Imagine that you go to another country. You can't speak this country's language, and so you grab a travel book with some common sentences. Let's say you go to France, and you see that the way you ask for the hotel is "S'il vous plaît, où est l'hôtel?". Try to say it. If you don't know how to read French, perhaps you are trying to read it as an English sentence. That's just not good, because you might read it something like "Sil vooze plate, ow ehst low-tel", which doesn't sound like the French sentence you wanted to say.
IPA and CXS
To prevent such things from happening, an alphabet was invented to represent the most common sounds in all languages on Earth, called the International Phonetic Alphabet, or IPA. In order to represent each sound by a single symbol, the designers of the IPA have created a great many new letters — upper- and lower-case Roman letters treated separately, modified versions of Roman letters that rotate them, flip them around, or add hooks, loops, or bars to them, Greek letters, and a few symbols that are partly or completely new. IPA is a very valuable tool for linguists, but there's also a downside to all those special symbols. They aren't easy to add into electronic documents — there isn't a key on a standard keyboard for, say, ɰ — and they're hard to display, too — with all those little details, they require higher resolution than ordinary text (that is, plain ASCII), and many computer programs can't display them at all. To get around these problems, people have invented various alternatives to IPA using standard ASCII characters (though of course these alternatives sometimes need more than one character to represent a single sound). For this book, we're going to use one such alternative designed by and for conlangers, called Conlang X-SAMPA, or CXS for short. (The full name, in case you're curious, is "Conlang Extended Speech Assessment Methods Phonetic Alphabet".)
There is a chart of CXS sound names available in the appendix. Don't worry, you don't need to know all these sounds right now; the next section will introduce you to some of the more common English ones. There are a few general patterns that may help make sense out of the overall arrangement of CXS. (Don't put yourself out trying to memorize these patterns; it should be enough later that you've been exposed to them now.) Most consonants start with a letter; if any additional symbols are needed, the additional symbols usually aren't letters, and if there is an additional letter it's preceded by an underscore, like "p_d" or "B_o". Most digits are used to represent vowels. The backslash, backquote, tilde, and semicolon symbols, " \ ` ~ ; ", are only used as modifiers after something else to form a different sound — though it will turn out that, of these, backquote, tilde, and semicolon modify the sound in uniform ways, while backslash usually means "something similar to", and occasionally is used to make a new symbol with no relation at all to the old one. The underscore symbol, "_", is only used as a separator between other symbols within the representation of a single sound.
CXS is a handy single phonetic alphabet if you only get to learn one, and we won't expect you to learn any other phonetic alphabets for this book. But, that said, you'll also find IPA being used routinely by conlangers in finished documents (like slide presentations, or language descriptions made to be printable on paper); so it will probably pay off for you to become somewhat familiar with IPA as well. To help you along with that, we've set up a chart showing both IPA and CXS together, for comparison.
Distinguishing sounds from letters
You may have noticed during the Beginner lesson on sounds that there was some trouble distinguishing between the sounds that we say and the words that we write. To solve this problem linguists tend to write sounds — that are spoken — between a pair of forward-slashes (//) and letters — that are written — between angle brackets (<>).
For example, the English word ship is written <ship> and pronounced /SIp/ (in CXS, or /ʃɪp/ in IPA).