A guide for students and teachers

This is an article about articles, those words which seem so natural to native English speakers and totally confuse people whose native language doesn't have articles or has different rules for them. I have edited articles by Turkish professors whose English is near-perfect and still had to add the occasional “the” for the simple reason that Turkish has no articles, and there is no simple rule that tells you which nouns want “the” in front of them. Instead, learners are given lots of rules, and some of these rules don't make much sense. What follows is an attempt to make life easier for learners and teachers of English. The aim is not to give a complete set of rules for English articles, but to show a few general principles that will make sense of 99% of cases, including some which seem at first to break the rules.


Let's look at a simple example:

Yesterday I saw a dog. The dog bit me.

If we ask why the first sentence has an indefinite article (“a”) and the second has a definite article (“the”), the most common answers are “Because they already mentioned the dog,” or “Because we know which dog it is.” These explanations are true, but they aren't the whole story.

The first thing to remember is that any time you use an article – definite or indefinite – you are talking about a member (or members) of a set. If you use an indefinite article, you aren't specifying any particular member of the set (even though you might have one in mind); if you use a definite article, you are. This simple principle will guide you through the minefield of articles (even if you might have to hop around a bit) so it's worth looking at it a bit more closely.

Imagine the set of dogs. Inside this set we find everything in the universe that can reasonably be called a dog (including imaginary dogs like Scooby Doo). If I say “Yesterday I saw a dog,” I mean that yesterday I saw some member of the set of dogs. It could have been Fido or Fifi or Rex or the Hound of the Baskervilles – it doesn't matter, or at least it doesn't matter yet. When I say “The dog bit me,” I mean that one specific member of the set of dogs bit me. I might not know its name, but I'm being just as specific as when I say “Fido bit me.” Probably when I say “the dog,” I mean “the dog that I saw yesterday,” but that is because of the rules of conversation, not the rules of grammar. I could mean a completely different dog, but people who talk like that don't have many friends.

That brings us to our first clue: if you say “the X,” you should be able to answer the question “Which X?”

The dog bit me.
Which dog?
The dog that I saw yesterday.

To answer here “I don't know – just some dog,” would sound very strange. Similarly, it would be strange to say “a dog” when both you and the person you're talking to know which dog you're talking about. Let's say John and Mary have a dog called Fido. John can say “I'm taking Fido out for a walk,” or he can say “I'm taking the dog out for a walk.” In either case, Mary can be pretty sure which dog John is talking about. But if John were to say “I'm taking a dog out for a walk,” meaning Fido, then he is being logically correct but linguistically stupid, just like the person who said “The dog bit me,” meaning some other dog than the one they just mentioned.


If John can say “I'm taking the dog out for a walk,” and the dog's name is Fido, why can't he say “I'm taking the Fido for a walk”? Remember the principle: any time you use an article, you are talking about a member (or members) of a set. Fido is a member of the set of dogs (so we can call him “a dog” or “the dog”) but he is not a member of the set of Fidos. (He is a member of the set of dogs named Fido, but that is not the same as the set of Fidos, believe me.) “Fido” is just a label we stick on this particular dog, like “Robin Turner” is a name people stick on me. We don't say “the Fido” for the same reason people don't call me “the Robin Turner” or say “The Paris is the capital of the France.”

At this point, it's tempting to make a rule and say “Don't use an article with names of people, places etc.” This would be silly. We don't normally say “the Paris” or “the France”, but there are times when we can, and of course we say “the Netherlands” and “the USA”. On the other hand, if we go back to our first principle, these so-called exceptions make perfect sense.

Let's start with the hardest one and think when we could say “the Paris”. Obviously we wouldn't say “The Paris is famous for its night-life,” but we could say “The Paris of the 1890s was famous for its night-life.” What is happening here is that we have made “Paris” behave like a common noun (i.e., not a proper noun, or unique name) by making a set of Parises spread out through time: the Paris of the 1890s, the Paris of the Middle Ages, the Paris of Louis XIV and so on. This is not the same as the set of things named Paris because it doesn't include Paris, Texas or Paris, Prince of Troy. It is like the case where someone I was friends with as a teenager visits me for the first time in thirty years and says to himself “This isn't the Robin I remember.” This is unusual, but sticks to the principle. Other cases of using articles with people's names tend to be slightly humorous, like when I was teaching in a class which had five girls called Maria and kept saying things like “Not you, the Maria in the red dress,” or when Donald Trump's friends call him “the Donald” (as though his name had become a title like “the king”).

What about places that have an article all the time? They fall into four categories.

