RISK AND PUBLIC COMMUNICATION

The problem of environmental contaminants in breastmilk generates headlines that sensationalise and concretise very speculative conclusions in a rapidly changing research milieu where many questions remain unanswered or unasked. In these discussions, women's bodies are the subject of interest, and the cause of the problem is presented as localised in women's bodies. Often there is a simple solution offered to a complex problem - use infant formula. Infant formula industries profit each time that a woman chooses a commercial substitute because she fears her breastmilk is polluted.

Research on contaminants in breastmilk assumes that there is a clear boundary between body/self and the outside/external world. With the hostile polluted world invading the pure body, this discourse reinforces the idea that breastmilk may no longer be "safe" or "pure", but rather a vulnerable substance permeable to environmental toxins; this discourse blames the breastfeeding mother as the toxic source. It is thus very important to examine how this subject is presented to the public.

"The public" is not necessarily aware of the literature reviewed in the previous section of this report. It is specialised knowledge, mostly in the hands and heads of experts-medical researchers, toxicologists, ecologists, among others. "The public" has access to this specialised knowledge through newspaper stories, radio and television coverage, and books that `translate' science for public consumption. A number of accessible books draw on studies of breastmilk contamination to dramatise environmental pollution. Some of the key syntheses of literature on environmental hazards fail to place the evidence on contaminants in breastmilk in a broad perspective of costs and benefits to mothers. Others scare without suggesting solutions; most fail to discuss the risks of alternative feeding methods. This review considers first four popular works that discuss breastfeeding in the context of broader questions about environmental health risks, Our Stolen Future by Colborn, Dumanoski, and Myers (1996), Mad Cows and Mother's Milk by Powell and Leiss (1997), and Living Downstream and Having Faith by Steingraber (1997, 2001). Most authors writing on this topic refer to the importance of Rachel Carson's Silent Spring (1962) as the starting point for examining "matter out of place," in this case, insecticides in breastmilk. Carson wrote that "the breastfed human infant is receiving small but regular additions to the load of toxic chemicals building up in his body" (1962:23). Other writers have elaborated on this observation.

Our Stolen Future argues that "while prenatal exposure seems to pose the greatest hazard, health specialists also worry about the chemicals passed on in breastmilk because some sensitive developmental processes continue in the weeks immediately after birth. During breastfeeding, human infants are exposed to higher concentrations of these chemicals than at any subsequent time in their lives. In just six months of breastfeeding, a baby in the United States and Europe gets the maximum recommended lifetime dose of dioxin, which rides through the food web like PCBs and DDT. The same breastfeeding baby gets five times the allowable daily level of PCBs set by international health standards for a 150-pound adult" (Colborn, Dumanoski and Myers 1996:106-7). They note that the contamination of breastmilk has been particularly severe among indigenous people in the high Arctic, where many people still eat the wild food the land and sea provide. There, researchers have found that babies take in seven times more PCBs than the typical infant in southern Canada or the United States. Many of these statements are made without citations. Animal studies are crucially important to their argument.

After birth, the transfer of contaminants to the offspring continues through the rich, fatty breastmilk. While nursing, a mammalian mother (including humans) draws down her fat stores, dumping not only the fat but also the persistent toxic chemicals she has accumulated in her body fat over the years into her milk. In this way, a load of contaminants that it has taken the mother decades to accumulate is passed onto her baby in a very short time. By the time a baby beluga stops nursing at two years of age, it will have acquired a toxic load that, relative to its size, far exceeds that of its mother. (Colborn, Dumanoski and Myers 1996:145-6)

The analogies to humans are then further developed:

As with other animals, human infants take in heavy doses of PCBs and other contaminants in breastmilk, which exposes them to levels ten to forty times greater than the daily exposure for an adult. Several studies report that infertile men have higher levels of PCBs and other synthetic chemicals in their blood or semen, and one analysis found a correlation between the swimming ability of a man's sperm and the concentrations of specific members of the PCB family found in his semen. (Colborn, Dumanoski and Myers 1996:178)

One of the most widely cited statements in their book is that, "based on the concentrations in breastmilk fat of PCBs, some have estimated that at least five percent of babies in the United States are exposed to sufficient levels of contaminants to cause neurological impairment (Colborn, Dumanoski and Myers 1996:188). While acknowledging the benefits of breastfeeding, which provides infants with important immune protection and substances that enhance development, they argue that breastfeeding exposes infants to disturbing levels of chemical contaminants, including a number of known hormone disrupters. "According to various studies of breast milk contamination, nursing babies take in the highest doses of contaminants they will experience in their entire lives - levels ten to forty times greater than the daily exposure of an adult. Breastfeeding is the only efficient way to remove these persistent chemicals from the human body" (Colborn, Dumanoski and Myers 1996:215-16).

