This is part of Six Months In, a Slate series reflecting on half a year of coronavirus lockdown in America.

We’re living through the first draft of science. At the beginning of the pandemic, we did not know that much about the novel coronavirus, so we had to improvise. All that initial advice about extensive hand-washing and not wearing a mask and not touching your face was based on a heuristic—of other viruses and our norms for dealing with them—that turned out to not work very well in the context of this coronavirus. Meanwhile, researchers started working at a breakneck clip to begin studying it—“scientists are sprinting to outpace the novel coronavirus,” read a late-February headline in the Lancet. But science as a concept doesn’t change just because there’s a pandemic; it remains an iterative process, one that’s prone to mistakes, correctives, and at turns, wishful thinking and overzealous worrying. To borrow a quote from the French novelist Victor Hugo (and the title of a science blog), “Science says the first word on everything, and the last word on nothing.”

We’ve heard a lot of first words about SARS-CoV-2. Some of them turned out to be prescient, some of them turned out to not matter, some of them turned out to point us in the exact wrong direction. Together, they hint at what we’re up against in the months (years) to come, though none of them can tell us what, exactly, will happen. Holding up any single study as proof of a concept or even a tide shift in thinking can be a dangerous exercise—science after all is the sum of a vast collective of research, not a new paper. And yet, individual studies continue to make headlines, go viral, change our minds. Over the past several months, they’ve told us stories about what the virus is doing, what it could do to our loved ones, and what it could do to us, and given that context, we’ve tried to absorb it all. We’ve learned some things about the coronavirus, sure, but I hope we’ve also learned a little bit about how science works—slowly, and sometimes confusingly. Here, we’ll use 10 studies as signposts for an (incomplete) tour of what we’ve learned about SARS-CoV-2 so far, how our thinking has changed, and where we’ve gone wrong.

The Mystery Illness Gets an Identity

We will start off with not a study but a big research milestone that coincides with the beginning of COVID-19. On Jan. 10, when there was just one reported death due to, what was then, the mystery illness in Wuhan, China, a group of scientists from Fudan University in Shanghai uploaded a copy of the genome to Virological, a discussion forum. The new genome, sequenced and shared in record time, established that the virus was “SARS-like,” as it would be identified in headlines for the coming weeks, and would serve as a road map for exploring

With a limited number of people infected, it would take time to confirm that this newly sequenced virus was definitely the same thing causing the outbreak in Wuhan, Helen Branswell wrote in Stat News. She also noted that, even given a genome of the virus causing the illness—a starting point for vaccine and drugs design—experts “cautioned that this discovery does not mean the outbreak is over or the threat has passed.”

The Virus Came From What Animal?

In a scramble to figure out the virus’s origins, one group of researchers suggested it came from snakes in a paper titled “Cross‐Species Transmission of the Newly Identified Coronavirus 2019‐nCoV” published on Jan. 22 in the Journal of Medical Virology. Their argument had to do with an analysis that showed there were technical similarities between some of the DNA in the virus and that in snakes. But people who actually study coronaviruses were immediately skeptical. “Nobody has ever found a coronavirus in a snake as far as I know,” says Angela Rasmussen, a virologist at Columbia University, noting that speculation about the particular animal origins of the virus quickly turned into a racist blame game(pangolins were another early theory). “Speculation about which wild creature originally harbored the virus obscures a more fundamental source of our growing vulnerability to pandemics: the accelerating pace of habitat loss,” wrote Sonia Shah in the Nation not long after the snake theory was debunked. What animal it came from is still a mystery.

The First Inklings of Asymptomatic Spread

At the start of all this, “contagious” was closely linked to “sick.” This makes intuitive sense; it’s how we think about the flu. SARS in particular was typically accompanied by a fever and was successfully contained in 2003 in part with temperature screenings. One of the first hints that SARS-CoV-2 might spread from someone who didn’t have clear symptoms was a case of a businesswoman traveling from Shanghai, reported in the New England Journal of Medicine at the very end of January, “Transmission of 2019-nCoV Infection From an Asymptomatic Contact in Germany.” The letter noted that, though she didn’t feel ill until the plane ride home, she infected two people directly while she was in Germany. There were a couple problems with this paper: First, it actually describes what we now call pre-symptomatic transmission. Two, the researchers didn’t actually talk to the woman, who, according to a later report, said she felt warm and fatigued during her trip and took meds. We now know that transmission without super clear symptoms (at any point in the illness) is a huge driver in the pandemic—but as this case illustrates, it was a tricky concept to pin down. Public health experts were saying well into March that realizing you were sick, and staying home, was an effective option for controlling the spread.

Is the Virus Airborne?

At a March choir practice in Washington state, “a greeter offered hand sanitizer at the door, and members refrained from the usual hugs and handshakes,” according to a later report in the Los Angeles Times. Despite precautions, 53 of the 61 people in attendance showed signs of the coronavirus, as researchers cataloged in “Transmission of SARS-CoV-2 by Inhalation of Respiratory Aerosol in the Skagit Valley Chorale Superspreading Event.” “The risk of infection is modulated by ventilation conditions, occupant density, and duration of shared presence with an infectious individual,” the authors write. In other words: Don’t be lulled into a sense of security by sanitizer, hand-washing—remember all the hand-washing?—and even distancing, because this thing spreads through the air, not just on surfaces. This particular assessment has been slow to trickle up into formal health advice, but it did start a push to change how we live with the virus. Some of this is probably a coincidence—plenty of experts and much of Asia had been recommending masks for months—but shortly after the choir news broke, the Centers for Disease Control and Prevention recommended we wear face coverings.

