The Unsettling Ambiguity of Early Pregnancy

Binary tests offer an illusion of certainty, but pregnancy is a nonlinear experience, and confirming one is a process of slowly ruling out all other possibilities.

September 16, 2026

Illustration by Anna Parini

The earliest version of the chemical pregnancy test that we use today, the Aschheim-Zondek test, or A-Z test, was refined in Scotland in the nineteen-thirties. Doctors would inject sexually immature mice—and, later, rabbits and frogs—with a potentially pregnant woman’s urine and then test the animals’ uteruses after three days. If the mice’s uteruses showed signs of sexual maturity, it meant that the urine contained placental matter. These chemical tests developed quickly over the course of the twentieth century, eventually leading to the first rapid at-home tests, in the nineteen-eighties. The tests were refined over the years, but they were all based on the same principle: a positive result was determined by the presence of placental matter, which is indicated by the hormone human chorionic gonadotropin, or hCG.

According to at-home tests, I have been pregnant five times. I have peed endlessly on the most modern versions of these sticks and seen dozens of clear, confident blue lines (and, several times, the even more definitive “PREGNANT”). What I did not understand the first time I tested positive was that pregnancy tests draw a misleading, somewhat arbitrary, binary between a state of pregnant and not pregnant. You can be kind of pregnant, a little bit pregnant, almost pregnant, or maybe pregnant, and still test positive for pregnancy.

Levels of hCG can rise because of chemical pregnancies, in which most commonly found random genetic errors prevent the pregnancy from growing before it can show up on a scan. They will also rise as a result of ectopic pregnancies, which develop outside the uterus, and will almost never result in a birth. You might test positive if you’re going to have a miscarriage in a week, if you’re currently miscarrying, if you had a miscarriage a week ago, or if you have a fetus inside you that’s dead but not coming out. During my fourth pregnancy, last year, I was in this state of limbo for weeks: my at-home tests were unmistakably positive, but the hCG levels measured at my clinic were low enough that my doctors warned me I would probably miscarry, though they could not tell me definitively whether or exactly when that would happen.

Pregnancy tests, particularly at-home urine tests, capture a moment in time at which an embryo has implanted and is interacting with the uterus in some way. We define how we understand the beginning of pregnancy based on this snapshot, but such a freeze-frame contains limited information. Urine tests cannot quantitatively measure hCG levels over time, and they don’t include the ultrasound data required to assess a pregnancy’s viability. Most pregnancies (that is, most embryos that have implanted within a uterine lining) are not viable.

Before the advent of widespread chemical testing, the onset of pregnancy was much more ill defined. Most doctors and patients were slow to adopt the A-Z test because they felt no rush to determine pregnancy before a missed period. Many doctors objected to the concept of a home pregnancy test in the first place, partly because the tests were unreliable and partly because they wanted to protect their authority as gatekeepers.

And yet, despite these concerns, home pregnancy testing became a billion-dollar industry, and these plastic kits became a shorthand for pregnancy in popular culture—a shorthand that is apparent in the phenomenon of the online pregnancy announcement, in which women and partners and family members all react to a single peed-upon stick. Even when such videos are posted much later—after the standard twelve weeks that many health-care practitioners recommend—they are recorded fairly early in a pregnancy, often in the initial few weeks, when a couple is testing for the first time. This is far too soon to have enough information to assess viability, and that is presumably why these videos are filmed and stored until they can be used in an “official” announcement. But this genre of content obscures the fact that one data point can never tell the whole story, and that a test can detect only a pregnancy, not a baby.

In the course of a few decades, the biological anthropologist Kate Clancy writes in her new book “Pregnancy Interrupted,” out this August from Princeton University Press, “we went from ‘this test measures something, we don’t exactly know what, but it has to do with things that could grow in the uterus’ to ‘this tells you whether or not you are pregnant’ by the middle of the twentieth century. Why did we trade in our caution for a false sense of certainty?”

Clancy’s book examines the mythmaking of pregnancy: the “midnight transformation” that results in the “fairy tale ending of a healthy living child.” There are a seemingly infinite number of ways that pregnancy can be misunderstood, mislabelled, or oversimplified, whether because of deceptively binary tests or cultural narratives that tell us pregnancy is a linear experience. Clancy tries to complicate these ideas, and to capture what she calls the “physiological reality of pregnancy, which is that most of them end without a baby.”

As of this writing, I am about three weeks away from my due date. This is fairly late in a pregnancy, but I have not felt pregnant for very long at all. It was only very recently that I began thinking that I would probably actually have this baby. For most of this year, I wasn’t pregnant; I had not yet miscarried. I was pregnant for now, but who knows? After being pregnant so many times, what does that even mean? (Once, my friend, an ob-gyn, was trying to explain my chemical pregnancy to me, and I asked her, dryly, if it would be helpful to think of myself as definitely pregnant only after I had given birth.) Time moves differently when you are pregnant after multiple early losses. With every day, your odds of delivering a healthy baby get a little better, but new questions and uncertainties arise constantly that make you feel like you’re moving backward.

Part of the reason that viability is so difficult to determine comes down to arbitrary rules. For example, guidelines in the U.S. and the U.K. define one of the indicators of miscarriage as an empty gestational sac of different sizes—sixteen millimetres in the U.S. and twenty millimetres in the U.K. “This leaves a window of 4 millimetres—a huge margin when you consider the size of an embryo,” Clancy writes, “where US radiologists would diagnose miscarriage and UK radiologists would not.”

Medical practitioners typically determine early viability based on gestational age, which is normally calculated from the first day of a pregnant person’s last menstrual period. (According to this method, a fetus is “born” a couple of weeks before the sex or procedure that conceived it.) This measurement, known as the “last menstrual period” method, is highly inaccurate, because it assumes a standard twenty-eight-day cycle; women whose cycles are longer or shorter than four weeks routinely present embryos that can appear “too small” or “too big.” If the timing is off by a week or two, a healthy early pregnancy can seem as if it’s not progressing as it should.

