One of the generally accepted assumptions about drinking is that women who are pregnant should not do it at all. “No amount is safe” is the mantra. That was thrown at me by CBC’s Matt Galloway (an otherwise well-informed, critical interviewer) when I mentioned drinking and pregnancy and the stigma women face in the event they are seen drinking in public while noticeably pregnant.
Him: “But not even one is safe!?”
Me: “No. That’s not true.”
I was curt because I realized when you get into this thorny issue of women drinking and pregnancy you lose your audience: “What, he thinks pregnant women should drink? What a monster!”
Note, I don’t think anyone “should” do anything (except be critical of overly simplistic research).
The idea that pregnant women shouldn’t drink has become so broadly accepted it doesn’t even need critical assessment any more. It has become a truth. So true, in fact, that in Ontario bars are required to post a sign saying “drinking alcohol during pregnancy can cause birth defects and brain damage to your baby.” (I’ll come back to the modal verb “can” in a bit).
This is a problem
In its guidance document, the Canadian Centre on Substance Abuse and Addiction includes the recommendation (say it with me) that there is “no safe level of drinking while pregnant.” Yet, oddly, in a report ostensibly designed to use the most recent research to inform guidelines, they do not include any research on pregnancy.
The explanation is twofold. First, they argue that since a mother’s drinking affects the child, and since they are looking at guidance about health harms to the individual from drinking, the research on pregnancy does not fit.
(It makes sense until you consider that the evidence on accidents and violence is about drinking’s effect on other people…)
The other reason is more peculiar: they didn’t find any robust data to support this. “None met the mathematical modelling criteria” (Final Report, pp 20-21)
So, what’s a reasonable scholar to do? Well, when you don’t have convincing data, you don’t say anything about something. Right?
What did the CCSA do? It reproduced the same statement about “No safe level” and then doubled down by expanding it to “women who are pregnant or trying to get pregnant.”
Now before we talk about evidence I need to address something that is not captured in the scientific research: stigma. As numerous researchers from more sociological and historical perspectives have argued, the drinking during pregnant issue is generative of huge stigma. Not only does it represent a process whereby an autonomous individual (a woman) sees her body becoming somehow the property of the public (therefore her actions seem to be something that people feel they are able to comment upon) but the resulting pressure is problematic. A stigma against drinking is generative of health harm. I’m not alone in saying this: the CCSA, in a moment of remarkable cognitive dissonance (I’ll explain in a moment) says this:
“Gendered social attitudes about alcohol use and reproduction mean that those who use alcohol during pregnancy experience negative, punitive and stigmatizing attitudes. In some jurisdictions, these manifest as state sanctioned interventions, barriers to treatment or child apprehension.” (Final Report, p 37)
Why is this cognitive dissonance? The CCSA is arguing that stigma about women drinking and especially pregnant women drinking is a problem. Yet that stigma is created (or at least amplified) by bodies like the CCSA making excessive recommendations about “no safe limit.”
I will deal with this research more fully in another post when I have time (note: what I’m doing is unfunded and I have a job and students I need to attend to). But I want to point out something to consider when reading any research. The “break towards harm.”
When your perspective is to find harm, and you face research that is not definitive, you tend to favour the stuff that shows harm (that’s my whole point of the “Distortions” series of posts).
A sneaky phrase you’ll see used is “evidence is mixed.” when someone says “the evidence is mixed” what it means is there are no clear conclusions one way or another. This is entirely understandable. Research, most especially on nutrition and physiology, is extensive, on going, and complex.
What happens, though, is harm-oriented researchers will break towards harm. It is often followed by a statement that seems to have missed the part about the evidence being mixed, to amplify dire outcomes. See, for example, this line from the paragraph before the one I just quoted:
“There is mixed evidence on the impact of alcohol on pregnancy and delivery outcomes, with possible increases in miscarriage and placental abnormalities.”
This is standard neo-prohibitionist language. “Possible outcomes” leads you down a certain hypothetical path, but since it’s hypothetical, it is not necessarily accurate or likely. I may explode at any moment. . .
The other word you have to keep an eye out for is “May”: “Activity X may lead to harm Y”.
It may. But it may not. “May” is a modal verb, (just like “can” in the image above). These are words which, according to Grammarly, “are used to express certain hypothetical conditions, such as advice, capability, or requests… they’re used along side a main verb to change its meaning slightly.” In other words “May cause” suggests capability. But the equally valid way of saying it is “may not cause.”
However, each has a different degree of rhetorical power. It affects the impact of what we say in different ways.
“Drinking may cause cancer” is pretty scary and suggests doing something about this thing called drinking.
“Drinking may not cause cancer” might result in a shrug, unless you’re fighting with the person who said the thing about causing cancer.
(Can is different. It implies possibility. “Drinking can cause damage.” This is absolutely true. But what is not stated is that you have to drink a lot, and repeatedly, to do that. This sign does not say “drinking one drink can cause damage” …because it cannot. Ok, let’s be scholarly: The risk to a baby of one drink is so negligible as to be non-existent.)
So to summarize: if the “evidence is mixed” then it can be taken either way. If it “is mixed, but…” you are about to be led in a direction that the researchers want you to follow, regardless of the uncertainty of the evidence.
Some day I’ll put together a more sophisticated discussion of the mixed evidence on pregnancy and drinking. But for today let’s just pay attention to the rhetorical use the language to drive fear and self doubt.
(c) 2023 Dan Malleck