The Whole Hydroxychloroquine Mess

A preprint of a teeny, tiny study, “Hydroxychloroquine and Azithromycin as a Treatment of COVID-19: Preliminary Results of an Open-Label Non-Randomized Clinical Trial,” was, to the horror of scientists, published at about the same time that Trump declared (on Twitter, where else?) that the drug was “one of the biggest game changers in the history of medicine.”

“We had reason to believe that it could be helpful,” says Sandra Albrecht, an epidemiologist at Columbia and a proprietor of the Facebook page Dear Pandemic. The very early data on the drug was promising, and perhaps more importantly, “a lot of people wanted to believe that it would be helpful.” (Yes, some of those people were Dr. Oz, Sean Hannity, and Laura Ingraham.)

In the end, not only did more rigorous studies fail to show a benefit, but the journal where that tiny paper was eventually published ended up investigating the research after “concerns [were] raised on social media regarding the paper’s content, the ethical approval of the study and the process the manuscript underwent to be published.” The Food and Drug Administration revoked emergency use authorization for the drug in June.

The “Study” That “Showed” Joggers Spew Enormous Plumes of Virus

It is logical that as we were grasping that the virus could travel through the air, we were trying to grasp to what extent it could travel through the air. In April, this confusion manifested in a lot of yelling at runners. A viral “study” described in the Dutch press and a post on Medium with the title “Why in Times of COVID-19 You Can Not Walk/Run/Bike Close to Each Other” was a specific catalyst here. A simulation suggested the virus could spread really, really far; the advice of the researchers was that runners keep a distance of 13 feet or so when running behind one another and that bikers leave a whopping 65 feet of space when passing someone.

Epidemiologist were skeptical, as a thorough piece in Vice laid out (was enough infectious virus possibly making it so many feet between people, when they were merely passing each other?), but for some experts, the precaution made sense—we were in the early phases of a deadly pandemic after all. Now, while basic distancing and masking are still recommended outdoors, everyone is in more or less agreement that the real issue with airborne transmission is in stuffy rooms. Yell about—but not at!—the people indoor dining.

Dogs Can Get the Coronavirus?

Aren’t we lucky that the paper “Infection of Dogs With SARS-CoV-2,” published in Nature, did not end up heralding a pandemic C plot of horrible pet deaths? June saw some reports of possible COVID-19 in pets, including an infected dog that ended up dying within a few weeks. But that dog also had cancer. “You tell people that your dog was positive, and they look at you [as if you have] ten heads,” the owner told National Geographic in July. While dogs can get infected with the coronavirus, it remains uncommon. A small good thing in a time of tragedy.

Children Can Spread the Coronavirus but to an Unknown Extent

It was not clear for a while to what extend kids and teens were involved in spreading the virus, or to what extent they got infected at all. It is, in fact, not super clear now, either. A report on 65,000 people titled “Contact Tracing During Coronavirus Disease Outbreak, South Korea, 2020,” published in Emerging Infectious Diseases in July, briefly promised some clarity. The study traced the contacts of nearly 6,000 COVID-19 patients, focusing on how the virus spread from an outside contact and into households. Infection “rates were higher for contacts of children than adults,” the authors wrote. “I fear that there has been this sense that kids just won’t get infected or don’t get infected in the same way as adults and that, therefore, they’re almost like a bubbled population,” epidemiologist Michael Osterholm, told reporter Apoorva Mandavilli at the New York Times, for a piece headlined “Older Children Spread the Coronavirus Just As Much As Adults, Large Study Finds.” But a second study, by the same group, found flaws in the first, as Mandavilli explained in a follow-up piece, underscoring the fact that researchers are still figuring out exactly how kids are involved in the pandemic.

Can You Get the Coronavirus Twice?

Throughout the pandemic, there have been anecdotal reports of people who seem to have caught COVID-19 twice. There’s been worry that people will use antibody tests as get-out-of-quarantine-free cards and then get reinfected. The first evidence that someone actually did have COVID-19 multiple times—that wasn’t, potentially, just a case of someone having a false negative test in between two positives—came at the end of August. In “COVID-19 Re-infection by a Phylogenetically Distinct SARS-Coronavirus-2 Strain Confirmed by Whole Genome Sequencing,” published in Clinical Infectious Diseases, researchers describe a man who tested positive in March and then again in August. The good news, as Rasmussen laid out in a piece for Slate: The man didn’t have symptoms the second time; it’s unclear that he could spread the virus to others; and this far into the pandemic, there’s no evidence that you can get a severe case of COVID twice. Rasmussen writes, “Until we have evidence that reinfection is something to worry about, we should instead focus on reducing the transmission risk for everyone in the community—regardless of whether they have already been infected with SARS-CoV-2.”

A Vaccine in Limbo

We’ll end with a study on pause. A vaccine from a group at Oxford University and company AstraZeneca was once estimated to be “available by September” after some early promising results in monkeys. Now, Phase 3 trials, which test safety and efficacy in humans, are on hold are on hold after “a suspected serious adverse reaction in a participant,” as Stat News reported. A statement AstraZeneca shared with the outlet noted that a  “standard review process triggered a pause to vaccination to allow review of safety data.” What the nature of this reaction is, and if it will delay the trial, is unclear. But the idea of having a vaccine within 2020—tested, distributed, ushering us back to our lives—was always a fantasy. And as experts have pointed out, this kind of inconvenient data is just the fact of how science works.