Breastfeeding and the provision of human milk are associated with positive maternal and infant health outcomes immediately postpartum, across the life course, and across generations (AAP, 2012; Victora et al., 2016). Over the last two decades, there have been groundbreaking advances in the areas of human milk science and infectious diseases (Colt et al., 2017; Foeller et al., 2020; WHO, 2016). In the wake of COVID‐19, clinical case reports, retrospective and prospective studies, and systematic reviews related to perinatal maternal and child transmission of SARS‐CoV‐2 have been generated at an astonishing rate (Center for Humanitarian Health, 2020). Scientists around the world have united in a liquid gold rush to study human milk and COVID‐19. But, to what end?

A majority of the studies have been designed to determine if SARS‐CoV‐2 can be transmitted to an infant via human milk (WHO, n.d.). Some scientists are investigating maternal immune response to SARS‐CoV‐2, particularly how milk bioactives may modulate vertical transmission or otherwise affect clinical presentation/disease trajectories of infants (Fox et al., 2020 preprint). Others are engaged in research to ascertain whether Holder pasteurization or other techniques may be used to maintain a safe supply of donor human milk (assuming SARS‐CoV‐2 is even in the milk). The quest for effective COVID‐19 therapeutics raises questions about the possible excretion of drugs into milk, the relative risks of infant exposures to experimental drugs via milk, and the ethics of including pregnant and lactating people in clinical trials for COVID‐19 medicines and vaccines.

It is not easy to conduct human milk research during a pandemic. Recent systematic reviews highlight a consistent lack of quality evidence, largely due to small sample size (Lackey et al., 2020; Martins‐Filho, Santos, & Santos, 2020). Of the approximately 50 COVID‐19+ individuals who have had their milk tested, with some individuals having donated multiple samples, scientists were only able to find viral RNA in a small percentage, and repeat samples from the same individuals did not consistently yield identification of viral RNA. Further, viral RNA does not mean that the milk has any infectious SARS‐CoV‐2 in it (Chambers et al., 2020 preprint). Collecting milk samples for these kinds of studies is also difficult, given high risk of samples becoming contaminated through droplets the air, skin, or surfaces and containers. Lackey et al. (2020) note there are major gaps in the published methodology, particularly detailing how the milk samples were collected and analyzed, making it difficult to rule out external contagion as the source of viral RNA. Even so, they assert that methodological issues, such as matrix interference in the samples from the milk itself, rather than external sources of contamination probably explain the absence of detectable virus.

An even greater challenge is translating the limited and poor‐quality evidence into global public health and clinical practice guidance, a process that is fraught with uncertainty. Because breastfeeding and human milk are so critical to maternal and infant health outcomes, especially during public health emergencies (Gribble, McGrath, MacLaine, & Lhotska, 2011; Gribble 2017, developing recommendations for infant feeding must rely on complex decision‐making in which the risks, benefits, and costs of available alternatives are weighed (Angood, 2017). Currently, the WHO states: “…that mothers with suspected or confirmed COVID‐19 should be encouraged to initiate and continue breastfeeding. From the available evidence, mothers should be counselled that the benefits of breastfeeding substantially outweigh the potential risks of transmission.” (WHO, 2020, p. 42).

Nevertheless, clinical and laboratory‐based COVID‐19 studies are being used to cast suspicion on breastfeeding and the “quality” of human milk. For example, Wu et al., 2020 (p. 5) conclude that: “Our advice is against the use of breastfeeding even through breast expression; mothers with COVID‐19 should not breastfeed until after full recovery, when breast milk tests negative for the virus.” Fan et al. (2020, p. 6): “In our cases, breastfeeding is discouraged even though we did not detect SARS‐CoV‐2 in consecutive breastmilk samples during follow‐up.” Groß et al. (2020) suggest that more evidence is needed before health authorities can recommend “whether mothers with COVID‐19 should breastfeed (Groß et al., 2020, p. 1758)” despite inconsistent evidence of SARS‐CoV‐2 RNA in multiple samples of one person's milk. Even influential global health and healthcare authorities have put forth recommendations that contradict the WHO interim guidance for breastfeeding with COVID‐19 infection (Tomori, Gribble, Palmquist, Ververs, & Gross, 2020). Unlike Ebola, there is no consistent evidence yet that infants who acquire SARS‐CoV‐2 infection will experience severe adverse outcomes, at least if they have access to responsive quality health care (Walker et al., 2020). When critical care is needed and healthcare resources are limited, breastfeeding confers immunological protection to help infants have best chances of survival (WHO, 2020). Given all that is known about passive immunity and the evolutionary significance of human lactation in newborn adaptation to novel infections, and knowing the importance of breastfeeding in emergencies, how do we parse studies that portray breastfeeding and human milk as expendable during this pandemic?

