Up here in the Northern Hemisphere, the spring weather’s just barely warming, but regulators in the United States are already wringing their hands over a tricksy fall brew: the contents of the COVID shot that vaccine makers are prepping for autumn, when all eligible Americans may be asked to dose up yet again (if, that is, Congress coughs up the money to actually buy the vaccines). In a recent advisory meeting convened by the FDA, Peter Marks, the director of the agency’s Center of Biologics Evaluation and Research, acknowledged the “very compressed time frame” in which experts will need to finalize the inoculation’s ingredients—probably, he said, by the end of June.

Which is, for the record, right around the corner. A big choice is looming. And whatever version of the virus that scientists select for America’s next jab is “probably going to be the wrong one,” says Allie Greaney, who studies the push and pull between viruses and the immune system at the University of Washington and the Fred Hutchinson Cancer Center.

Unavoidably, several months will separate the selection of this autumn’s vaccine and the deployment of said shot. That’s eons in coronavirus time. Half a year ago, we were all still living in Delta’s world; now a whole gaggle of Omicrons are running the show. Any decision that scientists make in June will have to involve assumptions about how SARS-CoV-2 will shape-shift in the future, which exactly no one is eager to make. “We keep getting burned,” says Adam Lauring, a virologist at the University of Michigan. Perhaps the virus will stay on its Omicron bender, making an Omicron vaccine—a favorite for the fall’s jab jubilee—sound like a no-brainer. Or perhaps by the time summer’s through, it will have moved on to a Rho, Sigma, or Chi that sproings out from somewhere totally unexpected and undermines that Omicron shot. With so many people around the world harboring some degree of immunity, the virus is being forced to continually reinvent itself, and no one knows what new costumes it might try on next.

[Read: We might not need annual COVID shots]

Our choice of fall shot, then, is inevitably going to be a gamble and a guess. But with the clock ticking down, most of the experts I’ve been talking with think an ingredient swap is wise, and probably inevitable. “We should be updating the vaccines now or yesterday,” said Jonathan Abraham, a physician and immunologist at Harvard Medical School. Modeled on the version of the virus that kick-started the crisis more than two years ago, our current crop of immunizations is still guarding against severe illness and death. But that OG variant has long since fizzled out—leaving our shots, in this one sense, frozen in the past, while the real SARS-CoV-2 continues to race ahead. A 2022 revamp might finally give our vaccines a chance to close some of that gap.

The decision that regulators make in early summer won’t just be about a boost. In the recent advisory meeting, Marks emphasized that any vaccine updates would be expected to be comprehensive, replacing old formulations as both boosters and primary-series doses; after the changeover, people who haven’t gotten their first doses—who number in the tens of millions in the U.S. alone, and would include future generations of kids—might not be able to nab an original-recipe shot. “We would not be going backwards,” Marks told the committee. “It would be too confusing and potentially dangerous to have different regimens.”

The same system shuffles the populace through a new flu-shot formulation year after year, and it usually works just fine. Those viruses have been twining themselves into the human population for centuries; host and pathogen have settled into an uneasy rhythm, with a more or less set flu season playing out in most parts of the world each year. Last year’s successful flu strains tend to give rise to this year’s, which then sire next year’s—a phenomenon scientists call “ladder-like evolution,” because of its soothing stepwise shape. To concoct the forthcoming season’s flu shot, “we do surveillance; we figure out what to be prepared for,” Lauring told me. With SARS-CoV-2, however, “the dynamics are still so wacky.” Waves of infection crest and crash in different countries every few months; the virus is still sloshing out new variants and subvariants at breakneck speed. The emergence of coronavirus iterations has also been less ladder-ish and more radial, like spokes erupting out of the center of a bicycle wheel: Alpha did not beget Delta, which did not birth Omicron.