They conclude that too little is known to judge how the undeniable benefits of breastfeeding balance against the risks of transferring hormonally active contaminants. By saying that it is premature to advise women against breastfeeding, they have already raised fears, even though the transfer of contaminants before birth may have a greater impact than any transfer taking place during breastfeeding. They call for research to determine whether the concentrations of hormone-disrupting chemicals in human milk pose enough of a hazard to make breastfeeding inadvisable for some women, perhaps those having a first child later in life and who may have a high burden of persistent chemicals in your bodies. They write; "We cannot afford to ignore the pressing issue of persistent contaminants when weighing the merits of breastfeeding against alternatives such as bottle feeding with a formula based on cow's milk" (Colborn, Dumanoski and Myers 1996:216). After making dramatic statements about the risks of contamination, they ask how much this matters if children born to mothers with contaminated breastmilk have already been exposed in the womb. Will the additional exposure through breastmilk greatly increase the risk? Are there breastfeeding regimens that can lower the transfer rate of contaminants from mother to baby while maintaining the benefits? (Colborn, Dumanoski and Myers 1996:225).

Mad Cows and Mother's Milk provides six case studies of "poor risk communication" including mad cow disease, E. coli outbreaks, silicone breast implants, and contaminants in mother's milk. In the latter case study, Powell and Leiss (1997) document some PCB accidents that lead to toxic effects in humans. In 1985, about four hundred litres of PCB-contaminated oil spilled onto the Trans-Canada highway near Kenora. In 1988, fifteen hundred barrels of PCB-containing oil caught fire in a chemical storage warehouse near Montreal. Although the fire was quickly contained, there was concern about health risks. "The next day, nursing mothers were warned to stop breastfeeding," and the next month, they were told they could resume breastfeeding (Powell and Leiss 1997:186). These and other problems concerning the disposal of PCBs brought the problem to the public's attention. However, the authors acknowledge that the analysis and testing of PCBs is not always accurate and the scientific literature on PCBs is vast, complex, rich in qualifying statements and difficult to summarize (Powell and Leiss 1997:189), an assessment with which this author is in total agreement. Since 1982, PCBs in breastmilk in Canada compare favourably with those in other western countries, and are on the decline (Powell and Leiss 1997:191).

The authors use the case of the Canadian arctic where some Inuit communities living off marine resources such as seals, fish, walrus and whale were reported to have high body burdens of contaminants, including PCBs. Breastmilk, like other valued locally produced food, was considered superior to processed foods from the south. These strength-giving foods formed the basis of identity and social well being, placing Inuit in a double bind since fatty acids from marine animals may provide protection against cardiovascular diseases, and counteract mercury and PCB induced toxicity (Powell and Leiss 1997:199).

On the other hand, studies from Laval University in the eighties confirmed very high concentrations of PCBs in the breastmilk of Inuit women - many times the amount of PCBs considered tolerable by Health Canada. Nevertheless, the Inuit made a policy decision that breast is best. But will continued monitoring of breastmilk and its effects on infants be just as detrimental as saying that there are high concentrations of PCBs in breastmilk? A resource committee to link scientists and the population will address this concern. Once doubts are cast on traditional practices like breastfeeding, eating traditional foods and hunting sea mammals, how is confidence restored? Or is it like cutting down the last tree; confidence cannot be restored once it is gone. Continuing research on breastmilk fuels speculation that a problem exists (Powell and Leiss 1997:205). Certainly research has not stopped on this population. Dewailly and colleagues have published another study on toxins and breastfeeding among the Inuit. The most recent study examines the susceptibility to infections and immune status in breastfed and bottle-fed Inuit infants exposed to organochlorines, concluding that prenatal exposure to organochlorines could contribute to the risk of ear infections (otitis media). However, there were few differences between breast and bottle-fed infants; breastfed infants experienced slightly fewer acute episodes of otitis media and pulmonary infections. This study contrasted ever-breastfed with never-breastfed infants, but provided no evidence on the role of exclusive breastfeeding. Prenatal exposure to pollutants was determined from breastmilk samples. The authors recommend that "a reduction of organochlorine body burden in Inuit women of reproductive age seems desirable", in addition to encouraging the use of traditional foods low in contaminants, such as red char (Dewailly et al 2000:210). Women were advised not to modify their breastfeeding practices.

