care … I would also let you know, that the nurse in question, when I arrived at the hospital and asked her, as there was no doctor on call, to perform first aid, she arrogantly and in violation of elementary principles of Ethics, answered me that she did not have time and that hospitalized patients came first.⁷³

The subsequent investigation, which involved interviews with Olán, Zamora, and the night watchman, as well as consultations with doctors, found agreement on the order of events, but wide divergences in interpretation.

First, what happened: The couple arrived with Teodoro at the thirty-two-bed hospital sometime after 7:00 p.m. Zamora, the only medical staff present, asked them to wait while she finished with another patient. Olán refused, and Zamora sent him and an orderly to find the doctor on call, who had neglected his contractual duty to indicate his whereabouts. Meanwhile, the nurse, though not authorized to decide treatment on her own, began first aid. When she asked the mother why they had waited to bring Teodoro in, she replied that they wanted to, but the heavy wind made the ferry from the workers’ neighborhood in Allende to Coatzacoalcos too dangerous. (No bridge crossed the river until 1962.)⁷⁴ Despite Zamora’s efforts, which several doctors concurred were adequate, the child died.⁷⁵

Next, how this sequence was interpreted. The union representative deemed Zamora’s behavior “quite awful” and suggested she be reprimanded. At the same time, he said, it was imperative for a doctor to be present at night, as “it is understandable that a nurse does not have sufficient ability to care for a patient like” Teodoro.⁷⁶ The company investigator found that Zamora “did not receive [the family] with proper respect,” but that “she did provide appropriate medical care.” He explained Olán’s falsehoods by observing, “he was in a state of absolute desperation … and precisely the poor treatment received from Nurse Teodora Zamora Ramírez led to [his] complaint, of course with some alterations to give it more force.”⁷⁷ Ultimately, the company manager wrote to the hospital director: “Due to the psychological state of the families of patients cared for in the Hospital, and to avoid shameful incidents like [this] … I will ask you to order the staff … to show solicitousness in their services to avoid similar complaints from both Union Organizations and workers.”⁷⁸ He made no mention of having a doctor present at night.

The heart of the dispute was not whether the pediatrician had neglected his obligations, nor whether a doctor should be present all night. It was not whether more than one nurse should staff the hospital so that emergencies did not inevitably present as decisions between caring for one patient over another. It was not even, really, about whether the nurse had done all she could to save the child; though Olán suggested otherwise, union representative and company inspector agreed that she provided proper care. The kernel of the dispute was whether she did so politely enough. What could, from the circumstances of the case, be understood as a matter of doctors’ responsibilities, understaffing, and the inaccessibility of the hospital for workers who lived across the river, was instead refracted through the family–nurse interaction as a question of courtesy and respect.

Not all cases presented stakes as high or outcomes as tragic. Nonetheless, conflicts between patients and nurses followed discernible patterns, in which complaints made against nurses for lack of politeness reflected the gap between what derechohabientes considered, often with reason, to be their entitlements, and what nurses, often with reason, were willing or able to provide under the circumstances. The gendered division of labor turned the railroad’s under-provision of working-class welfare rights into interpersonal disputes underwritten by expectations of feminine courtesy, self-denial, and endless elasticity.

A striking example of such elasticity was a discovery made by the union in 1957. A doctor recommended two young women as nurses, citing their more than a year working as unpaid auxiliaries in the Coatzacoalcos hospital.⁷⁹ One of the most diligently enforced provisions of the labor regulations was a ban on unpaid trainees, and the union complained immediately. The doctor replied that they were not trainees but “‘Voluntary Auxiliaries,’ of which there are groups organized in almost all Hospitals in the Republic … [their position does not create] seniority rights nor obligations on the part of the Railroad to remunerate them economically.”⁸⁰ The hospital sought to exploit the historical links between women and charity in a new context: These were not wealthy women seeking social recognition for their service, but working-class girls laboring unpaid in hopes of landing permanent jobs.⁸¹

