How does the ELU community define a century? When did this century start 2001 or 2000?

Question

This discussion arose (well, for me at least) when I was confronted with one of the definitions used for defining centuries. This definition is relevant for defining community tags on specific centuries.

The two different viewpoints

The two viewpoints are very concisely set out on this Wikipedia page

I will briefly point out the differences:

View point 1: Strict usage, the first century started in 1 AD, therefore every century starts in '01 and ends at the end of '00

View point 2: Usage as defined in ISO 8601, which defines centuries as starting in '00 and ending at the end of '99

I am asking for community opinions on which definition should be followed within the ELU community. An example which would depend on this definition is the excerpt for tags relating to centuries, e.g. the 20th-century-language tag. Should this be stated as 1901-2000 or 1900-1999?

My viewpoint

I would argue to follow the ISO 8601 standard. That way, decades are associated with centuries, so the first decade of this millennium (2000-2009) is entirely in the 21-st century, rather than being split over the 20th and 21st.

I would also like to point out that ISO 8601 is not just something programmers use, to that end I am citing the Oxford Living Dictionary, which acknowledges the use of ISO 8601 in practice:

Strictly speaking, centuries run from 01 to 100, meaning that the new century begins on the first day of the year 01 (i.e. 1 January 1901, 1 January 2001, etc.).

In practice and in popular perception, however, the new century is held to begin when the significant digits in the date change, e.g. on 1 January 2000, when 1999 became 2000. Since the 1st century ran from the year 1 to the year 100, the ordinal number (i.e. second, third, fourth, etc.) used to denote the century will always be one digit higher than the corresponding cardinal digit(s). Thus, 1066 is a date in the 11th century, 1542 is a date in the 16th century, and so on

I would like to end with a quote from one of the more experienced users, because I think it is something that should be held in high regard, deciding democratically on one's own consensus:

"Due to the deliberately democratic philosophy behind Stack Exchange, a lot of the 'rules' are decided by community consensus. This is done in Meta. Even your post all rules in one place might become a 'rule' itself one day." -Lawrence

Answer 1

In the context of English, one year of resolution for a century is too fine a distinction. With history and English, we're talking about periods that don't correspond neatly to century boundaries. Is an author who started writing in 1890 but didn't get published until 1901 a 20th century author or a 19th century author? You could argue either way, because we're really interested in the environment that influenced their writing, not the precise time that the text was created.

So what good is limiting 20th-century-language to a particular 100 year range? Was the language in 1899 significantly different from the language in 1901? Is the language in 2001 significantly different from 1999? Old English is considered to roughly go from the 7th century to mid 12th century - would it be more useful to use a century tag corresponding to when my source was written or to use old-english?

I think 20th-century-language should be ditched and replaced with contemporary-english. The description of early-modern-english is "late 15th century to the mid to late 17th century". I don't know if we need to distinguish the 18th and 19th centuries from "contemporary" English, but if we do, I think we should name the time period instead of using century names.

Answer 2

According to USNO

Years of the Gregorian calendar, which is currently in use today, are counted from AD 1. Thus, the 1st century comprised the years AD 1 through AD 100. The second century began with AD 101 and continued through AD 200. By extrapolation we find that the 20th century comprises the years AD 1901-2000. Therefore, the 21st century began with 1 January 2001 and will continue through 31 December 2100.

This is one of the sources referenced by Wikipedia. There was no year zero AD. Logically, the first 365 days period in "the year of our Lord" was completed a second after 11:59:59pm, 31/12/0001 so the first 2000 years were completed a second after 11:59:59pm, 31/12/2000. In short, it is not really a "democratic" matter of interpretation or ELU policy: 1/1/2001 was objectively the first day of the 21st century, and that is what we ought to follow here...

