One night last year, I found myself sitting at the dining-room table, scraping the upper thighs of a baby doll with a chisel. It was quiet in the house; the kids were sleeping in their room, unaware of what I was doing to one of their favorite toys. A toe or an ear would have been easier to cut, but my daughters would have been alarmed at that. I took two small slices of the doll—small enough, I hoped, that the girls wouldn’t notice—and slid them into a wax-paper bag, which I would mail to a chemistry professor in another city the next morning.

I hated the doll from the beginning, since my daughter Vivie received her as a third-birthday present three years ago. Everything about her plastic body seemed artificial: Her skin was a peachy shade like a spray tan. Her painted-on blue eyes seemed startlingly wide, like someone slipped her an energy drink, and her mouth hung open eagerly, showing off two teeth whose color had worn from aggressive feeding. Her body was too hard to bend, though her arms and legs and neck could swivel all the way around.

The worst thing about her, though, was this: If she was “fed” a bottle of water, she’d pee it all out a few minutes later. It worked best when they dressed her in the mostly absorbent diaper she came with, but often they would give her a whole bottle and then forget to put the diaper on, leaving her lying on the rug in a puddle of her own water. The doll is from a line of Hasbro called Baby Alive; we called her The Pee Baby, or P.B. for short.

A few months after P. B. came into our lives, hoping to shift my daughters’ affections away from it and towards something else, I bought Vivie a new doll for Christmas. After combing through a catalog full of “natural” products in search of the anti-P.B., I spotted a cloth baby I thought she would like, a pudgy pink-cheeked cherub named Erik, made by a Swedish brand called Rubens Barn. When he showed up in the mail a few weeks before Christmas, it was all I could do not to carry him around the house on my hip: He’s squishy and plush, with fleece skin and curly brown hair, and he’s hefty and weighted like a real baby, with a padded bottom.

Like P.B., Erik was popular with Vivie for about three weeks after Christmas. Midway through January, he went under her bed to live with the family of stuffed animals, broken toys, and dust bunnies that had taken up residence there.

So how did I get to hacking at P.B.’s leg with a chisel? Soon after Vivie’s infatuation with Erik faded, I began wondering why, exactly, I had such a strong preference for one doll over the other and why I had hoped that my daughter would, too. I didn’t like that P.B. was plastic—but was it because it looked cheap and disposable, or because I thought that meant the doll was somehow bad for the environment? Was I being a snob, or just a consumer with good intentions? Magic Cabin, the catalog I ordered Erik from, says the company is making a “conscious effort to protect the environment.” It plants two seedlings for every tree used to create its catalogs. It’s the kind of business I felt good about supporting; Hasbro, which made P.B., doesn’t offer that same sort of feel-good consumerism.

At the simplest level, I preferred Erik over P.B. because he represented the kind of object I wanted to see in my living space. At some point my vision of happiness began to include this idea of sustainability—that much of whatever we bought or acquired needed to last, to stick around long enough to be traded or passed on. Otherwise, it qualified as junk.

But as a parent and as a consumer, it can be hard to look past words like “natural” and “organic,” to understand why even the products that carry those labels may not necessarily be as environmentally friendly as they’ve been branded. I buy wax-paper bags instead of plastic ones because it intuitively seems like the more sustainable choice; press me for details, though, and my answer would be a little fuzzy.

So after a while, I started second-guessing my basic reactions to the dolls, wondering whether my feelings were valid or a result of naïveté. To find out, I took P.B. and Erik to the office of David Tyler, a University of Oregon chemistry professor who studies the life cycles of goods and materials, looking at the sustainability of products through various lenses, like toxicity and carbon footprint. Tyler started conducting these assessments while teaching a class on sustainability at the university, and is now a common guest at “science pubs” around the country, gatherings and informal lectures in bars where people can have a drink and ask questions of a scientist. Tyler’s science pubs often focus on the environmental impact of everyday products, and I thought he might be able to tell me if I was correct in assuming that Erik, rather than P.B., was the more environmentally friendly of the two.

I took a seat in Tyler’s tidy office and laid the dolls between us. I told him about my instinctive dislike of P.B. and asked if he could explain more about how the dolls were made—and if my preference for Erik was off-base. After just a few minutes of talking, Tyler admitted that he wasn’t going to tell me which doll was more environmentally friendly—or, to be more specific, that he couldn’t. He didn’t know. The answer was far more complicated than “cloth good, plastic bad.”

For one thing, he said, it wasn’t totally clear what each doll is made of. Looking at P.B., Tyler suggested that it might be made of polyvinyl chloride, or PVC—and if that were the case, he told me, then P.B. was likely far more toxic: Plasticizers, the additives that give PVC its flexibility, are endocrine disrupters that have been linked to cancerous tumors, birth defects, and other developmental disorders. These additives could make their way from the doll and into a child’s bloodstream if, for example, a kid put one of its limbs in her mouth. That’s not to mention the environmental cost of manufacturing PVC, which generates air and water pollution near the factories where it’s produced.