Breastfeeding has long been at the center of infant feeding controversies that are deeply rooted in ideologies about mothers being intrinsically, biologically, a danger to their infants (Hausman, 2011; López, 2019; Palmquist, 2017; Palmquist, 2020; Tomori, 2015; Tomori, Palmquist, & Dowling, 2016). As illustrated above, much of human milk science remains firmly grounded in these ideologies and, wittingly or not, reproduces them. The fact is that simply isolating virus in human milk neither leads to more straightforward infant feeding recommendations (Tomori et al., 2020) nor does it necessarily lead to the desired behavioral outcomes (Tuthill, Tomori, Natta, & Coleman, 2019; Van Hollen, 2011).

The consequences of promoting breastfeeding cessation among infants at risk of perinatal HIV are instructive for COVID‐19. By the mid‐1980s human immunodeficiency virus (HIV) had become global pandemic. The initial WHO guidance recommended that breastfeeding continue, particularly in places where lack of access to clean water and a sustainable supply of infant formula would increase the risks of infant death due to diarrheal disease, acute respiratory infections, and malnutrition (WHO 1987). Early clinical studies estimated that the risk of transmission during breastfeeding was 29% to 36%, but that this risk was relative to both the severity and timing of maternal infection (Nduati & John, 1995). In response to a growing number of perinatal HIV cases in Botswana, the Ministry of Health (MoH) implemented a large‐scale formula feeding program to prevent HIV transmission through breastfeeding. Safeguards were implemented to offset the risks of formula feeding, such as ensuring caregivers had access to clean water, formula distribution in community clinics, and counseling about safer formula preparation and feeding.

The effects of replacement feeding were not as expected, however. A large randomized control trial for perinatal HIV showed no net benefit of formula feeding in infant morbidity or mortality at 18 months (Thior et al., 2006). Moreover, the unintended consequences of this intervention were cast into sharp relief after a massive flooding event in 2006 where nearly all hospital admissions and deaths were among formula fed infants (Arvelo et al., 2010; Creek et al., 2010; Mach et al., 2009). This dual public health catastrophe raised serious concerns about the relative ethics and safety of replacement feeding for HIV, as well as the need for comprehensive policies, guidance, and programming for infant feeding in emergencies (Angood, 2017). Today, the WHO (2016) recommends that HIV‐positive mothers continue to exclusively breastfeed for the first 6 months, along with combined anti‐retroviral therapy (cART) whenever available, especially in areas where formula feeding is not acceptable, feasible, affordable, sustainable, and safe. Experts have argued more recently that failing to support informed decision‐making for breastfeeding in the context of HIV infection, particularly in the U.S. where high‐quality HIV testing and cART are readily available, is unethical (Gross, Taylor, Tomori, & Coleman, 2019).