In recent months, though, the virus appears to have taken a different tack. Since the end of 2021, nearly everything’s been coming up Omicron. From BA.1 (a.k.a. Omicron classic) to BA.2, and now the rising BA.2.12.1, BA.4, and BA.5, the last few viral successions have all occurred within the Omicron clan. So our next move might seem obvious: counter with an Omicron-centric vaccine, a switch some experts have been favoring for months. On that front, Moderna and Pfizer might soon deliver. The two vaccine makers have each been testing, among other options, bespoke BA.1 versions of their shots that they say could be ready within the next few months, just in time for a pre-winter inoculation push. “We plan to have a data readout soon,” Jerica Pitts, a spokesperson for Pfizer, wrote in an email.

[Read: Why can’t we just call BA.2 Omicron?]

By numbers alone, there is a pretty strong likelihood that more BA-whatever subvariants will come down the pike. And as a booster, especially, an Omicron shot could have clear perks, shoring up the defenses laid down by previous doses while also, ideally, pushing a new batch of immune cells to wise up to the variant’s unique and never-before-seen quirks, says Marion Pepper, an immunologist at the University of Washington. The hope is that Pfizer’s and Moderna’s data will back that notion up and show that people boosted with Omicron’s spike are better at duking it out with most of the BA fam than those who are injected with the original recipe again. But there’s also a chance that the evidence won’t bear this out. A smattering of recent studies, some in animals, hint that chasing an original-recipe shot with something Omicron-y might not push the body to develop a ton of Omicron-specific defenses, at least not at first; studied head-to-head, a BA.1 booster and an OG booster performed about the same. Pepper still has faith that a lesson on Omicron’s spike will pay dividends—the effects just might take more time to unspool. Taia Wang, an immunologist at Stanford, agrees. “Boosting with Omicron will almost certainly provide more immunity against currently circulating strains,” she told me. Currently could quickly become previously, though, if another variant elbows in. Although the virus’s evolution might look sort of, kind of, more stepwise right now, “we’ve seen the different lineages pass the baton back and forth,” Siobain Duffy, a virologist at Rutgers University, told me. “There’s absolutely nothing stopping a similar large jump in SARS-CoV-2 evolution from happening again.”

Perhaps the bigger worry is whether BA.1 will end up being a terrible teacher when deployed as an unvaccinated person’s starter shot. The variant’s bizarro-looking spike, so unlike any that came before it, is such an outlier that it may fail to show an unsavvy immune system how to recognize other morphs of SARS-CoV-2. That’s not a problem if the future of the virus stays hooked on Omicron. But should it be booted by another variant more resembling Alpha, Delta, or something else, bodies schooled on BA.1 alone might be ill-prepared. Pfizer, which is testing a triple-Omicron series in a group of previously unjabbed people, could produce data to the contrary. Absent those, a premature pivot to Omicron might bias immune systems toward the wrong track.

[Read: Should we go all in on Omicron vaccines?]

If an Omicron-only vaccine is starting to sound like a possible lose-lose situation, maybe it’s no surprise that the experts I spoke with ran the entire gamut of opinions about it. “If I could get an Omicron booster now, I definitely would,” Wang told me. Harvard’s Abraham said that he’s in the same boat. Meanwhile, John Wherry, an immunologist at the University of Pennsylvania, was one of several scientists who said that option’s a “nope”—safer, they said, to keep something with OG. The most common refrain, though, was, I’m not sure, and I’m glad I’m not the one deciding.

There could still be a quasi-compromise: a dose that includes two spike variations, maybe more, in the same shot. So-called bi- and multivalent vaccines are already in the works; both Moderna and Pfizer are slurrying together spikes from BA.1 and the OG coronavirus variant, a recipe that Moderna executives have repeatedly described as their “lead candidate for fall 2022.” That tactic could simultaneously enhance and focus the body’s defenses, says Lexi Walls, a biochemist and vaccine developer at the University of Washington. Such combo shots are the wary vaccinologist’s hedge: They might offer both a reminder of a version of the virus that most immune systems have already seen, as well as a preview of what might still be to come.