Powell and Leiss (1997:205) also contrast contaminant information with contaminant gossip, almost as dangerous for Inuit mothers. The research cited raises the question of what concentration of PCBs, or any other substance in breastmilk, would constitute a threshold that would contraindicate breastfeeding (Powell and Leiss 1997:207). The issue is not what would contraindicate breastfeeding but how this information about risks impacts on mothers' decision-making process on infant feeding. For example, perhaps mixed feeding represents the worst of both worlds, so that the toxic load doubles. Without more information on exclusive breastfeeding and mixed feeding, these questions remain unanswered. Lack of exclusive breastfeeding means that an infant receives the toxins from breastmilk and the toxins from some uncalculated combination of water, cow's milk, and infant formula.

Sandra Steingraber's stunning analysis of cancer and the environment, Living Downstream (1997), uses toxic-release and cancer registry data to make the connections between environmental contaminants, the food we eat, and incidence rates of cancer. As an ecologist, she uses a vast range of evidence to document the environment's contribution to cancer. Breasts and breastmilk figure prominently in the story as the following quotes demonstrate: "Because of the immutable fact of biomagnification, human milk will always be more contaminated than soy or dairy milk as the machinery of the human breast distills its product from the plant and animal foods consumed by the nursing mother" (1997:9); breasts as "repositories for synthetic organic chemicals circulating within the female body." (1997:93); 25% of breastmilk is "too contaminated to be bottled and sold as a food commodity" (1997:238); and "...during the intimate act of nursing, a burden of public poisons - insect killers, electrical insulating fluids, industrial solvents, and incinerator residues - is shifted from one generation into the tiny bodies of the next" (1997:238). To her, breastfeeding is both a window to find out toxic levels in women's bodies, and a "sacred landscape in need of protection." This is the story that Sandra Steingraber addresses in her new book, Having Faith: The Ecology of Pregnancy, Childbirth and Breastfeeding (2001). In it, she integrates the scientific evidence about how the environmental toxins threaten all stages of infant development from the moment of conception with her own experience of pregnancy, childbirth and breastfeeding. She exposes the dangerous disconnections between scientific knowledge of toxins and what pregnant women are told. In both books, Steingraber uses the language of body burden, the sum of total of exposures, including all routes of entry (inhalation, ingestion, and skin absorption) and all sources (food, air, water, workplace, home, etc.), thus drawing attention to all bodies, and not just female bodies. Because of the power and clarity of her writing and analysis, her statements about breastmilk contamination in Living Downstream and Having Faith may be cited out of context, instead of being used to bring the breastfeeding and environmental movements closer together.

Breastfeeding as a media subject is both sexy and emotional. It is an irresistible topic because every story has the potential to be sensational and controversial. The emotional ambivalence of breastfeeding expressed by many individuals and groups heightens the drama of stories. In the case of environmental pollution and toxins in breastmilk, there are both contradictory messages from the experts, as well as differences in interpretation by the media. Breastfeeding advocates are vulnerable if they ignore or deny potential risks associated with breastfeeding. They are also at times defensive, for they speak on behalf of a product - breastmilk - with no commercial endorsement, and on behalf of a life-sustaining process-breastfeeding. Advocates may react with appropriate (or in the case of emotional overreaction, perhaps inappropriate) responses that will make a media splash, creating an even more sensational story. Breastfeeding mothers may be impervious to or overreact to perceived threats to breastfeeding because they are so immediately bound up in the protection of their breastfed infants. When mothers are no longer breastfeeding, and have more distance, they may well react with political action. As the workshop to be discussed below demonstrated, breastfeeding mothers in North America tend to turn "in," not "out" until they have ended breastfeeding. Since breastfeeding advocacy groups have had several decades of training in dealing with industry, they are now skilled in using the media. Consider, for example, the La Leche League response to Greenpeace, the work of IBFAN on Code Compliance, and WABA's World Breastfeeding Week.

