Here's a question for you: Is 2019 the last year of the decade?

Yes. No. Maybe. It's complicated.

Ever since I asserted, in October, that there were "three months left in the decade," I've had readers writing in to let me know that everything I'd assumed about time, the universe, and the recordkeeping of human history is based on a lie. While I'd originally intended to write a rebuttal explaining, in the simplest terms, how a decade works (it's just 10 years! Any 10 years!), when I set out to do so, I inadvertently tumbled down a wormhole of theology, philosophy, history, mathematics, astronomy, and the wildest conspiracy theories imaginable.

This much is for sure: Whether you believe the decade ends in 2019 or 2020, everything you thought you knew about the calendar is wrong.

Let's start with the conventional wisdom: that the year 2019 caps off a 10-year period (the teens? the tens?) that started Jan. 1, 2010. This way of counting reflects how we naturally talk about time; "the 60s," for example, means 1960, 1961, 1962, 1963, 1964, 1965, 1966, 1967, 1968, and 1969 — 10 years, a.k.a. one decade.

Easy enough to understand. But then what's up with all of these people?

It all comes down to the first decade of the past two millennia, known, amusingly, as the 0s. This was the era of Emperor Augustus, the decade when Ovid completed Metamorphosis, and John the Apostle was supposedly born; it also, oddly, lasted only nine years (one of two such times in history, the other being its reflection, the 0s B.C.)

But how can so-called decades be only nine years? Because there is no year zero. The only way for the 1960s to actually span from 1960 to 1969, or for 2019 to close out the decade, is to accept that the first decade lasted only from A.D. 1 to A.D. 9.

The problem started with a monk named Dionysius Exiguus — whose name has been translated to "Dennis the Short," or, less charitably, "Denny the Runt." He was tasked by Pope John I in A.D. 525 to calculate the next hundred years of Easter dates, a complicated process with its own fancy Latin name and based on things like the movements of the moon. While going about his work, Dionysius Exiguus was also able to figure out that it had been 525 years since Jesus was born; he decided to start a new calendar with the Nativity as its anchor, beginning with year I, followed by year II, then year III, all the way up to year DXXV.

But there are no zeroes in Roman numerals. How would one count the time before Jesus' birth?

Somehow, worrying about this detail basically got put off until another monk, Bede, was stumped by the question in A.D. 731. Bede wanted to talk about the year 60 B.C., but didn't know how to reference it using Dionysius Exiguus' system. Still short a zero, Bede started working backwards, creating the system we still use today with 1 B.C., 2 B.C., and so on.

But without that zero, things became irreparably broken; there was only one year between 1 B.C. and A.D. 1, which ... isn't how numbers work. "This made no more sense than counting backward from 2001 directly to 1999, skipping 2000," The Atlantic explains. Today, secular timekeepers might prefer the initials B.C.E. and C.E., signifying "before Common Era" and "Common Era," but the underlying problem of the missing zero remains.

Hence our current counting conundrum. The absent zero has caused headaches and embarrassments the world over; in 1956, for example, a well-meaning librarian at Columbia named Roland Baughman wanted to put on a celebration for the 2,000th anniversary of Julius Caesar's assassination. But he, like many before and after him, had forgotten about that zero; as a result, his calculations were thrown off, and only 1,999 years had passed between Caesar's death in 44 B.C. and 1956, when his exhibit was to be put on.

Still, following the "year 1" logic seems pointlessly pedantic, doesn't it? When I spoke to Charles Seife, a mathematician and professor at NYU and an expert on the number zero, he agreed that "ordinarily, this would be a completely arbitrary decision. Just as it doesn't matter whether you decide that Sunday or Saturday (or Friday or Wednesday, for that matter!) is the first day of the week so long as you're consistent, it really wouldn't matter if the decade started with the year ending in 0 or 1." But because of that quirk in our accepted calendar, "we have little choice but to start counting on January 1 of 1 A.D."

Chronométrophilia, the Swiss Association for the History of Timekeeping, pointed out a similar "practical" shortcut applies to the century system. "The 20th [Century] started with 1901, ended with 2000, so that the two first digits of the last year of any century give it its name," a representative wrote me back, adding, optimistically: "Hope this will give you final and definitive certainty!"

