Рroto-Indo-European

Introduction

Language is the most important aspect in the life of all beings. We use language to express inner thoughts and emotions, make sense of complex and abstract thought, to learn to communicate with others, to fulfill our wants and needs, as well as to establish rules and maintain our culture. Language is everything.

Have you ever thought how many languages are there in the world? Are modern languages descended from a common ancestor?

It’s estimated that up to 7,000 different languages are spoken around the world. 90% of these languages are used by less than 100,000 people. Over a million people converse in 150−200 languages and 46 languages have just a single speaker.

2,200 of the world’s languages can be found in Asia, while Europe has a mere 260.

Nearly every language uses a similar grammatical structure, even though they may not be linked in vocabulary or origin. Common features, especially common words, shared by many of the languages used in Europe, India, and Asia, led scholars to believe that these languages may have developed from the same source.

That source language was never written down and is now extinct, but it has a name: it is called the Proto Indo-European language, and the family of languages believed to have developed from it is called Indo-European.

English is spoken practically all over the world. It is spoken as the mother tongue in Great Britain, the United States of America, Canada, Australia and New Zealand. A lot of people speak English in China, Japan, India, Africa and other countries. It is one of 6 official languages of the United Nations. It is studied as a foreign language in many schools. English is related to German and Dutch, and they are all part of the Indo-European family of languages. These also include Romance languages, such as French, Spanish and Italian, which come from Latin.

Languages are grouped into families that share a common ancestry. English is related to German and Dutch, and they are all part of the Indo-European family of languages. These also include Romance languages, such as French, Spanish and Italian, which come from Latin.

The object of our research is a common ancestral tongue of the Indo-European languages — the Proto-Indo-European language. This language is a mother tongue of many languages, that’s why it is a language of a profound interest for all linguists. This field is not loosing its popularity among them, because this protolanguage is still in the process of its reconstruction and wide research. Many claims are still debated, in their details by different scholars, so in our coursework we will try to do our own research on the field of historical linguistics and will give you a wide information about Proto-Indo-European language.

1. Proto-Indo-European language

1.1 General information about Proto-Indo-European language

All living languages evolve over time, adding and losing vocabulary, morphological behavior, and syntactic structures, and changing in the ways they are pronounced by their speakers. Even without knowing how or why these evolutionary mechanisms operate, one can still get a feel for their effects; for example, they account for the differences between American and British English, and for the fact that neither Americans nor Brits can understand Beowulf at all without first being taught how to read the Old English language in which it was composed. Even the writings of Shakespeare — much more recent than Beowulf can be difficult for modern English speakers to interpret. The field of study that concerns itself with language evolution is called historical linguistics.

A large number of related languages form what is called the Indo-European macrofamily. The Proto-Indo-European language is the reconstructed common ancestor of the Indo-European languages, spoken by the Proto-Indo-Europeans 6,000 years ago somewhere in the general vicinity of the Pontic Steppe north of the Black Sea and east to the Caspian — an area that, perhaps not accidentally, seems to coincide with the land of the ancient Scythians, from the Ukraine across far southwestern Russia to western Kazakhstan. PIE was the first proposed ancestral tongue to be widely accepted by linguists. Far more work has gone into reconstructing it than any other proto-language and it is by far the most well-understood of all proto-languages at its time depth. During the 19th century, the vast majority of linguistic work was devoted to reconstruction of Proto-Indo-European or of daughter proto-languages such as Proto-Germanic and most of the current techniques of historical linguistics (e.g. the comparative method and the method of internal recounstruction) were developed as a result.