One of the patients that Clancy interviews was given, at different times, two fetal-age estimates, which varied by four weeks. Based on the patient’s last period, her doctors estimated that she was about ten weeks pregnant, but, according to the size of the embryo, she was about six weeks pregnant. When the patient started bleeding, which can be normal in early pregnancy, her doctors didn’t know if this was a six-week pregnancy that was potentially viable or a ten-week pregnancy that she had miscarried. (After two weeks of uncertainty, she was diagnosed with a miscarriage, and continued to bleed for another six weeks.)

Four weeks is a very large margin of error. A person might be diagnosed with either a continuing pregnancy or a miscarriage, and that diagnosis would depend on where they live, which hospital they visit, and who their attending physician is. But much of pregnancy exists in what Clancy calls a “significant middle space” between viable and nonviable, which clinicians attempt to capture with the term intrauterine pregnancy of unknown viability, or I.P.U.V. The terminology can be applied to pregnancies without a yolk sac, or pregnancies with a yolk sac but no visible embryo, or pregnancies with an embryo but no detectable heartbeat—cases that might otherwise be diagnosed as miscarriages. Clancy cites one research report that found that, of about a thousand cases of I.P.U.V. gathered in the U.K., approximately one-fifth turned out to be viable pregnancies.

The category of I.P.U.V. is helpful because “unknown viability” is often a more honest description of a pregnancy’s early stages than either viable or nonviable. But the uncertainty that comes with an I.P.U.V. diagnosis also intensifies the state of limbo that so often comes with the experience of miscarriage, during which most people, Clancy writes, have “no idea whether to hope or to begin to grieve.”

This waiting is not just a part of miscarriage. All pregnancies involve significant waiting: waiting for a heartbeat; waiting for an embryo to pass multiple stages, from successful implantation to placental approval; waiting for various test results at eight weeks, twelve weeks, sixteen weeks; or waiting for nothing in particular but hoping for the best.

This is part of what makes an online pregnancy announcement so strange: it implies that the experience of “finding out” is a singular moment. Many things can happen between peeing on a stick and having a baby, and confirming a pregnancy is not so much an experience of “finding out” as it is a process of ruling out all other possibilities, gradually homing in on a most likely outcome. “Life is a movie, not a picture,” is the way Clancy’s own doctor put it.

A lot of early miscarriages happen as part of a natural process of discarding “errors.” Sometimes the body defines these errors without us even knowing: with chemical pregnancies, a person is pregnant—that is, they have successfully completed gamete fusion and embryo implantation—and they miscarry around the same time that they expect their period. If they weren’t looking for it, they would never know they were pregnant.

These errors happen frequently, and they don’t always result in an embryo’s being discarded. Errors are a natural part of our cells; aneuploidy, or an irregular set of chromosomes, can be a shifting state. “All of us are a little weird when it comes to our genes,” Clancy says. Genes are often self-correcting, and mutations can appear and disappear, over time, on their own; this means that early genetic testing, including NIPT, a common blood test for certain chromosomal anomalies, can’t always distinguish between a pregnancy that could end in a birth and a pregnancy that won’t.

What does this mean, and how should it change the way we think about pregnancy? That pregnancy loss is common and even “natural” does not necessarily signify that we should write it off as unpreventable. Pregnancy can be interrupted by biology, but it can also be interrupted by the environment; by microplastics and other endocrine-disrupting chemicals; by diseases that, as a society, we don’t protect ourselves against, because of a lack of information or interest; by the many ways that women’s bodies are surveilled through customs, social norms, and legal frameworks; and, of course, by a woman’s choice.

It’s hard to say exactly what the value of early pregnancy testing is. It is understandable to want to know as soon as possible that you might be pregnant, and to know, too, how likely it is that the pregnancy will end in a birth. On the other hand, to approach a diagnosis—of pregnant or not pregnant, viable or nonviable, normal or abnormal—with an acknowledgment of uncertainty, Clancy writes, would offer “more humility, and accuracy,” when it comes to the reality of early pregnancy.

Near the beginning of my own pregnancy, I was informed that I had an “equivocal” level of immunity to measles, meaning that it wasn’t clear whether, if I was exposed, the fetus I was carrying would be at risk of contracting the disease. Two more antibody tests, from two different clinics, gave me two different results. More research might not have given me a clear answer about my risk of contracting measles, since every clinic seems to have its own definition of immunity; it would not have explained why I had low immunity to measles in the first place, since I’ve been vaccinated; and it couldn’t really protect my fetus from the measles outbreak then escalating in the United States. More research might have insured that I was better prepared for pregnancy, but there are endless factors to prepare for, to protect against.

I used to think that, if I could just make it to the end of my pregnancy with a safe and healthy baby, I would be able to breathe easy. But I was born safe and healthy, and I am now thirty-four years old, and my own mother still doesn’t breathe easy. “Knowing more is good for choice,” Clancy writes, and it means that “you understand the consequences of that choice”—but “knowing more” also includes knowing just how much you can’t know, or can’t choose. An embryo might be in danger of everything and nothing at the same time, at all times. No amount of evidence can help us feel fully in control during the experience of pregnancy, or, for that matter, of parenthood. These are, in some sense, futureless endeavors. A more truthful picture of both, I think, would stay centered on the uncertain, fearful, thrilling present. ♦

• Parents disagreed on whether to vaccinate their children. Then the kids got sick.

Apoorva Tadepalli has contributed criticism to the New York Times, The Atlantic, The Nation, and other publications.

Source link