Universität Bielefeld

Fakultät für Linguistik und Literaturwissenschaft

SoSe 2000

Veranstaltung: Word Formation

Thomas Gräfe

The Etymology of Derivational Suffixes in the English Language

Table of Contents

I. Introduction 2

II. Suffixes, derivation and etymology: some definitions and examples 3

III.

Table of derivational suffixes 8

IV.

Interpretation and Conclusion 19

Table of Abbreviations 24

Literature 25

---

2

I. Introduction

Etymology is the study of the history of words and morphemes, the basic elements in the

formation of complex words. The processes involved in word formation (esp. compounding

and derivation) are in many cases not easy to distinguish from each other. One important

reason for this is the fact that language is not stable, it changes permanently, although the

changes are mostly long-term processes and hardly noticeable without the help of an

etymological dictionary. Morphemes can for example change their semantic properties and

their syntactic position (from free to bound morphemes, form lexicalised words to

derivational affixes or the other way around). This has effects on the patterns of word

formation, for instance on the productivity of derivational affixes in the formation of new

words. But changes are not exclusively the product of isolated developments within the

grammar or lexicon of a language. They can also be promoted by the contact of a language

with other languages (the borrowing of new words can be a result), or by social, scientific and

technological changes that exert an impact on language (for example in the creation of new

words). (Williams 1975, 3-9.)

This paper concentrates on the etymology of derivational suffixes. It tries to draw conclusions

about changes in the patterns of word formation by derivational suffixation on the basis of a

list of 70 derivational suffixes that provides some basic information concerning etymology,

semantics and patterns of formation. (Marchand 1969, 229-355.)1 The conclusions will

remain very general because it is not possible to do a detailed examination of every suffix

occurring in the list. (III, IV) Beforehand some basic definitions concerning derivation and

suffixation that are relevant on the background of further investigations in parts III and IV are

given, and it is examined on the basis of some examples how etymology is involved in these

processes of word formation (II).

1 For five basic questions dealt with in the interpretation of the list of derivational suffixes see page 7.

---

3

II.

Suffixes, derivation and etymology: some definitions and examples

The most general definition of "suffix" has been given by Marchand: "A suffix is a bound

morpheme which in a syntagma AB occupies the position B." (Marchand 1969, 209.) A more

concrete definition would tend to replace the "syntagma AB" by "the derived word AB"

consisting of the lexical root or stem A and the suffix B. But this concrete definition is not

quite correct, although it is true that suffixes occur in derived words, because derivation is not

necessarily combined with prefixation and suffixation. Some words are coined by affixless

zero derivation, for example the word

spy

. It is derived from the verb

to spy

and although it is

spelled in the same way as the verb it has changed its syntactic category (Jensen 1990, 5f.):

[spy]v -> [spy]n

A second argument why the "simple definition" does not work is that we have to distinguish

between derivational suffixes and grammatical suffixes.

Derivation is a process whereby complex words are formed by combining a lexical root or

stem with one or more than one derivational affixes (or without any affixes in the case of zero

derivation). These affixes can, but do not have to, change the syntactic category of their

respective heads. (Hansen et al. 1982, 88 ; Cannon 1987, 164.) For example the suffix

­ion

can produce a noun out of a verbal root or stem.

[generate]v -> [generat + ion]n

But the suffix

­cy

does not produce any grammatical changes. It forms nouns with nouns:

[prophet]n -> [prophe(t) ­cy]n

In a suffixal derivative the suffix is the grammatically and semantically dominant element,

which influences the grammatical category of a lexical root. But a prefix does not have any

influence on the part of speech of a word. "It joins the category the unprefixed word belongs

to." (Marchand 1969, 228.) If the prefix

in-

is added to the adjective

decent

the derived word

remains an adjective, only the meaning is shifted. (Matthews 1991, 66.)

[decent]adj -> [in + decent]adj

Not all suffixes that have the capacity of changing the grammatical category of a word are

object of derivation. If we compare the words

citizens

and

citizenry

, we recognize that only

one of them is a derived word, although their structure is similar. Both words consist of the

lexical root

citizen

and a bound morpheme. But the derivational suffix

­ry

forms a class of

words with the semantic basis "group, collectivity of..." while the grammatical suffix

-s

indicates the grammatical category plural. (Marchand 1969, 209.) Both suffixes are non-

independent elements (bound morphemes), but

­ry

can, like all derivational suffixes, change

---

4

the part of speech or the basic semantic meaning of a word, which grammatical suffixes

cannot do. (Williams 1975, 124.)

It is the intention of this paper to concentrate on (the etymology of) derivational suffixes. This

is why it is necessary to take a closer look at the process of derivation and to distinguish it

from other possibilities of word formation, especially compounding. There is a formal and a

semantical difference by which the processes of compounding and derivation can be

distinguished:

1. Compounding is the concatenation or combination of two words or more into a

morphological unit whereby both words are usually free forms like

housewife

or

penknife

. (Marchand 1969, 11.) A derived word consists of only one free form and a

derivational affix (a prefix or a suffix or both), which is a bound form. For example

the lexical root

sheep

and the suffix

­ish

make up the adjective

sheepish

.

2. Usually both elements in a compound word (whether head or non- head) are lexical:

house

as well as

wife

are lexicalised words (in this case nouns) that can exist

separately, while in

sheepish

only the noun

sheep

is a lexical element.

But these two rules do not apply for all compound and derived words. There are compounds

that consist of one lexicalised form or even no lexicalised form at all.

1. In parasynthetic compounds one element is lexicalised and the other element cannot

be used outside the compound, at least not in the same grammatical or semantical form

as it appears in the compound. For example the compound word

milk shake

consists of

two nouns.

