AUTHOR: Ringe, Don
TITLE: A History of English
SUBTITLE: Volume 1. From Proto-Indo-European to Proto-Germanic
SERIES TITLE: A Linguistic History of English
PUBLISHER: Oxford University Press
David Stifter, Institut für Sprachwissenschaft, Universität Wien
[Preliminary remark: Because of the lack of diacritic symbols in the
email-format, in the following review for reconstructed forms a less than
adequate approximation to what they usually look like in Indo-European and
Germanic studies will have to be used.]
Don Ringe is Professor in Linguistics at the University of Pennsylvania. He has
published on a wide range of Indo-European (IE) topics, and he is the author of
a historical phonology of Tocharian. This volume is the first in a series by the
author which aims to trace the entire recoverable history of the English
language, from the earliest reconstructable stage, the Proto-Indo-European (PIE)
language spoken in the centuries around 4000 BC, probably in the river valleys
of Ukraine, to the present. The first volume describes the development of PIE
into Proto-Germanic, both of which are unattested, reconstructed stages in the
pre-English language history. Frequent references in this volume to the as yet
unpublished vol. 2 give evidence of the tight concept of the series.
Chapter 1: General introduction (pp. 1–3)
Ringe briefly sets out the aims and the methodology of the book. It is not
primarily aimed at comparative linguists, but he rather intends to present
reliable information on historical Germanic grammar to readers ''who have not
undertaken serious study of Indo-European or comparative Germanic linguistics,
nor of the history of English.'' Yet familiarity with the basic concepts of
modern linguistic theory is required for perusal of the book.
Chapter 2: Proto-Indo-European (4–66)
The second chapter begins with Ringe's cladistic tree of the IE language family.
How the subgroups of IE relate to one another is no matter of secondary
importance. As in philology, where the genealogical tree of manuscripts
determines editorial choices, in historical linguistics the chosen model of
subgrouping will have far-reaching consequences on how one has to assess the
evidential value of a specific feature attested only in a subgroup of languages.
Therefore the genealogical model should always be made explicit. This is what
Ringe does: for him, Germanic is part of a Central IE dialect continuum,
stretching from Indo-Iranian across Greek to Balto-Slavic and Germanic (p. 5–6).
Anatolian, Tocharian, and Italo-Celtic stand outside this central group.
Chapter 2 amounts to an outright introduction into the grammar of the IE
proto-language, setting out in a concise, but precise manner the essentials of
how PIE phonology and morphology are conceived of in modern IE studies. All
grammatical categories receive attention. Although they are not the primary
targets of the book, this chapter qualifies as a reference grammar or as a point
of departure even for students of non-Germanic philologies.
Chapter 3: The development of Proto-Germanic (67–212)
This chapter traces the phonological and morphological changes in the post-PIE,
but pre-Proto-Germanic period. To put it into perspective: this period covers,
in conservative estimation, 2500 years, from c.3000 BC, by which time the split
into the principal branches of IE must have occurred, to c.500 BC, when a stage
in the linguistic development was presumably reached that can sensibly be called
Proto-Germanic. That means that this period is approximately as long as the
period since then, and it is considerably longer than the period in which
Germanic languages have been actually attested in written texts.
The first, longer part of the chapter (67–151) presents the regular sound
changes which transformed PIE into Proto-Germanic. Most notable among them, and
well presented by Ringe, is the complex of changes known as Grimm's Law (the
great Germanic sound-shift) and Verner's Law, which radically altered the system
of obstruents, and which left only few lexical items of the language untouched.
But Ringe does not neglect very minor changes. For instance, on p. 116, he
mentions with reservations possible examples of the 'articulatory leap' of *w…kw
> *w...p in pre-Germanic. In this context, it may merit consideration whether
the unexpected treatment of the labiovelars in the cardinal numbers of Germanic
is connected with this limited rule: in the numerals '4' and '5', where a reflex
of PIE *kw would be expected (i.e. Gmc. **hw), what is found in Germanic
languages are reflexes of *p. Perhaps what happened was phonetic influence,
partly across the lexical boundaries, in the numerical series PIE *KWetWóres,
pénKWe, sWék's '4, 5, 6' > pre-Grm. *PetWóres, pémPe, sWéks (vel sim.). (See,
however, Schaffner 2001: 170 for an alternative application of the 'articulatory
leap' to the numerals).
