Key:
/ / Encloses a phonemic transcription
[ ] Encloses a phonetic transcription in the International Phonetic Alphabet (X-SAMPA transcription)
< > Encloses an orthographic transcription
* Marks a reconstructed earlier form
** Marks an unattested grammatical form
--> "Becomes" in historical linguistics
Language evolution is a fascinating phenomenon, and one that the young science of linguistics is only beginning to understand. No matter what, languages change and modify themselves; sounds merge together, other sounds split, the grammar changes, ambiguities are eliminated, new ambiguities are formed, things simplify, other things complicate, all going along with the inevitable change of the society the language was used in. The English language can now be considered an international language, and is a common lingua franca for international communications.
This is made all the more remarkable when we consider the humble beginnings of English. A crude and rough tongue, spoken in a time dubbed the "Dark Ages", due to lack of literary information. As much as I would like to discuss at length many of the finer points of this language, and its transition to the popular idiom we now speak, I am restricted by both time and space. This dissertation is concerned with the changes in nominal inflection morphology from Old English to Modern English, in particular the loss of case and grammatical gender. In it I will describe the changes and their effect on our language, and I also present the intriguing argument that Modern English is completely lacking in morphological noun cases. When I say "Old English" I mean specifically the Early West Saxon vernacular spoken by King Alfred and his contemporaries c. ~900AD; and by "Modern English" I refer to the dialect spoken by the Scottish populace c. ~2000AD.
Earlier I mentioned that languages change as the society changes; and since societal change (of any kind) is inevitable, so too is language change. Hence, when society undergoes massive change, so too does language. This is what happened to English. In 1066, William the Conqueror defeated King Harold at the Battle of Hastings; the subsequent societal change saw many things happen to the language. While the number of Normans actually arriving was not great (some estimates place the initial number at around 50,000), French became the official language of state. With Latin the language of the church, English was the commoner's vernacular, the "vulgar speech" spoken by the poor. Elements of French vocabulary began to seep into the English tongue through a variety of conduits, including the courts, literature, and colloquial speech of noblemen (wishing to imitate their new rulers).
When the Normans met Old English, they brought with them many Latinate spelling conventions. For example Old English uses the edh and thorn symbols (<ð> and <þ>), to represent the interdental fricative /T/). However, the Normans were unfamiliar with such symbols (<þ> especially, being derived from Fuþark runes), and replaced them with the digraph <th>. The ligature <æ> was likewise replaced with a simple <a>, as the phonemic distinction between these vowels began to disappear. However, the Normans didn't just remove letters, they gave us some new ones too. For example, in Old English <c> was always pronounced velar, as in [k]. However, in Old French, <c> was pronounced as [s] before the front vowels /e/ and /i/. In order to stop the Normans pronouncing everything incorrectly, the letter <k> was introduced to calm the confusion. The letters <v> and <z> were introduced, which helped as Old English's fricative allophony began to disappear. And finally, the Latin <qu> replaced Old English <cw>, for example cwēn becomes Modern English "queen".
Many laypeople are under the misconception that Modern English, like French and Italian, is a language derived from Latin. This is untrue. Old English, the ancestor of Modern English, is a Western Germanic language, descendant of Proto-Germanic, the mother tongue of all the modern Germanic Languages (e.g. German, Swedish, Afrikaans). This language was spoken around 2,000 years ago, and we have no written records of it. However, historical linguists using the comparative method have been reasonably successful in reconstructing it.
However, this does not mean that Old English bears no relation to Latin - both Proto-Germanic and Latin share the same roots. For example, compare the Latin and Proto-Germanic words for "mother"; Latin mater [mater], Proto-Germanic *mōðer [>mO:DEr]. They are part of a massive family of languages commonly called the Indo-European group. This family encompasses the majority of modern European languages, a great many Indian languages, and some Middle Eastern languages. Hence while Latin and Modern English do share ultimately common roots, they're not directly related. Modern English does have many words ultimately derived from Latin, via the many French words brought in during the Norman occupation.