Milk

is a lexicalised noun which has the same meaning in and outside the

compound. But the noun

shake

can only be found in a dictionary with the general

meaning "act of shaking or being shaken". It only acquires the meaning "category of

drinks" in the compound. (OALD 1989, 785, 1163.)

2. In neo classic compounds like

bibliophile

,

microscope

or

telegraph

there are no free

forms, but nevertheless the parts taken from Greek and Latin (like biblios...) are

productive. (Adams 1973, 31f.)

While there are obviously compounds without any lexicalised free forms, it is possible to find

derived words which have two, for example in derived words with a semi- suffix. A semi-

suffix is an element in word formation which "stands midway between full words and

suffixes". (Marchand 1969, 356.) Are elements such as

­like

,

-worthy

or

­monger

independent words or derivational suffixes? There are also other cases in which a distinction

between compounding and derivation is a matter of interpretation, although this may not be

obvious at first sight. In the history of language some derivational affixes had been

---

5

independent lexicalised words. The suffix

­dom

, like in

kingdom

, is an OE noun having the

meaning "judgment, authority".2 A similar case is the suffix

­hood

, like in

brotherhood

,

which has its roots in the OE noun

had

meaning "rank, condition".3 (ODEE, 282, 447.) If

dom

or

hood

were still accepted as independent lexicalised nouns, the formation of

kingdom

or

brotherhood

would have to be considered as compounding.

Kingdom

and

brotherhood

would

be endocentric compounds with

dom

and

hood

as heads,

king

and

brother

as determinants.

But neither the noun

dom

nor

hood

can be found in current English dictionaries.

Hood

is

present as a noun, but does not have the OE meaning that is needed to interpret

brotherhood

as a compound, and

­dom

is listed as a suffix that either forms nouns out of verbs and

adjectives (for example

freedom

) or is used with nouns as in

kingdom

. (OALD, 598, 358.)

Dom

and

hood

have changed from free forms to bound forms, form lexical to grammatical

status. (Adams 1973, 30; Marchand 1969, 210; Kastovsky 1992, 384.)

OE: [dom]n -> [king + dom]n

[hood]n -> [brother + hood]n compounding

head -> determinant + head

Mod. E: [king]n -> [king + dom]n

[brother]n -> [brother + hood]n derivation

head -> head + derivational suffix

These facts allow the conclusion that some English suffixes have been produced by

"reducing" free morphemes to bound morphemes. But the majority of English derivational

suffixes have always been bound morphemes so that there must be other ways in which

suffixes entered the English language and became productive in the formation of new words:

1. Reinterpretation and secretion of the second morpheme ("end- morpheme")

One possibility is the reinterpretation of the second morpheme of a word. For example

­scape

had been exclusively used in

landscape

. It was associated with the meaning "pictorial type of

view, scenary" so that it could be secreted from

landscape

and occurred as a productive

derivational suffix in other words like

cityscape

and

cloudscape

. Another well-known

example is

hamburger

:

-burger

was associated with "everything that resembles a hamburger"

and could be used to create derived words like

cheeseburger

or

porkburger

consisting of

2 The word doom still reminds us of the free form roots of the suffix ­dom, although doom has acquired a

different meaning than the original. Dom is of OS origin.

3 Had is of OS (hed) origin.

---

6

-

burger

as a derivational suffix and of a lexicalised noun specifying it. (Marchand 1969, 213f,

Hansen et al. 1982, 66.) But language is not stable: in current English dictionaries

­burger

is

explained as a suffix that forms compound nouns because

burger

itself (meaning: "informal

for hamburger or anything that is similar to a hamburger") has been lexicalised as a noun.

(OALD 1989, 151.)

Marchand 1969: [cheese]n -> [cheese + burger]n derivation

head -> head + derivational suffix

OALD 1989: [burger]n -> [cheese + burger]n compounding

head -> determinant + head

It can be concluded that it entirely depends on the interpretation of the second morpheme

("end- morpheme") of a word whether its formation has to be considered derivation or

compounding. As we have seen reinterpretation is not a one-way street:

Dom

and

hood

have

lost their status as lexicalised nouns and are no longer heads of compound words, but

derivational suffixes. In the case of

burger

the reinterpretation process has taken the opposite

way. First

­burger

was used as a derivational suffix to coin words denoting food which is

similar to a

hamburger

. Then

burger

was lexicalised as an independent noun so that words

like

cheeseburger

and so on now have to be interpreted as compounds.

2. "Imported" suffixes from other languages

The contact of English with other languages (esp. with Latin, French, or Latin via French) led

to the adoption of many loan words. With these words many derivative morphemes (pre- and

suffixes) entered the language and were on the one hand extracted from the loan words and

transferred to established words, while on the other hand the loans were used with established

derivational suffixes. (Marchand 1969, 210f.)

[eat]v -> [eat + able]adj

etymological origins: eat = ON, OS, Ofris, OHG4; -able = OF

[clear]adj -> [clear + ness]n

etymological origins: clear = OF; -ness = Ofris, OS, OHG5

The processes of creating suffixes by reinterpretation or by borrowing from a foreign

language do not exclude each other because most derivational suffixes have not been

borrowed as such, but have been constituents of loan words until they were "reinterpreted" or

newly interpreted as suffixes. (Hansen et al. 1982, 66f.) For example the suffix

­furter

is the

result of the reinterpretation of the second morpheme of the German loan word

frankfurter

4 ON, Ofris: eta, OS: etan, OHG: ezzan. (ODEE, 298.)

5 Ofris: -nesse, -nisse, OS: nessi, -nissi, OHG: -nessi, -nissi, -nassi. (ODEE, 607.)

---