Risk and the Media
How do different groups communicate about risk, and how do they act on information, which is often inaccessible and not "user friendly"? What are the interconnections between academic, policy and advocacy communication? For example, when the Laval studies confirmed the presence of toxins in the breastmilk of Inuit women, Greenpeace and other environmental advocacy groups drew media attention to contaminated mother's milk. Greenpeace uses mass media to change public opinion using the international news services and focussed political campaigns using direct action spectacles, mass media campaigning and diplomatic lobbying (Dale 1997:3).

Breastfeeding advocacy groups worked to contain the damage done by the reports on contaminated breastmilk by publicizing the contamination of water and other foods. Similarly, when the New York Times accused breastfeeding advocates of suppressing "scientific evidence" about HIV/AIDS transmission through breastmilk, infant formula manufacturers offered to sell infant formula to Ministries of Health in affected countries. In this way, they "redeem" themselves in the public eye, for their aggressive promotion of their products in those same countries that now accept their offers of subsidized infant formula. Because of the political implications of all these messages, communicators must walk the fine line between "scare tactics" and "suppressing the evidence". Neither toxins in breastmilk nor HIV/AIDS and breastmilk are amenable to the 30 second sound bite required by many media sources.

In the nineties, breastfeeding advocates and NGOs documented and tried to publicize the hazards of infant formula as determined by clinical evidence provided by medical research. But this material is rarely publicized by the media, and is primarily circulated among breastfeeding advocates. Health professionals have generally not heeded warnings concerning the hazards of infant formulas, stressing instead their convenience and reliability particularly in comparison to post-war home-made mixtures. Consequently, medical training and practice since the fifties has stressed the equivalence of breast and bottle-feeding, of breastmilk and breastmilk substitutes.

Scientific research about the uniqueness of human milk as a living substance with long lasting effects extending well into adulthood (for example, reduction in allergies and asthma) has moved into public discourse through radio talk shows, TV, and magazines. Yet, media also draw attention to every example where mother's milk has failed to nurture with a vigour totally absent from discussions of the hazards of infant formula which remains a discourse developed and publicized by breastfeeding advocates. For example, Newsweek (Spring 1999) in a special supplement on Women's Health, devoted one page to breastfeeding, headlined "Nursing Trouble." The
article discussed cracked nipples, engorgement, insufficient milk, and the death of breastfed infants by starvation and dehydration.

Breastfeeding is one point of entry for talking about the much broader questions of contamination, pollution, and environmental health. But there are often serious discrepancies in the way environmental pollution and breastfeeding are thought and talked about by environmentalists and breastfeeding advocates. The environmental literature speaks of breastmilk as a warning system for environmental exposure. But breastfeeding advocates speak of breastmilk as total nutrition for an infant from birth to 6 months. Breastfeeding advocates stress that breastfeeding provides some protection from breast cancer; while environmental groups point out the substances in breastmilk that are carcinogenic. Environmentalists accuse breastfeeding advocates of burying their head in the sand and suppressing information critical of breastfeeding. Headlines about polluted mother's milk signify ultimate sacrilege, but they seldom suggest solutions. Evidence of pollutants in breastmilk tell us about serious environmental problems for children. It does not, however, tell us about serious environmental problems caused by breastfeeding, as the previous research review demonstrated.

Michel Odent, founder of Primal Health Research (London), argues that intrauterine pollution represents the most serious health risk. Intrauterine contamination occurs earlier, is inevitable and is therefore of more concern. He argues that human livers are unable to detoxify these new chemicals that they have not been exposed to during evolution, and proposes that women should mobilize fat-soluble pollutants by fasting before conception. He also draws
attention to recent research suggesting the importance of male-
mediated developmental toxicity, caused by paternal exposure to man-made chemicals. Animal studies showed effects such as birth defects, tumours, and increased stillbirth and neonatal deaths, with some effects transmitted to second and third generations (Infante-Rivard and Sinnett 1999).