Now that I have explained the social and linguistic background of Old English, I shall move on to present an account of Old English's case system, and how it has changed. I will show the effects of these changes, and argue that Modern English has no morphological noun cases.
Old English had a case system for its nouns, like Latin and many other languages do. The four cases it possessed were: nominative; accusative; genitive; and dative. Nominative case was used for the subject of clauses; accusative for the direct, and sometimes indirect, object; genitive was used to represent possession, familial relationships, control of, and a variety of other uses, many covered by the Modern English preposition "of"; and dative was used for the indirect object, and also to represent various relationships, including familial, locational, instrumental. In truth, the dative was a jumble of many other cases merged into one from Proto-Germanic. Remnants of the instrumental case survived in the masculine and neuter demonstrative pronouns, þy and þon.
Since then, English has lost its cases. In terms of nominal morphology, there are no cases. This may seem a bold claim, and I shall soon address it. The pronouns, however, still show case; there is a nominative / accusative distinction ("I" versus "me" and "he" versus "him", etc): much like Old English and its instrumental pronouns.
The 3rd person accusative forms of Modern English pronouns are derived from the Old English 3rd person datives; him [hIm], him [hIm], and hire [hIrE] for masculine, neuter and feminine respectively. This is because in every person save third, the accusative form was the same as the dative form (mē, unc, and ūs in 1st person; þē, inc, and ēow in 2nd). This ambiguity, which is reflected in some noun declensions, caused the dative forms to mimic the accusative forms: an example of sound change by analogy. However, the Modern English 3rd person plurals ("they", "them", and "their") do not come from their Old English forms, hīe, hira, heora, or him. The words are too different, and the sound changes required to justify them fly in the face of what we already know. Where do they come from? Professor Nils Johannesson from Stockholm University explains:
The loss of a morphological case system in English runs an interesting parallel with other modern-day European languages. Modern German does not have such a case system, although it used to. Similarly, Vulgar Latin lost some of its cases in the transition to the modern Romance languages. Indeed, it can be observed that the majority of West Indo-European languages have undergone some loss of case. This indicates that the loss of this system is not limited to English, and linguistic collision is not the sole cause of it. This general diminishing of case in Europe can be explained when we consider the usage of cases. By their very nature, cases are imprecise and vague in their meanings - they usually indicate a number of relationships, only one of which is intended. Usually, context suffices to distinguish the difference: but when context fails, prepositions can be employed in addition to case in order to clear up any confusion. At this point the case may feel superfluous, and will cease to be used.
Europe is also home to some non-Indo European languages, however. Modern Hungarian has a case system with over fifteen cases – and its ancestor, Proto-Uralic, had only around seven. Other Uralic languages have been gaining cases too – and Basque, the famous European isolate, has a healthy case system that (as far as we know) has neither diminished nor increased over the years. Evidently, we are missing part of the puzzle here.
In Indo European, nominal inflections fused onto the noun root in a complex manner, affecting the sounds of the root. When the language split and underwent sound changes, including the loss of final sounds, this led to a tumultuous series of events that culminated in many cases merging, or even being lost. This happened to many Indo European languages, particularly the Western ones. Nowhere can it be seen more prominently than in English, which now has no morphological nominal cases whatsoever. The loss of case was more extensive in English because of the linguistic collision and the merging of inflections.
The loss of morphological case means Modern English has to resort to purely syntactic case, using prepositions and word order to indicate the role each word plays. An example of this is the sentence Se mann lufað sēo giefe. Literally, it means, "The man loves the gift." If we change around the word order, so that it becomes Sēo giefe lufað se mann, it still means, "The man loves the gift". If we wish it to mean, "The gift loves the man", strange as it may seem, we have to effect two changes to it; we must make the word giefe nominative case, and we must make the word mann accusative. Mann is the same in nominative as it is in accusative, so we can leave that. The nominative form of giefe is giefu. The verb stays unchanged, as the number and case of the subject remains the same. So, our new sentence reads Sēo giefu lufað se mann. We can alter the order of the words, Lufað sēo giefu se mann, Se mann lufað sēo giefu, and it still carries the same meaning. In Modern English, word order and word order alone determines if a noun is the object or subject of a sentence.