The Indo-European proto-languages themselves evolved, each giving rise to its own family of languages. Each family is identified with the proto-language from which it sprung; these families are conventionally listed in order, roughly from west to east with respect to the homelands their speakers came to occupy. The ten families, linked to modern maps of their homeland areas (which open in a separate window), are:

1. Celtic, with languages spoken in the British Isles, in Spain, and across southern Europe to central Turkey;

2. Germanic, with languages spoken in England and throughout Scandinavia and Central Europe to Crimea;

3. Italic, with languages spoken in Italy and, later, throughout the Roman Empire including modern-day Portugal, Spain, France, and Romania;

4. Balto-Slavic, with Baltic languages spoken in Latvia and Lithuania, and Slavic throughout eastern Europe plus Belarus and the Ukraine and Russia;

5. Balkan (exceptional, as discussed below), with languages spoken mostly in the Balkans and far western Turkey;

6. Hellenic, spoken in Greece and the Aegean Islands and, later, in other areas conquered by Alexander (but mostly around the Mediterranean);

7. Anatolian, with languages spoken in Anatolia, a.k.a. Asia Minor, i.e. Modern Turkey;

8. Armenian, spoken in Armenia and nearby areas including eastern Turkey;

9. Indo-Iranian, with languages spoken from India through Pakistan and Afghanistan to Iran and Kurdish areas of Iraq and Turkey;

10. Tocharian, spoken in the Tarim Basin of Xinjiang, in far western China.

Scholars estimate that PIE may have been spoken as a single language (before divergence began) around 3700 BC, though estimates by different authorities can vary by more than a millennium. The most popular hypothesis for the origin and spread of the language is the Kurgan hypothesis, which postulates an origin in the Pontic-Caspian steppe of Eastern Europe.

The existence of PIE was first postulated in the 18th century by Sir William Jones, who observed the similarities between Sanskrit, Ancient Greek and Latin. By the early 20th century, well-defined descriptions of PIE had been developed that are still accepted today (with some refinements). The largest developments of the 20th century have been the discovery of Anatolian and Tocharian languages and the acceptance of the laryngeal theory. The Anatolian languages have also spurred a major re-evaluation of theories concerning the development of various shared Indo-European language features and the extent to which these features were present in PIE itself.

PIE is thought to have had a complex system of morphology that included inflections (suffixing of roots, as in who, whom, whose), and ablaut (vowel alterations, as in sing, sang, sung). Nouns used a sophisticated system of declension and verbs used a similarly sophisticated system of conjugation.

Relationships to other language families, including the Uralic languages, have been proposed but remain controversial.

There is no written evidence of Proto-Indo-European, so all knowledge of the language is derived by reconstruction from later languages using linguistic techniques such as the comparative method and the method of internal reconstruction.

There are several competing hypotheses about when and where PIE was spoken. The Kurgan hypothesis is «the single most popular» model, postulating that the Kurgan culture of the Pontic steppe were the hypothesized speakers of the reconstructed Proto-Indo-European language. Alternative theories such as the Anatolian urheimat have also gained acceptance.

The satemization process that resulted in the Centum-Satem isogloss probably started as early as the 4th millennium BC and the only thing known for certain is that the proto language must have been differentiated into unconnected daughter dialects by the late 3rd millennium BC.

Mainstream linguistic estimates of the time between PIE and the earliest attested texts range around 1,500 to 2,500 years, with extreme proposals diverging up to another 100% on either side.

Indo-European studies began with Sir making and propagating the observation that Sanskrit bore a certain resemblance to classical Greek and Latin. In the Sanscrit language he suggested that all three languages had a common root, and that indeed they might further all be related, in turn, to Gothic and the Celtic languages, as well as to Persian.

His third annual discourse before the Asiatic Society on the history and culture of the Hindus with the famed «philologer» passage is often cited as the beginning of comparative linguistics and Indo-european studies. This is Jones' most quoted passage, establishing his tremendous find in the history of linguistics: The Sanscrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs and the forms of grammar, than could possibly have been produced by accident; so strong indeed, that no philologer could examine them all three, without believing them to have sprung from some common source, which, perhaps, no longer exists; there is a similar reason, though not quite so forcible, for supposing that both the Gothic and the Celtic, though blended with a very different idiom, had the same origin with the Sanscrit; and the old Persian might be added to the same family.