The section on the phonology is closed off by a table illustrating the relative
chronologies and dependencies of the sound changes (p. 152); a table which
readers of Ringe's book on Tocharian phonology will find familiar. Intimidating
as it may look at first, presenting the matter in that manner is an interesting
way of visualising the complex developments, and particularly of visualising the
The second part of the chapter is devoted to the changes in inflectional
morphology (pp. 151–211). Being much more complex than the developments in
nominal inflection (196–212), Ringe starts with verbal morphology (153–196):
what came out of the relatively intricate aspectual verbal system of PIE in
Germanic is a relatively simplified and unified tense-based system. The general
outlines of the transformations are clear, but some details are still puzzling,
like, for example, the origins of the Germanic weak preterite endings, for which
Ringe offers his own explanation (167–8).
Chapter 4: Proto-Germanic (213–297)
The final chapter offers a grammar of the Proto-Germanic language as it would
have looked in a synchronic description after the transformations had taken
place that are described in chapter 3. In a way, this repeats in a systematic
and detailed way what was alluded to in the previous chapter. Like in all
chapters of the book, the emphasis lies on phonology and inflection;
derivational morphology and syntax receive only a very short and passing
treatment at the end. On various occasions Ringe stresses the role which native
and foreign language learners had in transforming the phonology and morphology
of Germanic (e.g. pp. 171, 215, etc.). This sociolinguistic aspect is a new and
interesting angle to historical linguistics.
This is a highly welcome and useful book for scholars and advanced students of
comparative Indo-European and Germanic linguistics and the history of English.
It provides an up-to-date introduction to the earliest stages of the pre-history
of English. Yet, as stated by the author himself in the introductory chapter,
this is not really a book for the specialists in historical Germanic linguistics
who require much more detail in a reference book.
Most of the points of criticism that can be raised are minor ones points that
are really questions of personal judgment (and sometimes predilection). It is
only natural in a matter like the reconstruction of an unattested
proto-language, which to a large degree involves personal judgment, that no two
scholars will agree on everything. When in the following I pick out a few points
in the present book, where I think that more than just a single person could
express disagreement, this is not to be misunderstood as nitpicking, but as
highlighting how many details remain controversial even after so much progress
has been made in IE linguistics:
On p. 43, it is a pity that Ringe only remarks in passing that the ending of the
PIE thematic ablative singular goes back to an ''endingless loc. sg. in *-e plus
the adverb (postposition?) *ád [...], which clearly did not mean 'to' in
pre-PIE.'' This idea is quite interesting, but it goes against the prima-facie
evidence in many languages. It would have been helpful if Ringe had given more
semantic and morphological details here, even more so as the original idea was
formulated in a PhD-thesis by Ronald Kim, and therefore is not easily accessible.
On pp. 47–50, Ringe gives a great number of full inflectional paradigms of
reconstructed PIE nouns. While in many cases the paradigms are quite generally
accepted, I disagree with the inflection of 'woman' on p. 49. The very isolation
of the relevant paradigm in Old Irish (and partly in Gaulish) virtually
guarantees the archaism of its inflection in these languages, and should be
taken as a strong indication for the inflection in the IE proto-language. The
attested plural forms like nom./acc. OIr. mná, Gaul. mnas and gen. OIr. ban,
Gaul. bnanom (<< *banom) have to be derived from (pre-Celtic = PIE?) *gwnéh2es
and *gwnh2óm (or *gwnh2óHom in Ringe's system, with an ending that is an
outright morphological monster), forms, however, which are unreconcilable with
the PIE pre-forms *gwénh2es and *gwnéh2oHom set up by Ringe. It is a
methodological decision whether one reconstructs PIE grammar as an ideal and
internally coherent and logical system (which is what Ringe's system would be),
or whether one concedes that, like any real language, the grammar of PIE was
structurally imbalanced to a certain degree.