Erik, on the other hand, is made of polyester—the same material as Coke bottles. So despite my intention of buying a more natural doll for my daughter, the one I got is just another plastic. Some scientists argue that polyester, when made from recycled products, is greener than cotton, which uses more water and can require the extensive use of pesticides to grow. On the other hand, petroleum, where polyester comes from, isn’t a renewable resource.

“It kind of depends on what environmental problem you care about,” Tyler told me. “Is global warming more important than toxicity? The global-warming impact of most things made out of plastic is far smaller than, say, something made out of cotton.”

He says one of the things he’s learned at science pubs is that people are generally more worried about immediate impacts than long-range ones. “Say your kid chews on this doll, and it leaches out some of those plasticizers or PVC or whatever. That would be a much more serious concern than … some kind of far-off global-warming disaster,” he said. “Even though the end result is the same. I’ll either die or get sick. Toxicity issues seem to outweigh water use and global-warming potential. It makes sense. It’s more immediate.”

Other things to consider, Tyler said, included Erik’s heft and his washability. He says he’s learned that an object’s environmental impact is often proportionate to its size. “This doll is pretty heavy,” he said, lifting the pudgy Erik off the desk. “Just going by that general rule, there’s more stuff. Because there’s more stuff, there’s more impact.” Erik is also a cloth doll, he added, which means he’ll get washed several times over his lifetime, adding to his footprint. P.B. is heartier, almost impossible to stain.

Really understanding the environmental impact of each of the dolls, Tyler told me, could take months: I’d need to study not just what the dolls are made of, but their manufacturing processes, taking into account labor practices and economies of scale (Erik was made in small batches, each doll individually stitched by workers in China).

It was exasperating, I said, to realize how little I really understand about what I buy, despite my best efforts to be conscious of my choices as a consumer. But Tyler doesn’t spend time agonizing over purchases, he told me. Instead, he just tries to consume less, and to be mindful of the use he gets from what he buys.

“Look, your daughter gets a lot of enjoyment out of this doll, and if that makes her happy and she turns out to be a better person because she had this comfort as a child…” he said. He compared the doll to owning a dog—which, between pet-food production and packaging, waste disposal, and all those miles logged driving to and from the dog park, can have a surprisingly large environmental impact. “People always get really upset with me and say, ‘Hey, I get a lot of enjoyment out of this,’” he tells me. “I’m not saying, ‘Don’t have pets.’ I’m just saying, ‘Here’s what we know about the carbon footprint per dog.’”

Later that day, alarmed by Tyler’s uncertainty about P.B.’s material, I talked online with a customer-service representative from Hasbro who told me that Hasbro products don’t contain PVC. But when I asked what the doll was made of, she said she didn’t have a list of ingredients. She gave me a link to a video about the company’s social responsibility, then disconnected.

I clicked on the video, which announced that Hasbro had removed PVC from its packaging in 2013. But it didn’t say anything about PVC in the products themselves.

A week or so later, on Tyler’s advice, I emailed a second professor, Skip Rochefort, a polymer expert from Oregon State University, who said he could run samples of the doll through a Differential Scanning Calorimeter, a tool that scientists use to do forensic analysis. After shipping the chunks of P.B. in their wax-paper bag, I emailed Rubens Barn to ask what kind of polyester Erik was made of.

The Hasbro customer-service rep had been right. According to Rochefort’s analysis, P.B. is actually made from a kind of elastomer (a rubber-like material). Rochefort assured me that my daughter was “safe even if she chews on the doll, or actually swallows a piece.” I didn’t hear back from Rubens Barn.

I have to admit, a small part of me was disappointed. Which is an absurd thing, really, given that PVC would have meant bad things in my kid’s bloodstream and bad things for the environment. But it would have confirmed my knee-jerk reaction to P.B., and I would have been able to declare a clear winner of the two dolls. I want these sorts of certainties to help guide my way.

“Plastics are so complex that it is no wonder that consumers have trouble telling one from another,” Rochefort said to me in an email. Indeed. If I have to mail shavings of my child’s doll’s thighs to a lab just to find out what it’s made of, it doesn’t make me feel secure about buying much of anything.

I found P.B. last spring, legless, in my daughters’ sandbox. (After I had chiseled her leg, it became loose and Vivie pulled it off. Then, for symmetry I suppose, she pulled off the other one, too.) I was in a decluttering mood, and that’s what she had become—clutter. I threw her in the garbage. Perhaps she could have been recycled in some way, but I’d made up my mind about P.B. a long time ago.