A change in staffing policy with the 1968 merger formalized this practice. The medical service began relying on zero-hours nurses called extras. These women were paid employees and union members. However, extras were not guaranteed any hours but were called in on an as-needed basis. That FUS made no pretense of guaranteeing income or stability led some extras to be frank about the terms on which they were willing to work. In 1971, a physician’s assistant complained to the head of the medical service about extra Santa Teodora Morales. She had been about to leave after her shift when a “child derechohabiente” arrived for emergency care. The assistant asked Morales to give the child an injection, but she replied that her shift had ended. He continued, “at that moment I was busy attending to another consult and there was another person waiting for an injection. The nurse whose shift it was, was changing the sheets and distributing patients’ clothing.”⁸² Though Morales was off the clock, he implied that derechohabientes’ right to medical care superseded her labor rights. Abundant need for nursing labor meant many extras worked substantial hours, but the system transformed fluctuations in that need into risks borne almost entirely by extras. The flexibilization of the nursing workforce formalized expectations that women workers assume the burden of ensuring the consistency of company medical provision without reciprocal assurance of consistent income or work.

Cooking, cleaning, laundering

Cooks, cleaners, and laundresses, while distinct jobs, were the most fungible occupations: Women moved between them over the course of their careers and substituted for each other, something not permitted for nurses or teachers. Cooking, cleaning, and laundering were also the clearest extensions of domestic work; at least one cleaner worked in the railroad manager’s home.⁸³ Domestic work, paid or unpaid, was overwhelmingly performed by women, making these job categories the most obviously and deeply gendered.⁸⁴ The women hired to perform these jobs shared a demographic profile slightly different from that of nurses or teachers. They were older, had more children, and were often widows. Several were illiterate, though most could read. While some eventually became nurses, others continued in these positions for their entire careers.

The proximity of cooking and cleaning to unpaid domestic labor affected the women performing this work from hiring onward. Like nurses, they faced the problematic of elasticity; like teachers, they faced management’s suspicion that their work was not “railroad work.” Unique to these women, however, was the divide between formality and informality—that is, between being official, salaried, unionized employees of the railroad, and working por su cuenta, as independent contractors without employment. Of all the workers discussed here, cooks and cleaners were the ones most ambivalently and precariously included in the formal sector.

Camp cooks were often only hired after they had established a commercial relationship with a nearby work crew, serving laborers out of their own kitchens. Why FS decided to hire women cooks, rather than allow the informal kitchen economy to endure outside its purview, is to some extent a matter of speculation. (FUY, for instance, did not hire women directly but rather subcontracted a woman cook to feed patients in the hospital—but FUY did not construct new railroads during this period and so did not have to support remote crews.)⁸⁵ Documentary sources provide some hints. Work crews expressed partiality toward certain women and requested they be hired.⁸⁶ The company could require employed cooks to follow crews and redistribute them to meet changing need. It could also require them to work certain hours and exert greater control over vacations and sick leave.⁸⁷ Cooks’ wages were among the lowest of all employees, so hiring them was not a major financial undertaking.⁸⁸

When FS hired cooks, it altered, but did not fundamentally remake, existing economic relationships. Camp cooks became waged and unionized employees. However, the company didn’t pay for food; workers paid cooks for meals, and cooks bought supplies from those earnings. Their wages compensated them for their knowledge of local markets and management of what were essentially extended household budgets. Credit lubricated the entire operation: On several occasions, cooks had to appeal to companies to get workers to pay their debts because the cooks were indebted to local vendors.⁸⁹ The camp cook system thus partially internalized the quasi-domestic economies first created by independent cooks. In so doing, it preserved for cooks a degree of autonomy, but also caused tensions between the normative eight-hour workday and the demands of cooking labor. Asked what time she did the daily shopping, Anacleta Ramos replied 5:30 a.m., but “sometimes at 6, others at 7.”⁹⁰ Eugenia de la Cruz was retroactively paid four hours of overtime per day for an entire year, suggesting she was routinely working twelve-hour days.⁹¹ Manuela Gómez received eight days of paid leave because FS called her in to work four days from 6:00 a.m. to 9:00 p.m.⁹²

Food provision was intimate, and workers became attached to the women serving them. If workers believed the company was treating their cook unfairly, they might advocate on her behalf. In 1941, thirty workers wrote to the chief construction engineer lauding cook Ángela Requena, who had been working without a wage for weeks, asking that she be paid.⁹³ Similarly, in 1961, workers in Tancochapa expressed their appreciation for “our cook” Alberta Ortega, rejecting any suggestion that her prices were too high.⁹⁴ However, intimacy could breed contempt as well as fondness, and workers weren’t shy with complaints.⁹⁵ Underlying even favorable assessments of a cook’s work was a sense of entitlement to the woman and her labor: “our cook.”