On p. 46 et passim, on several occasions Ringe reconstructs the PIE word for
'sun' as *soh2wl, i.e. a proterokinetic paradigm with o-grade of the root,
instead of an e-grade root *seh2wl, as would be expected. While this is totally
irrelevant for Germanic historical linguistics (since *oh2 and *eh2 both had the
same outcome *o: in Proto-Germanic), it should be stressed that there is no
compelling piece of evidence for the preform used by Ringe. Masculine Latin
so:l, which could be cited as possibly relevant, does not continue the neuter
appellative, but rather an animate internal possessive derivative *seh2wo:l,
possibly referring to a sun-god conceived of as a male person (J. Schindler p.c.).
On p. 154, Old Irish do•formaig 'adds to, increases' does not belong to the root
*mogh (or rather *magh) 'to be able', but rather to a Celtic root *mak 'to
raise, to increase' (Schumacher 2004: 466); whether rare Old Irish mochtae
'mighty?' belongs here, is not absolutely sure either.
On p. 186, in the matter of long-vowel perfects of C(C)eC-roots, Schumacher 2005
offers a PIE solution.
On two occasions (p. 196 and 263) Ringe makes reference to an alternative
present paradigm of the verb 'to be' formed from a stem *bi-, which survives
intact only in Old English, beside the ordinary present of the stem *iz-, which
underlies all other Germanic languages. Rather defiantly Ringe remarks that a
reconstruction of the pre-forms of this alternative present is not possible, and
he remains somewhat undecided as to the semantics of the verb. But see now the
extensive treatment of the matter in Schumacher (2007: 185–200), where it is
argued that a habitual present stem *bi- of the verb 'to be' is a common
West-Germanic innovation, brought about by contact with Celtic languages.
Twice (pp. 194 and 265) Ringe calls the exact etymology of Gothic iddja, OE eode
'went' one of the mysteries of Germanic historical linguistics. The most recent
literature he refers to is more than thirty years old. Actually, the two most
recent contributions to the question are Schumacher 1998  and Eichner
2005, who, however, offer competing solutions.
On p. 270, Ringe ascribes Verner's Law alternations across the Germanic
languages in neuter a-stems like 'blood' or 'gold' to shifts of the accent,
brought about by a derivational rule in the plural formation. But some of the
examples are mass nouns; a plural formation would not be expected in the first
place. Verner's Law alternations in these instances must either be analogical to
neuter a-stem nouns with a regular plural formation, or alternative explanations
are called for (see, for example, Schaffner 2001).
The book not only condenses established wisdom, it also presents new
suggestions, like the development of unstressed (that is, after the stress had
shifted to initial syllables) *ew > *ow > Pr-Germ. *aw (pp. 125–126). Germanic
would thus join in a wide-spread tendency of NW-European languages (Italic,
Celtic, Balto-Slavic) for rounding assimilation of *e to *o before the labial
glide (modulo the details in the individual languages).
Another point is the following: a phonological feature that connects Germanic
with neighbouring Balto-Slavic is the merger of the inherited non-high back
vowels in Germanic: the short vowels resulted in a sound that is traditionally
reconstructed as *a, the long vowels apparently merged in *o: (but see also
Schrijver 2003 who claims that *a: and *o: remained distinct in certain
contexts, and who has a completely different approach to the problem of the
various treatments of long, back non-high vowels in final syllables, treated in
this book on pp. 73–75). But it has been suspected that the product of the
merger of the long vowels had been low *a: at first as well, which had then
shifted to mid-high *o: only within Proto-Germanic. In that sense, Ringe calls
the rounding of *a: to *o: one of the ''latest reconstructable sound changes'' in
the prehistory of Germanic, ''possibly spreading through an already diversified
dialect continuum'' (p. 147). The evidence he cites is Gothic Rumoneis
/ru:mo:ni:s/ 'Roman', which was supposedly borrowed from Latin Ro:ma:ni: at a
time when Germanic had not yet acquired *o: (in that reckoning, Latin o:
therefore had to be substituted by Germanic *u:), and at a time when Germanic
still had *a:, which subsequently shifted to *o:. Intriguing as this scenario
may look in theory, it does not take into account the historical situation.