With this said, there tends to be a group of scientists and healthcare professionals that generally operate as if their work is neither political nor biased. Yet, anthropologists and others have produced decades of scholarship describing how science and biomedicine are inherently political and biased (Benton, 2016; Franklin, 1995; Hahn & Kleinman, 1983; Hardeman, Murphy, Karbeah, & Kozhimannil, 2018; Lock & Nguyen 2010; Martin, 2016; Rhodes, 1990; Sangaramoorthy, 2014; Skloot, 2011; Tallbear, 2013; Wilce Jr., 2003). Black, Indigenous, People of Color (BIPOC) scholars of have also drawn critical attention to how racism and anti‐Blackness in science and medicine have been weaponized over centuries to maintain white supremacy through reproductive harm (Bridges, 2011; Davis‐Floyd, Gutschow, & Schwartz, 2020; McLemore et al., 2019; Mullings & Wali, 2001; Owens & Fett, 2019; Roberts, 2011; Torres, 2019). COVID‐19 has only amplified these biases (Hall et al., 2020), and there is mounting evidence that the science used to support perinatal separation policies for COVID‐19, including strongly advising against breastfeeding or provision of human milk with SARS‐CoV‐2 infection (Tomori et al., 2020) are disproportionately harming BIPOC (Allers, 2020; Davis‐Floyd et al., 2020; Furlow, 2020a, 2020b; Thayer this issue). Unequivocal recommendations to disrupt lactation due to COVID‐19 reveal a blatant disregard of the potential harms that hang in the balance for parents and infants. It reflects a willingness to make infant feeding recommendations uncritically, ahistorically, and without regard for the basic human right to informed‐decision making and patient‐centered care (Tomori et al., 2020).

The largest geographic concentration of human milk laboratories and scientific training centers for human milk research are located in North America. Disparities in COVID‐19 cases in the U.S. and Canada provide a stark reminder of how racism, colonialism, xenophobia, and all the related structural inequities, which are also embedded in science and medicine, disproportionately harm BIPOC (Hooper, Nápoles, & Pérez‐Stable, 2020; Poteat, Millett, Nelson, & Beyrer, 2020; van Dorn, Cooney, & Sabin, 2020). Human milk scientists studying SARS‐CoV‐2 must reckon with the limitations and unintended, yet predictable, consequences of extractive clinical and laboratory‐based studies that do not adequately account for context, power and privilege, racism, colonialism, or nuance in designing human milk studies and interpreting scientific data. The politics that permeate scientific studies of human milk are made visible by interrogating fundamental assumptions driving the research questions, study methodologies, (in) consistencies in interpretation of findings across the literature, and lack of diversity, equity, and inclusion among the study participants and within investigator teams. Like other science, technology, and engineering fields, the field of human milk research is characterized by inequitable access to capital, resources, laboratory space, institutional infrastructures, to educational and training opportunities, and most importantly, access to human milk. There are serious racial disparities in access, power, and privilege that continue to drive human milk research across this global and interdisciplinary field, which is in many cases is extractive, mired in financial conflicts of interest, and largely disengaged from conversations regarding broader political economic contexts of the work. When the scientific teams of clinical research studies do not include BIPOC as principal investigators, fail to engage in ethical, community‐centered, co‐creation of biobanks and research studies, and do not articulate a commitment to racial equity in the research enterprises, they risk simply reproducing harmful, and deeply racist scientific enterprises. Anti‐racist, decolonized methodologies and ethical funding streams are needed in this work (McLemore et al., 2019; Scott, Bray, & McLemore, 2020).

Perhaps one of the most critical perspectives that anthropology and other disciplines can offer is that studying human milk outside of human lived experiences is not only extremely limited, it is potentially harmful. Anthropologists have long studied both endemic and emergent infectious diseases as one way of understanding the confluence of ecological, biological, and sociocultural factors that have shaped human evolutionary history, biological variation, and contemporary disease patterns (Inhorn & Brown, 1990; Sangaramoorthy, 2020; Zuckerman, Harper, Barrett, & Armelagos, 2014). Anthropological studies of human milk provide clues to the breadth and depth of human biological diversity and adaptability (Hinde & Milligan, 2011; Klein et al., 2018; Miller 2018; Miller et al., 2013; Power & Schulkin, 2016; Quinn 2016; Quinn & Childs, 2017). Research is needed that more fully contextualizes lactation within the complexity of COVID‐19 maternal‐infant disease dynamics, in which human milk as part of a more holistic picture of diverse maternal‐infant responses to infection (Table 1). Human milk cannot be understood separately from human life and more importantly, human milk cannot be understood separately from breastfeeding (Cassidy & El Tom 2015; Tomori, Palmquist, & Quinn, 2018; Van Esterik, 2015).