Unless this community comes to a consensus decision to follow not that technically correct interpretation of the Gregorian calendar, but

View point 2: Usage as defined in ISO 8601, which defines centuries as starting in '00 and ending at the end of '99

An excellent suggestion by a member:

Since our @1006a has already suggested the clever compromise of saying "the 1900's" "the 2000's" etc to bypass avoidable paradox, we can easily choose to use that terminology here.

Answer 3

The twenty-first century started on 1 January 2001. 2000 was the last year of the twentieth century, just as 100 was the last year of the first century.

The hoo-hah over the major change from 1999 to 2000 was not a celebration of a new century, but rather the change from 19 to 20. The so-called "Millennium bug" was a factor in both the hoo-hah and a popular misconception that it's the first two digits which define the boundaries. It's true that ticking over from 1999 to 2000 seems a far greater change than from 2000 to 2001, but it simply means that you have entered the 2000th year; it doesn't mean that you have completed 2000 years or twenty centuries.

ISO 8601 is a standard form for data interchange. It doesn't define language. Indeed, it defines an entirely artificial Year Zero and has no specification of centuries anyway.

Answer 4

I’m sure that you understand the vagueness of, e.g. “the 14th century” and realise that nothing here is going to clarify that vagueness

As a century is, as per OED

II. A period of one hundred years.

It doesn’t matter when it starts: “In 1848 the family was one of the richest; a century later that had changed.”

In another sense, and the one you mean - OED

5.a. Each of the successive periods of one hundred years reckoned forwards or backwards in time from a conventional starting point, esp. the date traditionally accepted for the birth of Christ. Frequently with preceding ordinal numeral.

Since the traditional year of the birth of Christ is a.d. 1, the first century is traditionally regarded as a.d. 1–100, the twentieth century as 1901–2000, the first century b.c. as 100–1 b.c., etc. However, more recently a ‘century’ in this sense has been popularly taken as spanning the years in which the last two digits of each year range from 00 to 99 while any preceding digits remain the same; in this reckoning, the twentieth century is regarded as the years 1900–99 inclusive.

“the first century is traditionally regarded as a.d. 1–100” is probably better understood as “the first century runs from January 1, 1 a.d. to December 31st, 100 a.d.” In practical terms, and accepting Christ’s birth as Dec. 25th 1 a.d., we see that the birth had a retroactive effect on the “a.d.” designation of preceding months, although the “real” a.d.1 was very short indeed.

Thus “Since the traditional year of the birth of Christ is a.d. 1,” is probably better understood as “Christ’s birth signalled the start of those years labelled “a.d.”” (As Christ and Herod were allegedly alive at the same time, and as Herod died in 4BC, it is all pointless anyway.)

However, such references as “the 14th century” is not helped by (i) Christ’s birth being taken as December 25th, (ii) the Roman year starting on January 1st at the time of Christ’s birth, and (iii)years, in the English speaking world, often starting in March to coincide with the start of Spring and the agricultural year and thus the financial year – in the UK the financial year still runs from 1st April. http://www.webexhibits.org/calendars/year-history.html

The problem with “the ISO 8601 standard” is that the vast majority of earth’s population has never heard of it and they continue to refer to centuries as they see fit. It is accepted that “century” is a vague term - it is never used in contracts or where money is concerned without a clear definition that uses actual dates.

Given this universal freedom to define as one sees fit, it should also be noted that the actual definition is usually unimportant and would create controversy only in reference to people and events between XX99 and XY01. For these, a slight rewording to avoid “century” is probably in order – either that or use your own interpretation. If anyone argues, refer them to this thread.

Answer 5

Quite honestly, following the ISO 8601 will make no good at all. If you follow that, the current century (21st century) will be 2000-2099. Over a billion people already define a century as a 100-year period starting on a year ending on a one, and ending on a year ending on a zero. In this view, the 21st century will always be 2001-2100, with no exceptions.