The WWF-UK report, Chemical Trespass: A Toxic Legacy (1999) reviewed the potential effects of global contamination on humans. Following many pages of tables of research results showing dangerous chemicals found in human body fat, the report highlights contaminants in human milk, and the levels of contamination in UK breastmilk. The recommendations include investigating the extent of human breastmilk contamination. The report received wide
media coverage in Europe and worldwide. As a result of raised public awareness about pollutants, the European Union made a political decision to adopt the precautionary principle rather than
demanding more research studies to prove that pollutants are at a level to cause harm. The decision resulted in use restrictions on a flame retardant, pentaBDE, considered to be an endocrine disrupting chemical. While the European Union acknowledged that more information was required, the precautionary principle was invoked to begin risk reduction activities immediately (National Toxics Network, Australia). The report was intended to raise awareness about environmental pollution not to undermine breastfeeding. WWF says breastfeeding advocates should communicate about the hazards of infant formula, environmentalists should communicate about toxins in breastmilk. They are "clean" as long as they say breast is best. But the most important story is in the placing of breastfeeding in a broader environmental health context.

Responding to the WWF-UK report, Ros Coward begins her newspaper article entitled "If breast isn't best" with a statement few in the women's health movement could dispute: "Media health scares come and go, doing little apart from inducing anxiety and confirming existing prejudices" (1999). This statement assumes that the dispute is about discourse not about dioxins. She argues that breastfeeding is a contested area with vested interests; however, the vested interests of not breastfeeding are more significant. Her
article accuses breastfeeding advocates of "preventing further discussion" because they don't like to talk about problems with breastfeeding. Similarly newspaper reports accused UNICEF of suppressing information on HIV/AIDS transmission through breastmilk.

Newspaper headlines do not always reflect article content. Nevertheless, these headlines from European and North American newspapers collected by GIFA (Geneva Infant Feeding Association) from 1980-2000 provide a sample to demonstrate the way information about contaminants in breastmilk is presented (compared with problems with infant formula, much more rarely discussed):

THE HIDDEN DANGERS IN MOTHER'S MILK
MANY POISONS IN MOTHER'S MILK
DIOXINS IN BREAST MILK REACH ALARMING HIGH
POISONS THREAT TO BREASTFED BABIES
BABIES IN POISON PERIL FROM BREASTFEEDING
SCIENTISTS FIND DEADLY TOXINS IN MOTHERS' MILK
BREASTFEEDING ISN'T ALWAYS PICTURE PERFECT
CALL FOR ACTION OVER POLLUTED BREASTMILK
MUM'S MILK 'POISON RISK'
IF BREAST ISN'T BEST
IS BREAST BEST?
BREAST WAS BEST
BREASTFEEDING MOTHERS MAY PASS TOXINS TO BABIES
POISONS FOUND IN BREAST MILK
ALARM OVER TOXINS IN MOTHERS' MILK
NEW STUDY INTO TAINTED BREAST MILK IS LAUNCHED
NEW SWEDISH STUDY ON CONTAMINANTS IN BREAST MILK: BREASTFEEDING INCREASES THE RISK OF CHILDHOOD CANCER
POLLUTANTS THAT 'PUT BREAST-FED BABIES AT RISK'

In contrast, these headlines related to problems in infant formulas and baby foods make no reference to toxins or poisons:

COVER-UP ON BABY MILK- CHEMICAL IN BABY MILK FORMULA
OPENING ANOTHER CAN OF WORMS (BABY MILK FORMULA)
NO USE CRYING OVER BABY'S MILK
DELAY OVER BABY MILK FINDINGS A "SHAMBLES"
PESTICIDES IN BABY FOOD
ALUMINUM IN BABY MILK (NO PROBLEM)
ALUMINUM IN POWDERED BABY MILK IS
"TOO HIGH"
THREE COUNTRIES BAN CHEMICALS AT
CENTRE OF BABY MILK ALERT
OUR MILK FORMULA SAFE FOR CONSUMPTION, SAY TWO COMPANIES

Popular media do not write about conflicting discourses nor muse about discursive formations. This is the domain of academic rhetoric. Nevertheless, discourses matter in a very policy-relevant way. Newspaper articles can have direct immediate policy consequences. A newspaper article in the Bangladesh Observer on toxins that pass through breastmilk and the placenta included the following: "with new information on the hazards of breastfeeding and the link
between dioxins and cancer, it may be necessary to review our position on advocating breastmilk..." (Bangladesh Observer, Sept. 13, 1989). Bangladesh has an infant mortality rate of 69.68 per 1000 live births (Dowling 2000); any decline in breastfeeding would significantly increase that mortality rate.