The quasi-separate-sphere of the camp kitchen reflected the company’s expectation that when it hired a cook, part of what it was purchasing was her budgeting ability.⁹⁶ But a quasi-separate-sphere was just that, and the company did not hesitate to intervene when evidence emerged that a cook was not stretching her budget to the utmost. FS instituted a modified version of the camp kitchen system in its antituberculosis ward. Hospitalized workers received meals for free, so FS paid Anacleta Ramos $7 per patient, per day to procure foodstuffs. Five men interned in the unit complained about her meals.⁹⁷ The company and union stepped in and asked Ramos’s assistant, Teodora Hernández, if she considered the food to be worth $7. Hernández answered in the negative.⁹⁸ Five days later, a company investigator interviewed Ramos, attempting to catch her in a lie about the cost of the food she purchased. At this, Ramos became indignant: “After many days, it is not possible for me to remember exactly what I give each patient to eat, because for that I would need to write down everything that happened every day.” She challenged the union and management to do the shopping for a day “so they can see if $7 is enough.” The investigator determined—without taking Ramos up on her suggestion—that the food she was buying should cost only $20 per day and that the company had cause to fire her.⁹⁹ Instead, FS relocated her to a distant camp.¹⁰⁰

Precarity was another result of the proximity of cooking, cleaning, and laundering to domestic labor. While workers had rights to schools and medical care, they lacked legal entitlements to food or laundry service. These women’s inclusion on company payrolls was thus far more ambivalent than nurses’ or teachers’. Many women in these roles worked short-term positions for years before landing permanent jobs. Others never even made it onto payroll: Ángela Requena paid an unnamed assistant out of her own wages—meaning the assistant was not formally employed and had no entitlements to benefits, retirement, or severance.¹⁰¹

Other features of cooking and cleaning work reinforced this precarity. Supervisors assumed that the work was unskilled and could be performed by any woman, so female relatives were permitted to informally substitute for ill cooks/cleaners, or other women workers of different job categories were required to step in. When the laundress Lucía Ríos fell ill, the company did not find a substitute but rather expected the cleaners to do laundry in addition to their other work: “[A]s you know,” a worker in Teapa wrote, “the Cleaners have complained to the Union that they are obligated to do the work of laundresses, Cleaners, and Nurses at once … the laundresses … are overloaded with work.”¹⁰² Repeated demands for more cooks, cleaners, or laundresses were rarely satisfied. FS instead enforced and exploited the fungibility and elasticity of women’s work to patch holes in its provision.

Relatedly, but specific to the case of cooks, was an expectation that married men did not need company catering. Their wives would provide. When a crew in Tancochapa requested that their cook stay on, management inquired of the men, “Are they single?”¹⁰³ A provision in the 1967 contract between FUY (which did not keep cooks on payroll) and STFRM gave employees stationed in “places where it is hard for workers to obtain supplies” the right to an annual rail pass to the closest market so that they could “send a person who would do their shopping for them.”¹⁰⁴ Similarly, the inclusion of workers’ wives, but not husbands, as benefits-eligible dependents indicated an assumption that women would not have access to benefits on their own account but would instead work in the home, transforming husbands’ wages into a comfortable living for families. Only where men—and by extension companies—could not rely on unpaid family labor would FS step in to subsidize workers’ consumption.¹⁰⁵