Cramming several spikes together isn’t a perfect solution. A recipe that’s half BA.1 and half OG won’t necessarily yield an immune response that splits the difference. Such a concoction also doesn’t fully solve the problems of an Omicron-only vaccine. The pesky delay between design and deployment always puts the humans behind: BA.1 may no longer be the most relevant form of Omicron to use, because it’s rapidly being ousted by speedier siblings. And a body trained on BA.1 might have some trouble tussling with some of its more irksome kin, which appear to circumvent some of the antibodies their predecessor lays down. The BA subvariants, for now, share the name Omicron, but in reality, some of them are “just as divergent as some of the variants of concern that have their own Greek letter,” says Jemma Geoghegan, a virologist at the University of Otago, in New Zealand.

Several experts, including UW’s Greaney and Michigan’s Lauring, told me that, in an ideal world, they would have liked to see BA.2’s spike slotted into the next shot instead. That’s not necessarily a reason to forgo an upgrade to BA.1, though, because that could still better familiarize bodies with other Omicron offshoots than if they were left none the wiser. Strain-vaccine mismatches happen all the time with flu shots, Geogeghan points out, and even so, those vaccines “are still really good at protecting against severe disease and death.”

Experts won’t know for sure how bivalent vaccines will fare until Moderna and Pfizer publish data from their ongoing trials. Omicron-only shots might outperform them; original-recipe boosters might still trounce them all; none of those data will have clear bearing on the next theoretical variant to rise. Abraham, for one, isn’t quite sold on the idea of a bivalent vaccine. “We don’t know what the second-best antigen would be” after Omicron’s spike, he told me; pick the wrong one, and it may just end up wasting space in shots. He’d prefer to lean into Omicron’s ongoing monopoly, he said, and model the next shot on only that. (Moderna is also trialing a Beta-OG bivalent shot—remember Beta?—that the company says is performing well, even against BA.1.)

Vaccines may not always need to lag variants this much. Geogeghan expects that the pace at which new, antibody-dodging variants sprout off the coronavirus family tree will eventually slow down. And researchers such as Walls, at UW, are working on universal vaccines that may be able to guard against a whole menagerie of coronavirus iterations—perhaps even ones that haven’t yet been detected—so that the game of variant whack-a-mole can end.

Until then, experts are working with limited options, based on limited data—and there is yet another option that may feel like the easiest of all: Do nothing, and stick with the vaccines we have. They are, after all, still performing extraordinarily well, especially when delivered in full rounds of at least three doses; it’s what’s known, and maybe, what feels safe. Among the dozen-plus experts I spoke with for this piece, there wasn’t consensus on what our next vaccine’s main ingredients should be. Still, most agreed on this: The worst thing to do would be to stay stagnant with our shots—to miss an opportunity to move our understanding forward when the virus has already gained so much ground. “We’re always playing catch-up,” says Karthik Gangavarapu, a computational biologist at UCLA. “But if we don’t do anything, we’re for sure not going to be able to win the race.”

In the early years of the 1980s, I was fooling around with a novel that explored a future in which the United States had become disunited. Part of it had turned into a theocratic dictatorship based on 17th-century New England Puritan religious tenets and jurisprudence. I set this novel in and around Harvard University—an institution that in the 1980s was renowned for its liberalism, but that had begun three centuries earlier chiefly as a training college for Puritan clergy.

In the fictional theocracy of Gilead, women had very few rights, as in 17th-century New England. The Bible was cherry-picked, with the cherries being interpreted literally. Based on the reproductive arrangements in Genesis—specifically, those of the family of Jacob—the wives of high-ranking patriarchs could have female slaves, or “handmaids,” and those wives could tell their husbands to have children by the handmaids and then claim the children as theirs.

Although I eventually completed this novel and called it The Handmaid’s Tale, I stopped writing it several times, because I considered it too far-fetched. Silly me. Theocratic dictatorships do not lie only in the distant past: There are a number of them on the planet today. What is to prevent the United States from becoming one of them?