Plus, the year 2020 had just started, and are you going to consider that we have entered a new decade already? So, the "previous" (?) decade (2010s) was 2010-2019? I've never heard anyone having an idea to define a decade like this - a ten-year period starting on a year ending on a zero, and ending on a year ending on a nine...? So, this "decade" is 2020-2029? All completely false. We are still on the 2010s (2011-2020), and the 2000s ran from 2001-2010. So the 2020s will be 2021-2030.

It's worth pointing out that there is no such thing as a "year zero". Thus the first year was year 1, not zero. And, if you define a century like how you usually would, the first century would have only 99 years. And the first decade would have only nine. Also, millenniums would progress as 1-1000, 1001-2000, 2001-3000, etc. So it was a massive, sadistically enormous issue when billions of idiots celebrated the new millennium on the year 2000, while billions of other people had done so on the year 2001 which is the valid time to celebrate.

You notice, this is a huge, serious problem. There was no "year zero" and if you persist to believe in ISO 8651, decades would last for 9-10-10-etc, centuries would last for 99-100-100-etc, and millenniums would last for 999-1000-1000 years. We need to consider every decade/century/millennium to last for 10/100/1000 years with no exceptions at all. So the ELU defines a century as how I've explained (the "strict" method), and the 21st century (the current century) will be pretty much (?) unanimously defined as 2001-2100.

---

So the current decade/century/millennium will be marked as:

• Decade: 2010s (2011-2020)

• Century: 21st century (2001-2100)

• Millennuim: 3rd millennium (2001-3000)

Answer 6

I would like to propose a new method which will happily be adopted by all communities. First the problems with the two methods, and then, following, the solution that works around both.

Problems: - It is so obvious that a century should change with the digits: 1899 goes with the 1800's and 1900 should obviously then go with the 1900's. But... - There is no year 0, causing the first year of the first century to be year 1, and then logically

Caveat: The second point was explained to all of us in 6th grade history. And we all got used to it. Or some of us got used to it and other's didn't care. Or no one got used to it, but those grade grubbing little bastards at the front of the class with their glasses and their hand-raising and their 'think-they're-so-smart'ing somehow remembered it past the test.

Solution:

CREATE A YEAR ZERO OUT OF ONE BC AND SHIFT EVERYTHING BEFORE THEN BY ONE

The implications of this proposal are profoundly simplifying.

• we'll be able to use 'century' as you suggest and everyone else will be able to label things logically: 1900 is in the 1900's.

• for all those people who would complain, having their birthdays all messed up because of the BC year change, hah, suck, you're all dead. Which is to say no one will complain because no one cares.

You're all welcome.

Next problem: 1950 is in the 20th century? That's messed up.

Answer 7

I would argue to follow the ISO 8601 standard. That way, decades are associated with centuries, so the first decade of this millennium (2000-2009) is entirely in the 21-st century, rather than being split over the 20th and 21st.

I would also like to point out that ISO 8601 is not just something programmers use, to that end I am citing the Oxford Living Dictionary, which acknowledges the use of ISO 8601 in practice:

Strictly speaking, centuries run from 01 to 100, meaning that the new century begins on the first day of the year 01 (i.e. 1 January 1901, 1 January 2001, etc.).

In practice and in popular perception, however, the new century is held to begin when the significant digits in the date change, e.g. on 1 January 2000, when 1999 became 2000. Since the 1st century ran from the year 1 to the year 100, the ordinal number (i.e. second, third, fourth, etc.) used to denote the century will always be one digit higher than the corresponding cardinal digit(s). Thus, 1066 is a date in the 11th century, 1542 is a date in the 16th century, and so on

---

Since there is some confusion as to what ISO 8601 says, let me quote paragraph 3.1.2.23:

3.1.2.23

century

time scale unit (3.1.1.7) of 100 calendar years (3.1.2.21)duration (3.1.1.8), beginning with a year whose year number is divisible without remainder by 100

EXAMPLE:

The 19th century covers the years 1800 through 1899.

Note 1 to entry: Century is also used to refer to an arbitrary duration of 100 years, however century is not used as such in this document.