Instead it is much more likely that *Ru:mo: 'Rome' and *Ru:mo:nas are common
Germanic loanwords from conjecturable Gaulish *Ru:ma: and *Ru:ma:nos, perhaps
reflected in inscriptions as ''Matronae Rumanehae''. Putative Germanic Bacenis
Silua (not mentioned by Ringe), the ''Beech Wood'' in Julius Caesar's Commentaries
on the Gaulish War (1st c. BC), is not probative for the existence of *a: at
Caesar's time either, because it could have reached the Romans in Gaulicized
shape (i.e. substitution of Gaul. /a:/ for /Germ. [å:] or [o:]); the same is
true for a number of other loanwords. There is ultimately no proof at all that
the inner-Germanic shift from *a: > *o: occurred at any particularly late point
of time. In fact, it is not possible at all to make any chronological inferences
for this change (see Stifter forthcoming for more details).
In the context of this putative change, Ringe makes the somewhat mysterious
claim that the ''latest possible date for direct contacts'' between Germanic and
Latin was 113 BC at the battle near Noreia (p. 146). Is this a typo for
''earliest possible date'' (or ''latest possible date for first direct contacts'')?
Surely Roman-Germanic contacts intensified in the decades after that battle; one
needs only to think of the Suebian king Ariovistus and his dealings with the
Roman senate and Julius Caesar during the latter's conquest of Gaul.
Although the above remarks are minor, a few desiderata are more strongly felt:
Ringe takes a cautious position where the evidence allows no clear answer. While
this is fine for an introduction, the specialist would like to have presented
all reasonable contributions to a controversial problem, and (s)he would like to
have a reference section where the most important and the most recent literature
for a topic can be found. This is lacking in the book: for example, in the
section which deals with Verner's Law's effects in nominal inflection (pp.
269–271), no mention is made of Schaffner 2001, and in the section on consonant
stems (pp. 278–279) no reference is made to Griepentrog 1995. Despite Ringe's
cautiousness, a few phenomena have been treated too idiosyncratically: for
example, it is quite astonishing that in the chapter about the reflexes of PIE
laryngeals in Germanic no mention is made of the so-called 'Verschärfung', i.e.
of the development *-RH- > *-RR-, a sound change which, as far as I can see, is
largely accepted among specialists in Germanic linguistics. Only on p. 241, in a
completely different context, does Ringe make a short, dismissive remark. Even
if the author does not believe in a laryngealistic explanation of the
phenomenon, it is inexcusable in this case not to cite works such as Lühr 1976
or Jasanoff 1978.
The handbook's prime purpose is a formal-linguistic one. Language and grammar
are treated as abstract and interrelated systems. The prehistoric sound changes
that constitute the Germanic languages are not simply enumerated, but Ringe
discusses them as phenomena with all their effects on morphology and the
language system as a whole. This is highly appreciated. But at the same time
Ringe's rather abstract approach entails that the book becomes conspicuously
unphilological. Languages are spoken by real persons at concrete points in
history, and the language material that has come down to us from earlier periods
has been transmitted under particular circumstances of production and
preservation. The knowledge of all of these factors is important in order to
properly assess the evidential value of the linguistic material with which we
historical linguists work. To do the author credit, he does on occasion remark
that a particular form on which a theory hinges is attested only a single time
(like, for instance, Gothic lais 'I know'). But more could have been said. It is
my conviction that in a book intended for beginners in historical linguistics,
the role of philology cannot be stressed highly enough. After all, it is
philology that not only supplies us with our working material, but which at
times produces physical proof for our theories and hypotheses, for example when
a form that we postulated suddenly finds attestation on an inscription. In this
context, more Runic texts and Roman-age inscriptions from Germanic areas could
have been cited.
