… says Christy P, forwarding me another article about a study that will be said to show that co-sleeping is dangerous.
She also sent me the .pdf of the journal article (Mendoza et al., Pediatrics 2009 v. 123 pp. 533-539), which I am pretty sure I don't have the legal right to reproduce in entirety. I have only had time to skim it but didn't see any outrageous anti-co-sleeping recommendations on the part of the scientists—that kind of extrapolation is usually the domain of journalists and "consumer advocates." However, Christy P assures me she will assign some epidemiology students to "tear apart their methods" for a homework assignment. Perhaps she will favor us with some commentary as part of her lesson planning.
The glaring assumption error that I usually notice is a failure to distinguish between families who sleep together daily (and thus develop nighttime co-sleeping awareness and skills) and families who usually sleep separately (and thus never develop awareness of each other). As far as I can tell, there is some reliable data and a reasonable argument that taking your baby into your bed for one night is relatively risky—or rather, that failing to sleep together regularly creates a risk that rears its head should you and your baby happen to drop off to sleep together! But I have yet to see any data that regular co-sleeping is more dangerous than regular separate sleeping.
As an aside, that seems to me a single representative from an entire class of logical errors. The common thread among them is a sort of assumption of linearity. Generally: We are comparing two contexts of behavior, say Context A and Context B. Less neutrally, but perhaps with more clarity, we could call them the "Conventional Context" and the "Alternative Context." A context is made up of a set of practices in a certain environment. What we see people doing is observing that a given practice is more common in the alternative context than in the conventional context. Then they observe (or sometimes merely imagine!) "When the practice occurs in the conventional context, something bad happens." Then they extrapolate that the alternative context is itself harmful, because it contains so much of the practice that's proven (or imagined) harmful in the conventional context.
Like: "When people take their babies into their beds for one night, there is a higher risk of suffocation or entrapment than if everybody stayed in the sleeping environment they were 'used to' that night. Therefore, sleeping with your baby every night puts them, every night, at a high risk of suffocation or entrapment."
Homeschoolers are really familiar with this, especially in the imaginary versions. Our context is so very different from the context of an institutional school that very little extrapolation from the experiences of one can be made to the experiences of the other. I remember my mother sniffing about the homeschoolers in her district, "You see those kids outside playing at all times of the day! They can't be spending much time on schoolwork." And yet, it turns out that in the home school, you don't actually have to spend much time on schoolwork to accomplish learning. The context (even simplified down to: three kids vs. 30) makes a huge difference. Or I'll meet a teacher who says to me with a straight face, "How can you teach anything with younger kids underfoot? A baby in the classroom seems like a big distraction." Well, perhaps the fact that the home is a normal context for a baby (as well as a normal context for learning) makes it not quite so distracting.
bearing blog 