[Read: Margaret Atwood on envy and friendship in old age]

For instance: It is now the middle of 2022, and we have just been shown a leaked opinion of the Supreme Court of the United States that would overthrow settled law of 50 years on the grounds that abortion is not mentioned in the Constitution, and is not “deeply rooted” in our “history and tradition.” True enough. The Constitution has nothing to say about women’s reproductive health. But the original document does not mention women at all.

Women were deliberately excluded from the franchise. Although one of the slogans of the Revolutionary War of 1776 was “No taxation without representation,” and government by consent of the governed was also held to be a good thing, women were not to be represented or governed by their own consent—only by proxy, through their fathers or husbands. Women could neither consent nor withhold consent, because they could not vote. That remained the case until 1920, when the Nineteenth Amendment was ratified, an amendment that many strongly opposed as being against the original Constitution. As it was.

Women were nonpersons in U.S. law for a lot longer than they have been persons. If we start overthrowing settled law using Justice Samuel Alito’s justifications, why not repeal votes for women?

Reproductive rights have been the focus of the recent fracas, but only one side of the coin has been visible: the right to abstain from giving birth. The other side of that coin is the power of the state to prevent you from reproducing. The Supreme Court’s 1927 Buck v. Bell decision held that the state may sterilize people without their consent. Although the decision was nullified by subsequent cases, and state laws that permitted large-scale sterilization have been repealed, Buck v. Bell is still on the books. This kind of eugenicist thinking was once regarded as “progressive,” and some 70,000 sterilizations—of both males and females, but mostly of females—took place in the United States. Thus a “deeply rooted” tradition is that women’s reproductive organs do not belong to the women who possess them. They belong only to the state.

Wait, you say: It’s not about the organs; it’s about the babies. Which raises some questions. Is an acorn an oak tree? Is a hen’s egg a chicken? When does a fertilized human egg become a full human being or person? “Our” traditions—let’s say those of the ancient Greeks, the Romans, the early Christians—have vacillated on this subject. At “conception”? At “heartbeat”? At “quickening?” The hard line of today’s anti-abortion activists is at “conception,” which is now supposed to be the moment at which a cluster of cells becomes “ensouled.” But any such judgment depends on a religious belief—namely, the belief in souls. Not everyone shares such a belief. But all, it appears, now risk being subjected to laws formulated by those who do. That which is a sin within a certain set of religious beliefs is to be made a crime for all.

Let’s look at the First Amendment. It reads: “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.” The writers of the Constitution, being well aware of the murderous religious wars that had torn Europe apart ever since the rise of Protestantism, wished to avoid that particular death trap. There was to be no state religion. Nor was anyone to be prevented by the state from practicing his or her chosen religion.

[Read: The challenge of Margaret Atwood]

It ought to be simple: If you believe in “ensoulment” at conception, you should not get an abortion, because to do so is a sin within your religion. If you do not so believe, you should not—under the Constitution—be bound by the religious beliefs of others. But should the Alito opinion become the newly settled law, the United States looks to be well on the way to establishing a state religion. Massachusetts had an official religion in the 17th century. In adherence to it, the Puritans hanged Quakers.

The Alito opinion purports to be based on America’s Constitution. But it relies on English jurisprudence from the 17th century, a time when a belief in witchcraft caused the death of many innocent people. The Salem witchcraft trials were trials—they had judges and juries—but they accepted “spectral evidence,” in the belief that a witch could send her double, or specter, out into the world to do mischief. Thus, if you were sound asleep in bed, with many witnesses, but someone reported you supposedly doing sinister things to a cow several miles away, you were guilty of witchcraft. You had no way of proving otherwise.

Similarly, it will be very difficult to disprove a false accusation of abortion. The mere fact of a miscarriage, or a claim by a disgruntled former partner, will easily brand you a murderer. Revenge and spite charges will proliferate, as did arraignments for witchcraft 500 years ago.

If Justice Alito wants you to be governed by the laws of the 17th century, you should take a close look at that century. Is that when you want to live?