母乳喂养和提供母乳与产后即刻，整个生命历程以及几代人之间的母婴健康积极结果相关（AAP，2012； Victora等，2016）。在过去的二十年中，人类乳科学和传染病领域取得了突破性的进展（Colt等人，2017年; Foeller等人，2020年;世卫组织，2016年））。随着COVID-19的出现，临床病例报告，回顾性研究和前瞻性研究以及与围产期SARS-CoV-2的母婴传播有关的系统评价令人惊讶（人道主义卫生中心，2020年）。世界各地的科学家联合起来，进行一次淘金热，研究人乳和COVID-19。但是，目的是什么？

大多数研究旨在确定SARS-CoV-2是否可以通过母乳传播给婴儿（WHO，nd）。一些科学家正在研究孕妇对SARS-CoV-2的免疫反应，特别是牛奶生物活性物质如何调节垂直传播或以其他方式影响婴儿的临床表现/疾病轨迹（Fox等人，2020年预印本）。其他人正在进行研究以确定是否可以使用Holder巴氏灭菌法或其他技术来维持供体人乳的安全供应（假设SARS‐CoV-2甚至存在于牛奶中）。对有效COVID-19疗法的追求提出了以下问题：药物可能从乳汁中排泄，婴儿通过乳汁暴露于实验药物的相对风险，以及在COVID-19药物和疫苗的临床试验中包括怀孕和哺乳期人群的道德规范。

在大流行期间进行人乳研究并不容易。最近的系统评价强调了始终缺乏质量证据，这主要是由于样本量较小（Lackey等人，2020年； Martins-Filho，Santos和Santos，2020年）。在大约50名接受过牛奶测试的COVID-19 +个体中，有些个体捐赠了多个样本，科学家只能在很小的百分比中找到病毒RNA，而同一个体的重复样本并不能始终如一地鉴定出病毒RNA。此外，病毒RNA并不意味着牛奶中含有任何传染性SARS-CoV-2（Chambers等人，2020年预印本）。考虑到样品被水滴，空气，皮肤，表面和容器污染的风险很高，因此收集这类样品的牛奶也很困难。Lackey等。（2020年）指出，已发表的方法学存在重大差距，特别是详细介绍了如何收集和分析牛奶样品，这使得很难排除外部感染是病毒RNA的来源。即使如此，他们仍认为方法问题，例如牛奶本身样品中的基质干扰，而不是外部污染源，可能解释了缺乏可检测的病毒。

更大的挑战是将有限且质量差的证据转化为全球公共卫生和临床实践指南，这一过程充满不确定性。由于母乳喂养和母乳对母婴健康结局至关重要，尤其是在突发公共卫生事件期间（Gribble，McGrath，MacLaine和Lhotska，2011年; Gribble 2017年），为婴儿喂养制定建议必须依靠复杂的决策，其中权衡可用替代品的风险，收益和成本（Angood，2017）。目前，世界卫生组织指出：“……应鼓励怀疑或确诊COVID-19的母亲开始并继续母乳喂养。从现有证据中，应该建议母亲，母乳喂养的好处远大于传播的潜在风险。” （世界卫生组织，2020年，第42页）。

尽管如此，临床和实验室COVID-19研究仍被用于怀疑母乳喂养和母乳的“质量”。例如，Wu et al。，2020（p。5）得出以下结论：“我们的建议反对即使通过乳房表情也要使用母乳喂养；患有COVID-19的母亲在母乳中对该病毒呈阴性反应之前，应等到完全康复后再进行母乳喂养。” 范等。（2020年，第6页）：“在我们的案例中，即使我们在随访期间未在连续的母乳样本中检测到SARS-CoV-2，也建议不要进行母乳喂养。” Groß等。（2020年）表明，在卫生当局建议“ COVID-19的母亲是否应该母乳喂养之前，需要更多的证据（Groß等人，2020，页。1758）”，尽管一个人的牛奶的多个样本中存在SARS-CoV-2 RNA的证据不一致。甚至具有影响力的全球卫生和医疗机构也提出了与WHO的COVID-19感染母乳喂养暂行指南相抵触的建议（Tomori，Gribble，Palmquist，Ververs和Gross，2020年）。与埃博拉病毒不同，目前尚无一致的证据表明，至少在获得SARS-CoV-2感染的婴儿可获得优质医疗服务的情况下，他们会遭受严重的不良后果（Walker等人，2020年）。当需要重症监护且医疗资源有限时，母乳喂养可提供免疫保护以帮助婴儿获得最大的生存机会（世卫组织，2020年）。鉴于所有关于被动免疫的知识以及人类泌乳在新生儿适应新型感染中的进化意义，并且知道母乳喂养在紧急情况下的重要性，我们如何分析将母乳喂养和母乳视为在这种大流行中可消耗的研究？