Despite this, Modern English nouns have two main inflections (beside the plethora of derivational morphology affixes: but they are beyond the scope of this essay) – and these are the plural marker and the possessive marker, both usually taking the form of "s" (with phonemic realisations of /s/, /z/, and /Iz/, depending on environment). The possessive marker is, as one would expect, the remains of the genitive case, indicating only ownership rather than the full set of meanings the original genitive granted. Many times earlier have I made the perhaps audacious claim that this is not a case. It is my opinion that it is a clitic[1], not a case, for reasons I will now outline.
Evidence that it is a clitic and not a case exists in the fact that "s" can be fixed onto the end of entire noun phrases, not just nouns. For example, "the boy in the red coat's book." It is not the coat's book; it's the boy's book. In highly formal writing this displacement of relationship would not be allowed, yet despite this, it is not uncommon to see this construction in official documents and literature.
As a clitic, it has less variety of meaning than were it a full case. We can say "The city London", but not **"London's city". (Note that the phrase "London town" used to be quite common, however it is idiomatic and has since fallen into disuse.) Yet both of these constructions are acceptable in Old English (however the first would seem contrived to a native speaker[2]): Se burgstede of Lundene and Lundenes burgstede. Another example is "Hamish of clan MacDonald", but not **"clan MacDonald's Hamish". Again, both versions are acceptable in Old English.
Support against my argument of course exists. If "s" were really a clitic, it would suffix onto words regardless of syntactic category. As G Sandi has observed,
It is possible to construct a table showing the possible forms of "boy":
| Singular | Plural | |
| Unmarked | boy | boys |
| Possessive | boy's | boys' |
To counter his argument, I reproduce the table, but with phonemic characters to show the pronunciation of the words.
| Singular | Plural | |
| Unmarked | boI | boIz |
| Possessive | boIz | boIzIz |
Bearing in mind that English orthography is often illogical, this is "boys'" being pronounced as if it were **"boys's". This is a recent innovation, and began when proper nouns ending in /s/ or /z/ were made possessive, e.g. "James's" /dZeImzIz/, in analogy with common nouns, e.g. "houses" /haUzIz/. It is not present in all dialects though, and certainly is not considered "Standard English". Nevertheless, it exists in many American accents (notably Californian), and also in my native Dumfriesshire accent. I have also heard this phenomenon in the idiolects of Glaswegians, where the ending is often centralised to /@z/.
To further consolidate my point, I present this table with the Old English strong masculine noun bāt, "boat", fully conjugated into each of its forms. Following the table is a replica, with corresponding Modern English attempts at translating the meaning each case held (the plural row has been omitted for clarity).
| Nominative | Accusative | Genitive | Dative | |
| Singular | bāt | bāt | bātes | bāte |
| Plural | bātas | bātas | bāta | bātum |
| Nominative | Accusative | Genitive | Dative | |
| Singular | boat | boat | boat | to the boat, of the boat, for the boat, from the boat, with the boat, in the boat, etc |
So how did this affect the way we speak and write? The major benefit of loss of case was removing ambiguity [4]. For example, the noun ēare, meaning "ear", has only two singular forms: ēare and ēaran. These two forms account for four cases! Also, ēaran, the singular genitive and dative, is also the form of the plural nominative and accusative. In all Old English nouns, the plural nominative and accusative are indistinct. The majority of distinctions between noun cases were different vowels on the end of words, for example scipu and scipa. This is helped in no way by the tendency of Old English (and Modern English) to make any unstressed vowel into a schwa, further blurring these case endings. Given this evidence, it seems only inevitable that English lost its cases.