Endometriosis is a condition affecting many Canadian women of reproductive age. It has been a focus of women's health advocacy. The Endometriosis Association Newsletter (1992 13(4):5) contained the following:

...What about breastfeeding? One wonders whether there are testing services available as there are for water. If indeed one's breastmilk is contaminated, would it make sense to pump it and dispose of it (as hazardous waste presumably)? (And don't forget to hold that baby close when bottle feeding her/him to help create the emotional closeness of breastfeeding!) The true tragedy of toxic pollution is clear in an issue like breastfeeding . pollution could rob us of the ability to have children and to engage in something as beautiful and powerful as breastfeeding! ...

Many of the scientific reports about toxins in the breastmilk of Inuit women were also reported in newspapers and television. The news that their breastmilk contained chemicals left some Inuit women frightened and desperate. One mother decided to stop nursing in an effort to protect her new baby. After several weeks of being bottle-fed a mixture of water and Coffee-mate, the baby was hospitalized (Colborn, Dumanoski and Myers 1996:108).

Who should be translating scientific and policy language into "user friendly" language? Certainly corporations contributing to contamination or producing products affecting women's health should not be translating "science" for the general public. Journalists present their versions of scientific research with an aim to making it understandable to the public, but without careful consideration of the
importance of breastfeeding, they often raise unnecessary fears. First Nations communities are particularly sensitive to this issue since media have focussed attention on the high rates of contamination in Inuit and First Nations women's breastmilk, particularly because of their reliance on subsistence fishing and hunting. Yet in the video produced by the Indigenous Environmental Network and Greenpeace, Drumbeat for Mother Earth, the issue of pollutants was handled in a sensitive manner, highlighting not only the pollution in breastmilk, but pollution throughout the food chain and the body burden shared by all indigenous peoples.

Breastfeeding and HIV/AIDS
In some parts of the world, breastmilk is regarded as a potential source of destruction as well as nurturance. Although highly valued, breastmilk is considered vulnerable and susceptible to being tainted by spirits, the evil eye, black magic, pregnancy, or the mother's behavior, including her emotions and diet. Women know they and their breastmilk may be blamed for a child's sickness and death. A conversation in Toronto with an old shopkeeper from Eastern Europe was a powerful reminder of a public perception concerning breastfeeding and breastmilk that may be quite widespread, although seldom articulated. On seeing my WABA button promoting breastfeeding, he lectured me:

. ..in my day women could breastfeed because they ate pure food, lived in a pure environment and were pure... sexually... then they produced pure milk... now they eat bad food, live in a dirty environment and are sexually loose... so they can't produce good breastmilk and should use pure commercial milks.

This distrust of women is also reflected in the HIV/AIDS discourse. Traditional healers in Africa claim that AIDS has become epidemic because women no longer observe the old sexual customs such as fidelity and ritual bathing following intercourse. Tradition has been reinvented to control women (Schoepf 1998:112).

There is some congruence between blaming mothers for the quality of breastmilk, and contemporary concerns about a number of substances that may pass from mother to child through breastmilk, including HIV and chemical residues. Both subjects require a careful weighing of public health risks and attention to the needs of mothers. In both cases, there are significant information gaps, and widely contested claims. For example, we do not know the rate of HIV transmission in exclusively breastfed, formula fed and mixed fed babies, nor the mortality rates in infected and uninfected babies by feeding methods. In the case of chemical residues in breastmilk, the long term health consequences for infants and children are unclear. Yet both topics may be presented by the media in ways that reduce women's right to accurate information and undermine public health support for breastfeeding.
HIV/AIDS transmission from mother to child can occur through breastfeeding, although there is a greater chance of transmission during pregnancy and delivery; as with chemical contamination. It is difficult to distinguish between intrauterine, perinatal and postnatal transmission. A child breastfeeding from a woman who is HIV positive has about a 15% risk of infection through breastfeeding when neither mothers nor babies receive any anti-retroviral drug treatment. Considering communities with a 20% HIV infection rate, only three infants out of 100 are likely to be infected through breastfeeding, leaving 97 infants who would benefit from breastfeeding. For HIV positive mothers who have chosen to breastfeed or for whom breastfeeding is the only option available (particularly in resource-poor settings, where mothers cannot bottle feed safely), it is possible to promote safer breastfeeding. Women can be informed of the dangers of mixed feeding and supported to breastfeed exclusively, breast problems can be prevented or treated early, and the use of condoms can help avoid further transmission of HIV during the period of lactation (Linkages 1998). Other options include the use of heat-treated expressed breastmilk. But HIV/AIDS policy making is deeply political and not solely based on whose science is right; in some countries women are advised or told not to breastfeed. Yet some studies suggest that exclusive breastfeeding has as good an outcome as exclusive use of infant formula (Coutsoudis 1999).