  1. Countries where the type of state is built into the name, like the United Kingdom, the Soviet Union, the United States of America, the Czech Republic and so on. Here the article applies to the type, so there is a set of kingdoms, states, republics or whatever. Remember the “Which X?” clue. Which republic? The People's Republic of China. Which union? The European Union.
  2. Regions. These often take a definite article because they are seen as members of a set of regions within a country (like the Ruhr) or, alternatively, at a level above countries (like the Balkans or the Middle East). On the other hand, they might not, either because they are seen as places in their own right, like Micronesia or Anatolia, or for historical/linguistic reasons. A good clue is that if the region is an administrative province or state (Guangdong, Texas, New South Wales) or if it was at some point an independent country (Aquitaine, Valencia) it probably won't take an article.
  3. Names given to geographical features. “The Mediterranean” is short for “the Mediterranean Sea”; “the Rockies” is short for “the Rocky Mountains”; “the Philippines” is short for “the Philippine islands”: “the Rhine” is short for “the River Rhine” (changing the word order doesn't make any difference). In all cases, the article applies to the type of geographical feature (sea, island, ocean, mountain range, lake etc.) even if it is only implied and never said (I don't think I've ever heard anyone say “the Himalaya Mountains” but the mountains are lurking there in your mind). Individual mountains normally don't take an article as they are seen as individuals; where they do, like the Matterhorn or the Old Man of Hoy, it is because the name is meaningful.
  4. Meaningful names. The Netherlands has an article because the name is a description of the country — the place got called the nether (= down there) lands because it was so low down and flat. It is an odd case because the article side of it takes it as a meaningful phrase, but the verb side takes it as a singular name: although some people say “the Netherlands are”, most say “the Netherlands is”. Similarly, its administrative capital is called the Hague as an Anglicisation of Den Haag, which just means “the harbour”. When the name and the description are the same, things get complicated, like some people's nicknames: somebody could be called Slasher or The Slasher, depending on what is going through people's heads when they choose the name (to my ears, the former sounds like a nickname given by friends, while the latter sounds more criminal, implying that he has actually slashed some people).

We can see that in most cases we can apply the set principle to tell us if the name of a place will take an article, but there are a few cases where it's not enough and you need some extra historical or linguistic knowledge. A good example is (the) Ukraine, where there is some controversy about the article, with modern Ukrainians objecting to the article on the grounds that they are an independent state, not a region. (Most style guides now advise dropping the article, but in the twentieth century it was nearly always “the Ukraine”.)

The post office, the sun and the economy

One of the first questions students of English are taught to ask is “Can you tell me how to get to the post office?” When you think about it, this is quite strange. The questioner does not have one specific post office in mind – in fact, any post office will do – so it seems to go against our principle. However, it is not so different from John taking the dog for a walk, or going to the pub. We want an article because there is a set of dogs, a set of pubs and a set of post offices. We use the definite article because there is one specific dog that John will take for a walk, one specific pub that he will go to (probably his local) and one specific post office that we want directions to – the nearest one. The questioner doesn't know which post office is nearest, but we hope that the person who answers does. If there were sensible alternatives, then we would probably ask directions to “a post office” rather than “the post office”. (Although John talks about going to the pub, you wouldn't ask directions to “the pub” because there are plenty of pubs you might want to go to – you'd just ask something like “Is there a pub round here?”) This might also explain why we say “You should go to the doctor.”

Now for a harder question: why do we say “the sun”, “the sky”, “the ground” and so on? Some textbooks give the rule that we use the definite article if there is only one of something. This is not helpful because normally if something is unique, then it just has a name – it is the fact that there can be more than one of something that makes us use articles. It doesn't work to say that there are billions of other planets which also have their own suns, skies and so on, because the English language was developed by people whose knowledge of astronomy was pretty limited. Anglo-Saxons were good at sailing, fighting and writing poems (mainly poems about sailing and/or fighting) but astronomy was not their strong point. On the other hand, even if they didn't know about other planets, they certainly could imagine other worlds. This means that they would have had some idea of a set of suns, skies, moons and so on, so that when they said “the sun” they might have been thinking “the sun of Middle Earth” (yes, that's where Tolkien got the idea from) or, in other words, “the sun of our world.” So really, it's not that different from John going to the pub.

Now I can't swear that the above explanation is correct, but it's a convenient way of thinking about it, and it means we don't need to invent another rule. It also ties in nicely with a lot of abstract nouns. Why do we say “The economy is in trouble,” but “Economics [no article] is a difficult subject”? The answer is simple: there is a set of economies (the American economy, the world economy etc.) but there is no set of economicses. Abstract nouns often indicate general qualities or concepts, and since these are not items, they do not form sets. An exception is phrases like “the economics of development”, “the politics of race” or “the love of God”. These work like “the Paris of the 1890s”; they create a set of types of economics, politics or whatever. “The economics of development” is short for “the type of economics that is concerned with development.” This gives us another clue about articles: phrases with “of” don't guarantee an article (think of Helen of Troy) but they make it more likely because they often specify members of a set.


While we're thinking about social and political issues, consider why we say “the public”, “the aristocracy”, “the working class”, “the individual” and so on. This ties in with the principle that parts of something form a set – all of these are parts of society (which doesn't take an article when used in this way). It also follows the same principle as “the economy”, since “the working class” generally means “the working class of our country”.

We also sometimes make groups by adding “the” to an adjective: the young, the old, the rich, the poor, etc. There was an outbreak of this on TV in the 1990s with shows like The Young and the Restless or The Bold and the Beautiful. Here, the article applies to the unspoken noun after the adjective, usually “people” (“the rich” means “the rich people”.)

Organisations are more complicated. Names of companies often have no article, perhaps because a company is thought of as a person (which legally, it is). Other organisations usually take a definite article: the Labour Party, the United Nations, the OECD, the IMF, the IRA, and so on. Acronyms, where the letters in an abbreviation are pronounced as a word, are an exception: we say “NATO”, not “the NATO”, “ETA”, not “the ETA”, “UNESCO”, not “the UNESCO” and so on.

Summary: the Principles

This article does not cover all uses of articles in English, but these basic principles will do fine for nearly all cases.

  1. If it is one particular member of a set, (or several particular members) use “the”.
  2. If it is a member of a set but no member in particular, use “a”;
  3. If it isn't a member of a set, don't use an article.