You Could Look It Up (6 page)

As we might expect for an early work in any genre,
Erya
has its weaknesses. The definitions are short, and they often do little to explain the meanings of the words. The organization of words into thematic clusters is an interesting effort to make sense of the universe, but later eras have found it counterintuitive and hard to consult. But it was groundbreaking in its day, and it remains illuminating even long after it was superseded by other dictionaries. Moderns find
Erya
a valuable source for identifying the animals and plants, for instance, that appear in older Chinese literature. Even more important, it reveals how early Chinese writers organized their understanding of the social world, the natural world, and the divine world.

The original
Erya
led to a wide array of editions and annotations. The earliest is the
Hsiao erya
, or
Abbreviated Approaching Elegance
. There was also a
Kuang-ya
, or
Expanded Elegance
. In the third century
C.E.
came
Guangya
, or
Extension of Erya
, by Zhang Yi, with 2,345 entries. Early in the fourth century, Go Pu wrote
Erya zhu
, or
Annotations on Erya
, and around the year 1000, Xing Bing turned out
Erya shu
,
Explanations of Annotations on Erya
. As late as 1775, Chinese scholars were working on new versions of and annotations on the
Erya
, including Hao Yixing’s
Erya zhengyi
, or
Meaning Verification of Erya
, and Wang Niansun’s
Guangya Shuzheng
, or
Annotations and Textual Criticism of Guangya
.
9

The
Erya
also prompted imitations, new dictionaries that attempted the task of defining words from scratch rather than extending the original work. Probably in the early third century
C.E.
, Liu Xi wrote his
Shiming
, or
Explanation of Names
, a collection of 1,502 definitions in eight volumes and twenty-seven chapters. As Xue Shiqui explains, “its special feature is its use of phonetic explanations. It explains the meanings of words in terms of other characters of the same or similar sound, and makes
assumptions about etymology based on the sound of a character”—or, as some have described it, semantics based on puns, or connections between the sounds of words. “At times,” Xue goes on, “this method is misleading and far-fetched, but it did motivate scholars to analyze the meaning of a word from the viewpoint of the spoken language.”
10

In the year 837,
Erya
was named one of the Thirteen Confucian Classics, joining the
Book of Changes
(or
I Ching
), the
Classic of Poetry
(or
Shijing
), and the
Analects
of Confucius. As the Sinologist Endymion Wilkinson wrote, “This greatly enhanced the influence of the
Erya
on the interpretation of the classics, and no doubt also on the development of the language itself since generations of scholars memorized it.”
11

Several centuries after the greatest early dictionary of China came the greatest early dictionary of India, the
Amarakosha
—probably compiled around the fourth century
C.E.
, though some favor a later date. It had to be before the early ninth century, since other writers were referring to it by that period.
12
As with the
Erya
, little is known of the author, Amarasimha, but tradition says he was named one of the Navaratnas or “nine gems,” the most extraordinary people in the court of king Chandragupta
II
.

The
Amarakosha
, also known as the
Nâmalingânusâsana
, is the “Immortal Treasure” that organizes the entire Sanskrit vocabulary into a logical order. It has remained an important work in the ancient literature of India ever since. It is not the earliest Sanskrit dictionary—that honor belongs to the
Nighantu
, a collection of words drawn from the Vedas. We have the names of other Sanskrit lexicographers, including Katya, Brihaspati, Vyadi, Bhaguri, Amara, Mankala, Sahasanka (Vikramaditya), Mahesa, and Jina. These early lexicons tend to fall into two categories, the dictionaries of synonyms and the dictionaries of homonyms. The synonymous dictionaries, organized topically, give clusters of words that stand for related ideas. The homonymous lexicons collect multiple meanings for a single written form, and they are usually organized not thematically but by letter, sometimes the first but often the last letter of a word.
13

Amarasimha himself acknowledged that the
Amarakosha
resulted from his work of codifying, adapting, and abridging the works of
those who came before him. The organization, though, seems to be original with him. The whole work is in poetry rather than prose, around fifteen hundred lines. It is organized into three
khandas
or books—
Svargadikhanda
, treating the gods and the heavens;
Bhuvagardikhanda
, on earthly things like animals and towns; and
Samanyadikhanda
, treating common words, especially on grammar—each of which is further divided into
vargas
, or sections.

It is not an easy book to use, as this entry makes clear:

Blood. Flesh used in sacrifice. Heart. Marrow, fat. Diaphragm. The tendon forming the nape of the neck. Any tubular organ of the body, as an artery, vein, intestine, etc. Kidneys. Brain. Any bodily excretion. Entrail. Spleen. Tendon, muscle. Liver. Saliva. Concretion on the eyes. Excretion of the nose. Wax of the ear. Urine. Feces, dung. Glene. Bone. Skeleton. Spine, backbone. Skull. Rib.