By the 1970s, FUS was unwilling to do so on terms that provided secure employment to cooks. The company forced most to retire—some on partial pensions—in 1969. But patients still needed meals. Ángela Ballesteros vda. de Pizano was too young for retirement. She lost her job and the rights that came with it—to healthcare, sick days, vacation, a pension—but kept the work, now as an independent contractor.¹⁰⁶ The hospital no longer contracted with vendors that provided foodstuffs for waged, unionized cooks to prepare; it contracted with Ballesteros on a per-meal basis.¹⁰⁷ FUS thus unwound the internalization of cooking labor that FS had effected over the prior thirty years. Without knowing her expenses for food and labor (at least one of her daughters worked with her, though such family labor was often un- or underpaid), it is hard to say whether Ballesteros made as much as a contractor as she had as an employee.¹⁰⁸ What is certain is that her income depended on fluctuating occupancy rates—and she had lost the labor rights and union representation that would buffer against this insecurity.¹⁰⁹ The costs of her precarity reverberated through to patients, who complained of watered-down milk, rancid meat, and days-old beans the cook “tries to preserve with some ingredient.”¹¹⁰ A month later, the hospital was forced to increase her daily budget twenty percent. In a telling slip, a note on this agreement said that if “problems continue … we will lay her off.”¹¹¹ This was revised to reflect that Ballesteros had, in fact, already been laid off. Should problems continue, management would “cancel the agreement she has with the Company.”¹¹²

The formalization of care work and its removal from domestic space could be reversed or repeatedly permuted across the formal/informal divide. While this enabled nonnormative, women workers some access to the kinds of economic and social citizenship widely considered the prerogative of male breadwinners, men remained the normative derechohabientes, and it was in relation to their assumed needs that the company organized welfare. Women in these positions were relegated to the lowest tier of the wage hierarchy and liable to cycle in and out of employment. Managers insisted their work was “useless.”¹¹³ Meanwhile, male workers demanded more of their labor, while simultaneously relying on women’s proximity to domestic workers to not only make claims on them, but also denigrate them. Cleaner Ernestina Canché recalled that a coworker once lashed out, “that who was I to reprimand him, that I was no one … that I was a gata [a derogatory word for a woman domestic servant].”¹¹⁴

Conclusion

Mexico’s postrevolutionary state aspired, on paper, to guarantee universal social and economic rights. In practice, a patchwork quilt of public and employer-provided welfare services made full access to those rights contingent on certain kinds of employment; more than rights, these were privileges. In the exemplary but not unique case of railroads, these privileges—and the political and economic stability they underwrote—depended, in turn, on the exploitation of women workers.

As unions, labor law, and geographic exigency forced the valorization of some domestic-adjacent work within the rubric of industrial employment, companies and male workers attempted to preserve and exploit the elasticity, indeterminacy, and flexibility that had historically characterized “women’s work.” They were largely successful. But folding this work into railroad companies and unions on a formal basis was also transformative. It opened, implicitly, questions about what “women’s work” was and the terms on which women workers would provide it.

On the one hand, women’s work was an entitlement of railroad workers and families. Women’s exploitation was motivated by the overlapping pressures of worker demand for welfare rights and railroads’ interest in minimizing the costs of those rights. Both workers and companies interpellated these women as deferent and self-sacrificing, so that when things went wrong—when a child failed to learn to read, a sick toddler died, the breakfast budget ran out—those shortcomings and tragedies could be interpreted as failures of femininity, rather than as companies persistently under-providing resources for the welfare of workers and families. But such forms of femininity always exceeded attempts to articulate them straightforwardly to productive (or reproductive) processes.¹¹⁵ The workplace itself, with its union and legal protections, but also simply with its wage bargain, provided toeholds to contest this instrumentalization of gender. Moreover, it would be a mistake to overlook the simple, profound truth that these jobs provided women with rising real wages, union benefits, and stable employment—all without necessary dependence on a male breadwinner.

Taking labor rights not as abstract, static entitlements but as complex processes constituted by others’ labor shows that they were neither sewn from whole cloth nor permanently guaranteed. Rather, they depended on the exploitation of existing inequalities and could be hemmed in or revoked—with devastating consequences for both derechohabientes and the women workers who lost their jobs in the process. Making rights real was conflictual and quotidian, unfolding not just at the bargaining table or on the streets but in the interstices of everyday life, in the ways workers and companies laid claim to women’s labor, and in the conflicts over the terms on which women worked. These conflicts molded the archives that preserve them. Cook Eugenia de la Cruz died on August 5, 1977, after three decades on the railroad. Her death certificate lists her occupation: labores del hogar. Housework.¹¹⁶