The media focus great attention on the transmission of HIV through breastfeeding, particularly in developing countries. A study in Thailand on older people's knowledge about HIV/AIDS transmission confirmed that about 85% of the study population agreed "somewhat" that a person will contract the AIDS virus from breastfeeding (Im-Em, Vanlandingham, Knodel & Saengtienchai 2001:23). The headlines are by now familiar: Is Breast Always Best? Infant feeding and HIV (cover story in The Health Exchange, April, 2001). Although formula companies agreed not to exploit the HIV crisis, they have positioned themselves as the protectors of infant health by providing free infant formula for distribution to HIV positive mothers. In South Africa, Nestle offers Pelargon brand infant formula instead of the UN advised generic formula, a public relations windfall and an opportunity for increasing sales of Pelargon in a country with low rates of infant formula use. Until breastfeeding advocacy groups publicised it, Nestle claimed that Pelargon 'killed bacteria from contaminated water' (Greiner 2002:9). Where is the evidence that exclusive use of infant formula will improve the survival and health of infants born to HIV positive mothers? For complex and contested topics such as HIV/AIDS and environmental contamination, care must be taken on the way relative risks are communicated to mothers.

Speaking the Same Language
When there are contradictory messages, powerful interests can promote one message over another and further polarize complex issues into either/or options, benefiting corporations that both pollute and offer a "solution" to the problem of pollution. This further reduces the possibilities for coalition building between environmental groups, breastfeeding advocates and women's health activists. But these groups are co-travellers who must travel together to advocate for social change. Of concern here is how they can travel further
together.

What are the elements of a shared discourse among advocates for women's health, cancer prevention, environmental protection, and breastfeeding promotion? What is the relation between these constituencies? Can a shared discourse motivate action strategies for local, national and international campaigns without damaging other constituencies? Splitting these natural allies only helps toxic industries and infant formula companies - an outcome damaging to all advocacy concerns. Women's reproductive rights groups are increasingly interested in the influence of endocrine disruptors on reproductive health. This is an interest shared with environmental groups.

Environmental groups and breastfeeding support groups can work together and produce messages that further multiple objectives. For example, a feminist group in Costa Rica active in breastfeeding support, provided an example of a public service ad from Pro Eco (Programa Ecologico en Centroamerica) urging readers to take their cars to be checked for lead emissions, with visuals showing a feeding bottle filled with lead pellets. This group recognized that communicating about environmental toxins does not have to target breastmilk as a source of pollution.

The Center for Health, Environment, and Justice in Falls Church, Virginia developed from the advocacy work of mothers following the Love Canal chemical exposure. The center provides information and referral services to empower citizens and communities to prevent or eliminate exposure to hazardous chemicals. They work to translate science into accessible user-friendly knowledge through the distribution of fact sheets and guidebooks. The newspaper ads in papers such as the New York Times, March 15, 2000, called on the United States to support international efforts to negotiate a treaty to phase out POPs, with the slogan: "when it comes to children's health, America should be leading the way, not standing in the way". More detailed brochures produced by a coalition of environmental groups made specific reference to breastmilk, stressing the dangers of prenatal exposure. They list the toxins found in adipose tissue, brain, blood, cyst fluid, liver, placenta, semen, and umbilical cord blood as well as breastmilk. Inside the brochure, they address the fact that breastmilk is the most frequently tested fluid, while reiterating the claim that with the levels of dioxins found in breastmilk in most countries, it would be too contaminated to be legally sold as food in Belgium, Netherlands, and France. However, they do conclude with a paragraph that this does not mean that women should stop breastfeeding. The Center for Health, Environment, and Justice also provides a comprehensive list of studies of human contamination. The list of contaminants by country and medium shows breastmilk as the medium in nearly two-thirds of the cases. It would have been helpful to include a paragraph to explain that the predominance of breastmilk as a medium reflects the fact that it is convenient and cheap to test, not that it is "more toxic" than other substances. It is clear that breastfeeding advocacy groups could work together with these groups to create integrated messages with
regard to environmental pollution and breastfeeding. Cooperation might involve the following principles:
. acknowledge what is known about contaminants in breastmilk
. stress prenatal exposure as contributing to the body burden of all babies, not just breastfed babies
. identify the source of the pollution (chemical industries), not the source of evidence (breastmilk)
. stress the risks associated with artificial breastmilk substitutes and the risks of not breastfeeding when communicating about contaminants in breastmilk
. draw attention to alternatives to toxic products not alternatives to breastmilk
. avoid metaphors of downloading toxins from one body to another
. avoid "pump and dump" as a solution to concern about breastmilk
. make clear in media reports that any testing of breastmilk is done for bio-monitoring programs not for advising individual mothers on the condition of their breastmilk
. draw attention to contaminated milk not contaminated mothers
. suggest practical actions to reduce contaminant loads, such as limiting consumption of fatty meats, fish from polluted lakes, and dairy products