Well, what it does tell me is that Dec. 31, 1999 wasn't actually the last day before the new millennium, despite the billions of people worldwide who celebrated it as being such; Dec. 31, 2000, which was celebrated like any other year, was the actually significant date. And yes, that also means 2019 isn't the last year of the decade we're living in; next year will be. All those best-of-the-decade lists, including my own, ought to be tossed right into the trash.

But let's face it; that's not actually how anyone counts time. When I reached out to James Nye, the chairman of the Antiquarian Horological Society, a "learned society dedicated to the study of antiquarian clocks, watches, and other forms of timekeeping" based in London, he wrote back that "the argument about the 'no year zero' may be valid, but clearly in normal parlance the Thirties did not include 1940, the Seventies did not include 1980." Nye added that "this may offend those who wish to base their calculations from 1 Jan [1 B.C.], but no one does, in regular cultural exchange. Decades are generally given names that are extremely closely linked to the principal first digits, hence Fifties, Eighties, Nineties."

After all, if you were serious about starting the annual clock at 1, then "the Sixties" actually runs from 1961-1970, which means they're not really the 60s at all but the 60s-plus-one-year-of-the-70s-and-not-counting-1960-which-is-actually-a-part-of-the-1950s-now, and that doesn't exactly roll off the tongue. And no one was ever out there buying CDs called Now That's What I Call the 80s-plus-one-year-of-the-90s; MTV never had a show called I Love the 80s, Except Specifically Not the Year 1980. Colloquially, everyone on Earth understands the concept of "the decade" as years that share the same number in their tens columns: 1980-1989 is the 80s, full stop. In other words, we have all — at some unspoken point — decided to give that original nine-year decade a pass.

Naturally, here's where it gets complicated.

Purists who argue that a nine-year decade is a logical fallacy ("a decade has to have ten years!!!!," I can hear someone furiously emailing me right now) will then also have to reconcile that the entire basis of our recordkeeping system is based on an error. Dionysius Exiguus made a mistake when he calculated the year of Jesus' birth; historians and theologians agree Jesus was more likely born in what we would understand to be 4 B.C., since King Herod, who Joseph and Mary were fleeing after they ended up at a now-famous manger, died in 3 B.C., "years before the supposed birth of Christ," writes Seife, the NYU professor, in his book Zero: The Biography of a Dangerous Idea. Although the whole world has collectively agreed to ignore this fact for the sake of consistency and not breaking our brains thinking about time anymore than we already have to, that would mean that, yes, if you are being loyal to the Christian timetable, the decade technically started in 2006 ... and ended after 2015. In other words, it's already "the 20s."

Maybe the ancient Mayans had things right. As monks were running around Europe pulling their hair out trying to figure out how to make a calendar work, the sages of Mesoamerica had long since dodged the problem by inventing zero. If the Western world had just followed their far more more logical system, the last day of this decade would instead be denoted as "13.0.7.2.6.," which I won't even bother trying to explain but will note doesn't look like anything particularly worthy of fanfare.

Science has since caught up to the Mayans. Short of adopting their system, astronomers have thrown up their hands over the B.C.-A.D. crossover and created their own systems of timekeeping. "Consider Halley's Comet, whose journey brings it past our planet approximately every 75 years," points out The Atlantic. "The comet cannot accommodate Bede's sloppy math ... by truncating its orbit and losing a year as it crosses the B.C.-A.D. interface." Several alternative calendars have since emerged to help arrange time in a way that is more accommodating of the cosmos ... and, well, basic math. "Astronomers as well as an ISO norm for dates add a year 0000 to facilitate calculations and computer programming," notes the Chronométrophilia representative.

If you ask me, Baughman, the librarian who prematurely celebrated the anniversary of Caesar's assassination, put it best to The New York Times back in 1956. "It seems to be almost a philosophical question as to when the calendar started," he told the paper of record, adding: "I'm going on with the 2,000th anniversary celebration and swelter in my error, if there is one."

And at risk of overthinking the nonsensical constructs of time anymore than absolutely necessary, I would advise everyone else to do the same.