But not all Modern English nouns pluralise neatly with "s". Some undergo a vowel shift, like "foot" / "feet"; and some have different inflections, like "ox" / "oxen". The simple reason is that these words are relics from Old English that have refused to conform to the rest of the inflectional standardisation. The more complex reason is that the ones undergoing vowel shift were previously masculine or feminine nouns affected by the Old English i-mutation. In the mutation, the short low front vowels (/e/ and /æ/) are raised (to /i/ and /e/), and the back vowels (both long and short /a/, /o/, and /u/) are fronted (to /æ/, /e/ and /y/). (Strangely enough, /o/ does not retain its roundedness, becoming /e/ and not /œ/ as one would expect.) This is an example of umlaut, where the vowels front in response to an /i/ in the following syllable. (The phonemes that cause the i-mutation are from Proto-Germanic, and are subsequently lost. Hence, the i-mutation occurs seemingly randomly, in both Old and Modern English.) Thus words like tōþ, "tooth", become tēþ, "teeth", in the plural. It is interesting to note that the majority of the feminine nouns became regular (e.g. bōc --> "book", but bēc --> "books", not **"beek" as one would expect if analogy was not such a strong force): while more of the masculine nouns kept their older forms (e.g. mann --> "man", menn --> "men"). As for the irregularity in "oxen," the cause of this is a movement, towards the end of Old English and the beginning of Middle English, to regularly form the plural in "-en" (much like Modern Dutch). The Modern English remnants of this are few, "children" and "oxen" being the only common examples. However, this is reflected in Southern English dialects (where this movement originated), where the word "housen" used to be a common plural for "house". It must be recognised that this non-acceptance of regularity on the part of these nouns is no bad thing. For example, were I to say "man's", the listener would be aware that I am speaking in the possessive, not the plural. Such irregularity is useful in eliminating ambiguity.
On the subject of plurals, the Indo-European proto-language possessed three number forms: singular, dual, and plural. This is reflected in Old English by the dual forms of the 1st and 2nd person pronouns, e.g. wit means approximately "we two". These pronouns are no longer with us in Modern English. But are all traces of the dual form gone? No, they are present in adjectives like "both" and "either" – with their plural counterparts "all" and "any".
Much like most other West Germanic languages, Old English possesses noun gender. This is not a sex distinction; it's an arbitrary classification system, which splits nouns into one of three categories: masculine, feminine, or neuter. Each gender has its "strong" nouns and its "weak" nouns, with various sets of noun endings. There are likewise masculine, feminine and neuter pronouns, used depending on the gender of the noun to which they refer. Adjectives must be inflected to "agree" with the gender (and case) of the noun that they are modifying. The definite article (Modern English "the") had three distinct forms (per case and number) dependent on the gender of the noun it was defining. A particular point of interest is that masculine and neuter inflections are more similar to each other than they are to the feminine inflections. For example, in "strong" nouns the masculine and neuter genitive is formed by suffixing -es, whereas the feminine genitive is formed by suffixing -e. Another example is that the 3rd person genitives are his for both masculine and neuter, and hire for feminine. This carried on to Middle English, and is often reflected in Chaucer's writing, where he uses "his" where speakers of Modern English would say "its".
The Ormulum, mentioned above by Professor Johannesson, is a metrical harmony of the Gospels written in the East Midland dialect around the twelfth century. Contained within its poetic verses is a language that uses gender entirely dependently on meaning. Instead of being a mere grammatical complication, it had become a practical method of expression. Soon, other dialects would follow in its path. The cause for such a momentous change in just two centuries is associated with another change, which is present in the Ormulum – the loss of inflection. Adjectives had ceased to mark case or gender through suffixes, and the definite article þe was used indeclinably, as in Modern English. On line 3266 we see the phrase þe laferrd crist, meaning "the lord Christ". Its form in 10th-century Old English would be roughly se hlāford Crīst, the word hlāford or laferrd being masculine. And later in line 3277, the phrase þe boc, "the book", is used. However, boc is a feminine noun – in 10th-century Old English one would say sēo bōc. And here we are, only a few hundred years later, using þe almost indiscriminately. Thus, the gender of a noun had no effect on a sentence; save influencing which pronoun it would take. And so, grammatical gender evolved into semantic sex, as we now know in Modern English.