长期以来，母乳喂养一直是婴儿喂养争论的焦点，这种争论深深植根于意识形态上，即母亲在本质上，生物学上对婴儿构成威胁（Hausman，2011 ;López，2019; Palmquist，2017 ; Palmquist，2020; Tomori，2015 ; Tomori，Palmquist和Dowling，2016年）。如上所示，许多人类乳科学仍然牢牢扎根于这些意识形态，无论有意还是无意地复制了它们。事实是，仅分离母乳中的病毒既不会导致更直接的婴儿喂养建议（Tomori等人，2020），也不一定会导致所需的行为结果（Tuthill，Tomori，Natta和Coleman，2019年）; Van Hollen，2011年）。

在围生期感染HIV的婴儿中促进母乳喂养的后果对COVID-19具有指导意义。到1980年代中期，人类免疫缺陷病毒（HIV）成为全球大流行病。世卫组织最初的指导意见建议继续母乳喂养，特别是在缺少清洁水和可持续供应婴儿配方奶的地方，这会增加由于腹泻病，急性呼吸道感染和营养不良造成婴儿死亡的风险（WHO 1987）。早期临床研究估计，母乳喂养期间传播的风险为29％至36％，但这种风险与母体感染的严重程度和时机有关（Nduati＆John，1995年））。为了应对博茨瓦纳围产期艾滋病毒病例的增加，卫生部（MoH）实施了大规模的配方奶喂养计划，以防止艾滋病毒通过母乳喂养传播。采取了保障措施来抵消配方奶喂养的风险，例如确保护理人员能够获得清洁水，在社区诊所分发配方奶，并就更安全的配方制备和喂养提供咨询。

但是，替代喂养的效果并不如预期。一项关于围产期HIV的大型随机对照试验表明，在18个月的婴儿发病率或死亡率方面，配方奶喂养没有净收益（Thior等人，2006年）。此外，在2006年发生大规模水灾之后，几乎所有的住院和死亡病例都在配方奶粉喂养的婴儿中，这一干预措施的意想不到的后果大大减轻了（Arvelo等人，2010； Creek等人，2010； Mach等人， 。，2009）。这种双重公共卫生灾难引起了人们对替代喂养艾滋病毒的相对道德和安全性以及紧急情况下婴儿喂养的综合政策，指导和计划的严重关注（Angood，2017）。如今，WHO（2016）建议艾滋病毒呈阳性的母亲在头6个月内继续纯母乳喂养，并在可能的情况下继续采用联合抗逆转录病毒疗法（cART），尤其是在不能接受，可行，负担得起的配方奶喂养的地区，可持续和安全。最近，专家认为，在艾滋病毒感染的情况下，不能支持做出明智的母乳喂养决策，特别是在美国，那里容易获得高质量的艾滋病毒检测和cART，这是不道德的（Gross，Taylor，Tomori和Coleman，2019）。