Where the entries end and the interpretations begin is far from obvious, and the book is useful only for nouns. But in combining the two common methods of organizing a dictionary, the homonyms and the synonyms, and in surveying the best literature in the Sanskrit language, Amarasimha developed a reference tool that served not only his own age but every succeeding generation that has worked through the literary tradition of the Subcontinent. As late as 1808, an English author could refer to “The celebrated Amara Kosha, or Vocabulary of Sanskrit by Amara Sinha,” praising it as “by the unanimous suffrage of the learned, the best guide to the acceptations of nouns in Sanskrit.”
14
Its influence extended even further, for another nineteenth-century
reader influenced by the
Amarakosha
was Peter Mark Roget. The grouping of words in thematic clusters inspired his approach to classifying synonyms in the famous
Thesaurus
of 1852.

Erya
and
Amarakosha
were entirely independent works—there is no evidence of one influencing the other—but they are remarkably similar. Both are concerned with only a selection of the entire vocabulary of their respective languages; both are concerned with making sense of the great literary classics of their traditions; both are organized thematically, giving us a glimpse of how their authors classified the world’s knowledge. Both also inspired long traditions.
Erya
is only the first in a long line of premodern Chinese dictionaries, culminating perhaps in
Peiwen yunfu
, put together by a team with support of the emperor in 1711, which contains 700,000 words—more than any other Chinese dictionary, even those created today—arranged by rhyme by their last character. Though there are no definitions, words and phrases are shown in use, providing a contextual clue to their meanings.
15
And the
Amarakosha
led to the
Anekarthasamuchchaya
of Shashvata, the
Trikandashesha
of Purushottamadeva, the
Haravali
of Purushottama, and others—not merely useful tools in making sense of older literature, but important works of literature in their own right.
16

The West has its own major premodern dictionaries. For centuries, no Greek lexicographer rivaled the great Suidas, who lived in the late tenth century: everyone who worked with ancient Greek texts found his dictionary essential. The only problem is that Suidas never existed. The book known as “the dictionary of Suidas” was not actually a book by an author named Suidas, but a dictionary called
Suda
, a late Latin word for “fortress.” Only in the twentieth century was its origin sorted out, but for a millennium this book, the high point of Byzantine scholarship, has been indispensable for students of Greek literature. It contains thirty thousand articles—some lexical, some encyclopedic, some somewhere in between—on language, literature, and history. The information is strangely miscellaneous. Some entries on important topics are short; some entries on trivial topics are long. The entire entry for Aaron, the brother of Moses and the first high priest of the Israelites, reads, “Aaron,
a proper name.” (“A proper name” is a favorite non-entry entry: “Abdiou [i.e. Obadiah], a proper name,” “Abeiron, a proper name,” “Aberothaeus, a proper name,” “Abim, a proper name” …) The
Suda
remains valuable for its extensive quotations of Greek writers whose works have been lost—sometimes our only evidence about a Greek writer’s style is the quotations in the
Suda
.

Advertisements for modern dictionaries always boast about all the new words that have been included since the previous edition, as if the sole virtue of a dictionary is its up-to-the-minute treatment of words like
locavore
and
hashtag
. But dictionaries can be valuable precisely for being old, for giving us a glance at the way a language was organized, and a world was understood, centuries or even millennia ago.

CHAPTER
2 ½

A FRACTION OF THE TOTAL

Counting Reference Books

“Touts les gens de Lettres sont d'accord,” lexicographer Antoine Furetière wrote in
1684
, “qu'il n'y sçauroit avoir trop de Dictionnaires”—“Men of letters agree that you can't have too many dictionaries.”
1
But is there such a thing as too many?

Even in antiquity, reference books were supposed to fix the problems of information fatigue, serving as life preservers in a sea of information. In the end, though, they may just contribute to the flood. It didn't take long for even the reference books to start overwhelming their readers. The cultural commentator Ilan Stavans reports on his own explorations in the card catalog: “Today there are dictionaries of Aramaic, ballet, gerontology, hip-hop, knighthood, Napoleon's wars, proteins, Russian slang, and
TV
.”
2

What, then, is the total? Counting reference books is like counting the stars in the sky—something, incidentally, that reference books help us do. No one knows the answer, but the number is immense. Major research libraries devote large rooms to reference books, where they may number in the tens of thousands. Even that is a tiny fraction of the total. The Library of Congress Online Catalog, when asked to display all its holdings with the word
dictionary
in the title, comes back with an error message: “Your search retrieved more records than can be displayed. Only the first 10,000 will be shown.” The same thing happens in a search for
encyclopedia
.