What is a balanced view of a subject so complex? One that considers what is wrong with alternative infant feeding products, one that considers social and cultural context more carefully. UNICEF presents a balanced view by calculating the number of days of life expectance lost by breastfeeding and not breastfeeding. These calculations are unlikely to be helpful to the public because they are easily misinterpreted. It will be critical to consider the purpose in communicating about toxins in breastmilk; is it to uncover previously unknown facts? to shock? to correct misunderstandings? to undercut industrial capitalism? to incite action? to change our environmental protection laws?

The Canadian Royal Commission on New Reproductive Technologies (RCNRT) recognized that exposure to harmful agents in the workplace can delay conception or reduce fertility, but stressed that there is no evidence on the effects of "more common levels of exposure". Policies in this area concentrated on removing at-risk workers; other approaches include using exposure standards, regulatory measures, education of workers, and improved health and safety legislation (1993:9). Breastfeeding was not considered in their report.

A number of breastfeeding advocacy groups, including INFACT, LaLeche League, IBFAN and WABA, have produced brochures for public education on this issue. The 1997 WABA folder, Breastfeeding: Nature's Way, argued in general terms that breastfeeding is natural, sustainable, and not polluting. It included examples of how the production of infant formula has proved vulnerable to contamination by bacteria, radioactivity, chemicals and foreign bodies, listing specific recalls and accidents. (For example in 1997, UK Milumil was withdrawn from sale after being linked with salmonella infection.) Water for mixing infant formula provides additional opportunities for contamination. It did not draw attention to breastmilk contamination, raising the issue briefly
under the header, "Breastfeeding is Preferable to any Alternative". Referring to the fact that PCBs and other pollutants have been found in some samples of cow's milk and breastmilk, it pointed out that WHO concluded that the advantages of breastmilk far outweigh any possible risks. Finally, it reassures the reader by pointing out that levels of toxins found in breastmilk fell by around 35% in Europe between 1988 and 1994.

Why have breastfeeding mothers not been at the forefront of advocacy action about pesticides, toxins, and industrial accidents? What is the difference between the Love Canal response where mothers acted because their children were at risk, and the response of breastfeeding mothers to information on pollutants in breastmilk? The experience from breastfeeding support groups suggest that mothers may be constrained by a misplaced bodily-based guilt, since contaminants are coming from mothers' bodies not from mothers' backyards. Breastfeeding mothers are unlikely to be able to address evidence on environmental toxins when they are breastfeeding; instead they will confront the full impact after the fact and react with anger, bitterness, and a feeling that it is "too late" for them. This may turn some breastfeeding mothers into activists. But the "Love Canal" scenario with breastfeeding mothers leading the advocacy fight against environmental pollution is less likely (cf Rosenberg 2000). Victims of toxic contamination often seek redress by making the problem public, organizing group actions, and mobilizing political allies (Reich 1991). It is difficult to act when users may not know the name of the pesticides they are using and consequently the active ingredients, and can only suspect who or what may be responsible for contamination. Media reports of this issue seldom stress the message that it is not mothers who are poisoning their babies, but chemical companies and identifiable industrial processes. Print media in particular take every opportunity to put breastfeeding groups on opposite sides of controversies about chemical contamination and breastfeeding, since there are real differences in emphasis and interpretation of evidence on the subject. Yet we do not have evidence about how media reports about "contaminated breastmilk" affect mothers' infant feeding decisions. We will never know who decided not to breastfeed "just to be safe".