如此说来，往往会有一群科学家和卫生保健专业人员在运作，好像他们的工作既不政治也不偏颇。然而，人类学家和其他学者已经获得了数十年的奖学金，描述了科学和生物医学在本质上是政治性和偏见性的（Benton，2016年; Franklin，1995年; Hahn＆Kleinman，1983年; Hardeman，Murphy，Karbeah＆Kozhimannil，2018年; Lock＆Nguyen 2010年;马丁（Martin），2016年;罗德（Rhodes），1990年;桑加拉莫西（Sangaramoorthy），2014年;斯库洛特（Skloot），2011年;塔尔比（Tallbear），2013年;小威尔斯（Wilce ），2003年）。黑人，土著，有色人种（BIPOC）的学者也引起了人们的极大关注，即如何在过去的几个世纪中将科学和医学中的种族主义和反黑人现象武器化，以通过生殖伤害维持白人至上的地位（Bridges，2011年； Davis-Floyd，Gutschow ，＆Schwartz，2020； McLemore等，2019； Mullings＆Wali，2001； Owens＆Fett，2019； Roberts，2011； Torres，2019）。COVID-19仅放大了这些偏差（Hall等人，2020年），并且越来越多的证据表明，用于支持COVID-19围产期分离政策的科学，包括强烈建议不要母乳喂养或提供SARS-CoV-2感染的母乳（Tomori等人，2020年）正在严重损害BIPOC。 （Allers，2020 ; Davis-Floyd等人，2020 ; Furlow，2020a，2020b; Thayer这个问题）。明确建议破坏由于COVID-19引起的泌乳，揭示了公然无视父母和婴儿平衡中潜在的危害。它反映了愿意不加批判地，从历史上提出婴儿喂养建议的意愿，而不考虑知情决策和以患者为中心的护理的基本人权（Tomori等人，2020年）。

北美地区是人类乳汁实验室和科学培训中心的最大地理集中地。美国和加拿大的COVID-19病例差异明显提醒了种族主义，殖民主义，仇外心理以及所有相关的结构性不平等现象，这些不平等现象也深深扎根于BIPOC（Hooper，Nápoles和Pérez-马able ，2020 ;波蒂，米利特，尼尔森和贝勒，2020 ;范多恩，库尼和萨宾，2020）。研究SARS-CoV-2的母乳科学家必须考虑提取性临床和实验室研究的局限性和意想不到但可预测的后果，这些研究并未充分考虑背景，权力和特权，种族主义，殖民主义或设计人的细微差别牛奶研究和解释科学数据。通过质疑驱动研究问题的基本假设，研究方法，文献解释结果的一致性以及研究参与者和研究者之间缺乏多样性，公平性和包容性，使渗透到母乳科学研究的政治变得可见。在研究人员团队中。与其他科学，技术和工程领域一样，母乳研究领域的特点是获得资金，资源，实验室空间，机构基础设施，教育和培训机会，最重要的是获得人乳。在获取，权力和特权方面存在着严重的种族差异，继续在这个全球性和跨学科的领域推动人类牛奶研究，在许多情况下这是提取性的，陷入了金融利益冲突中，并且很大程度上脱离了有关更广泛的政治经济学的讨论工作的环境。当临床研究的科研团队不包括BIPOC作为主要研究人员，未能参与以伦理为基础，以社区为中心，共同创建的生物库和研究活动，并且没有在研究企业中明确表达对种族平等的承诺时，冒险只会简单地复制有害的种族主义科学企业。反种族主义者，2019 ; Scott，Bray和McLemore，2020年）。

人类学和其他学科可以提供的最关键的观点之一是，在人类生活经历之外研究人乳不仅非常有限，而且可能有害。人类学家长期以来一直对地方性和新兴传染病进行研究，以此作为了解生态，生物和社会文化因素融合的一种方式，这些因素已经塑造了人类的进化历史，生物变异和当代疾病模式（Inhorn＆Brown，1990； Sangaramoorthy，2020； Henry ＆Brown，1990）。 Zuckerman，Harper，Barrett和Armelagos，2014年）。人类乳汁的人类学研究为人类生物多样性和适应性的广度和深度提供了线索（Hinde＆Milligan，2011; Klein et al。，2018; 米勒2018; Miller等，2013；Power＆Schulkin，2016年；奎因2016; 奎因和查尔兹（Quinn＆Childs），2017年）。需要进行研究，以便在COVID-19母婴疾病动态的复杂性中更全面地了解哺乳情况，其中母乳是更全面的母婴对感染反应的整体描述的一部分（表1）。不能将母乳与人类生活分开理解，更重要的是，不能将母乳与母乳喂养分开理解（Cassidy＆El Tom 2015； Tomori，Palmquist和Quinn，2018； Van Esterik，2015）。