Another absence that I felt is that the earliest transmission in secondary
sources and contact relations, doubtlessly important sources of information for
the prehistoric stages of Germanic, could have been considered more thoroughly.
To give an example, it could have been demonstrated that the change e > i in
stressed syllables before tautosyllabic nasals or before a following i was not a
Proto-Germanic development, but was only under way by the first century AD, as
evinced by material in secondary sources: e.g. the personal name Segimundus in
Tacitus, or the Finn. loanword teljo 'thwart, rwoing bench' (cp. OHG dilla
'board'; Kluge 1913: 129–130).
If it is allowed to formulate a wish-list for the future volumes of the series,
it would include: an introductory chapter that gives a survey of the most
important handbooks and secondary literature in Germanic studies, and an
overview of the various Germanic languages and dialects and their geographical
extension, their history, and their attestation. It must be conceded that not
much can be said with certainty on the latter topics in the case of
reconstructed languages (nevertheless, something can be said about these
matters), so this may be a lack particular to the first volume.
In conclusion it must be stressed that this is a very useful book. I await with
anticipation the second volume of _A Linguistic History of English_ (from
proto-Germanic to Old English), and it is to be hoped that it will follow soon.
Eichner, Heiner. (2005) Etymologische Notiz zu gotisch iddja und altenglisch
eode 'ging' aus sprachgeschichtlicher Sicht. In _Indogermanica. Festschrift Gert
Klingenschmitt_, ed. by Günter Schweiger, Taimering: VWT, 71–72.
Griepentrog, Wolfgang. (1995) _Die Wurzelnomina des Germanischen und ihre
Vorgeschichte_. Innsbruck: IBS.
Jasanoff, Jay. (1978) Observations on the Germanic Verschärfung. _Münchener
Studien zur Sprachwissenschaft_ 37, 77–92.
Kluge, Friedrich. (1913) _Urgermanisch. Vorgeschichte der altgermanischen
Dialekte_. Strassburg: Trübner.
Lühr, Rosemarie. (1976) Germanische Resonantengemination durch Laryngal.
_Münchener Studien zur Sprachwissenschaft_ 35, 73–92.
Schaffner, Stefan. (2001) _Das Vernersche Gesetz und der innerparadigmatische
grammatische Wechsel des Urgermanischen im Nominalbereich_. Innsbruck: IBS.
Schrijver, Peter. (2003) Early Developments of the Vowel Systems of North-West
Germanic and Saami. In _Languages in Prehistoric Europe_, ed. by Alfred
Bammesberger and Theo Vennemann. Heidelberg: Winter, 195–226.
Schumacher, Stefan. (1998 ) Eine alte Crux, eine neue Hypothese: gotisch
iddja, altenglisch eode. _Die Sprache_ 40, 179–201.
Schumacher, Stefan. (2004) _Die keltischen Primärverben_. Innsbruck: IBS.
Schumacher, Stefan. (2005) 'Langvokalische Perfekta' in indogermanischen
Einzelsprachen und ihr grundsprachlicher Hintergrund. In _Sprachkontakt und
Sprachwandel. Akten der XI. Fachtagung der Indogermanischen Gesellschaft_. 2000
Halle/Saale, ed. by Gerhard Meiser und Olav Hackstein, Wiesbaden: Reichert
Schumacher, Stefan. (2007) Die Deutschen und die Nachbarstämme: Lexikalische und
strukturelle Sprachkontaktphänomene entlang der keltisch-germanischen
Übergangszone. In Hans Hablitzel and David Stifter, eds. _Johann Kaspar Zeuß im
kultur- und sprachwissenschaftlichen Kontext_. Wien: Praesens Verlag, 167–207.
Stifter, David. (Forthcoming) The Proto-Germanic shift *a: > *o: and early
Germanic linguistic contacts. _Historische Sprachforschung_.
ABOUT THE REVIEWER
David Stifter is contract assistant and lecturer at the Department for
Indo-European Linguistics at the University of Vienna. His focus of research
lies in Celtic languages.