Undoubtedly the loss of gender has simplified our nominals somewhat. However, it has had the unfortunate side effect of introducing strange quirks into their usage. Some nouns take articles in some situations; some do not in the same situation. Some nouns, when coupled with a certain verb, will take a certain preposition, while another takes another.
Having looked at the majority of changes from Old English nouns to Modern English nouns, we can see that much has changed over the past millennium. We no longer have cases, genders, or classes of "strong" and "weak" nouns. These things have fallen away to a past age. But instead, we have greater syntactic complexity, varieties of interesting exceptions to the usual rules, and a much simpler pronoun system.
In summary, many things happen to the nominal inflectional morphology. Case is lost, gender is lost, some exciting new irregularities are formed, and the whole system is simplified. I have shown Old English to be a Western Germanic language, and the influence from French has lead to many Latinate spelling styles, such as <c>. French influence has also left many new words in our vocabulary. The loss of case resulted in Modern English having to opt for syntactic methods, such as word order and prepositions, more often. I feel that I have shown the possessive suffix "s" to be a clitic, and not a case. The loss of gender marks a simplification of an illogical system, and leads the way for a more simple grammar, which I'm sure is something English students across the world would agree is a good thing.
To
conclude, the study of English and its roots provides both a greater
appreciation of the language in its present state, but also a deeper
understanding of the complex history and how English has risen from the
language of barbarians on a lonely isle to a truly international language.
Through historical linguistics, we have been able to deduce facts about
features of the present language, allowing for better grammatical analyses. As
society is grounded in communication and relationships, by studying language we
are studying the very nature of human society.
Bibliography
Books
A Guide to Old English: Sixth Edition - Bruce Mitchell and Fred C. Robinson - 2002
Historical Linguistics - Theodora Bynon - 1983
The Making of English - Henry Bradley - 1957
From Old English to Modern English - Ben Clapp - 2003
Wordcraft - Stephen Pollington - 2002
The Pelican Guide to English Literature, Volume 1: The Age of Chaucer - 1976
The Concise Dictionary of English Etymology (Oxford Paperback Reference) - T. F. Hoad (Editor) - 1993
The Cambridge Encyclopedia of Language - David Crystal - 1996
Websites
William Z. Shetter's Language Miniatures (http://home.bluemarble.net/~langmin/)
University of Virginia English Department's "Magic" Declension Table (http://www.engl.virginia.edu/OE/courses/handouts/magic.pdf)
The Ormulum Project (http://www.english.su.se/nlj/ormproj/ormulum.htm)
Acknowledgements
Ða Engliscan Gesiþas for introducing me to Anglo-Saxon culture and language.
Ben Clapp for motivating me to do this, and also for those long nights discussing the finer points of his paper.
Gabor Sandi, J.S Burke, and Mark Rosenfelder for their enlightening discussions on Modern English morphology.
Ruth Staver and Stephen Pollington for their helpful explanations of Old English phonology, and general assistance with regards to the dark ages.
My teachers Mr T Birrell and Dr J Casey for their comments and support.
[1] A clitic is a morpheme similar to an affix, but it attaches to words of many syntactic categories. It is phonologically bound yet syntactically free.
[2] The Old English preposition of was more often used to represent features native to the dative case: "with" or "out of". The Modern English use of "of" has been heavily influenced by French de.
[3] From email correspondence with G Sandi
[4] Then again, loss of ambiguity may not necessarily be a benefit. From a literary point of view, ambiguity is a very useful tool to have at your disposal, and a language with no ambiguity would make for very plain literature. So perhaps "effect" is a more fitting word than "benefit".
Home Blog Writing Pictures Conlangs Mirror About Contact Links
All content Copyright © 2003-2007 Rory Turnbull, with the exception of some pictures on the pictures pages, and all of the content on the mirror page. They are the property of their respective owners.