The classical phase of Indo-European comparative linguistics leads from Franz Bopp’s Comparative Grammar to August Schleikher’s Compendium and up to Karl Brugmann’s Grundriss published from the 1880s. Brugmann’s grammatic re-evaluation of the field and Ferdinand de Saussure’s development of the laryngeal theory may be considered the beginning of «contemporary» Indo-European studies.

PIE as described in the early 20th century is still generally accepted today; subsequent work is largely refinement and systematization, as well as the incorporation of new information, such as the Anatolian and Tocharian branches unknown in the 19th century.

The laryngeal theory, in its early forms discussed since the 1880s, became mainstream after Jerzy Kuryіowicz's 1927 discovery of the survival of at least some of these hypothetical phonemes in Anatolian.

Julius Pokorny’s landmark Indogermanisches etymologisches Wцrterbuch («Indo-European Etymological Dictionary», 1959) gave a detailed overview of the lexical knowledge accumulated up until that time, but neglected contemporary trends of morphology and phonology (including the laryngeal theory), and largely ignored Anatolian.

The generation of Indo-Europeanists active in the last third of the 20th century developed a better understanding of morphology and, in the wake of Kuryіowicz's 1956 Apophonie, understanding of the ablaut. From the 1960s, knowledge of Anatolian became certain enough to establish its relationship to PIE.

As we said before there is no direct evidence of PIE, because it was never written. All PIE sounds and words are reconstructed from later Indo-European languages using the comparative method and the method of internal reconstruction. Anasteriskis used to mark reconstructed PIE words, such as *wуdr? 'water', *?w?n 'dog' (English hound), or *trйyes 'three (masculine)'. Many of the words in the modern Indo-European languages seem to have derived from such «protowords» via regular sound changes (e.g., Grimm’s law).

As the Proto-Indo-European language broke up, its sound system diverged as well, according to various sound laws in the daughter languages. Notable among these are Grimm’s law and Verner’s law in Proto-Germanic, loss of prevocalic *p — in Proto-Celtic, reduction to h of prevocalic *s — in Proto-Greek, Brugmann’s law and Bartholomae’s law in Proto-Indo-Iranian, Grassmann’s law independently in both Proto-Greek and Proto-Indo-Iranian, and Winter’s law and Hirt’s law in Balto-Slavic.

Many higher-level relationships between Proto-Indo-European and other language families have been proposed, but these hypothesized connections are highly controversial. A proposal often considered to be the most plausible of these is that of an Indo-Uralic family, encompassing PIE and Uralic. The evidence usually cited in favor of this consists in a number of striking morphological and lexical resemblances. Opponents attribute the lexical resemblances to borrowing from Indo-European into Uralic. Frederik Kortlandt, while advocating a connection, concedes that «the gap between Uralic and Indo-European is huge», while Lyle Campbell, an authority on Uralic, denies any relationship exists.

Other proposals, further back in time (and proportionately less accepted), link Indo-European and Uralic with Altaic and the other language families of northern Eurasia, namely Yukaghir, Korean, Japanese, Chukotko-Kamchatkan, Nivkh, Ainu, and Eskimo-Aleut, but excluding Yeniseian (the most comprehensive such proposal is Joseph Greenberg’s Eurasiatic), or link Indo-European, Uralic, and Altaic to Afro-Asiatic and Dravidian (the traditional form of the Nostratic hypothesis), and ultimately to a single Proto-Human family. A more rarely mentioned proposal associates Indo-European with the Northwest Caucasian languages in a family called Proto-Pontic. Etruscan shows some similarities to Indo-European, such as a genitive in — s. There is no consensus on whether these are due to a genetic relationship, borrowing, chance and sound symbolism, or some combination of these.

The existence of certain PIE typological features in Northwest Caucasian languages may hint at an early Sprachbund or substratum that reached geographically to the PIE homelands. This same type of languages, featuring complex verbs of which the current Northwest Caucasian languages might have been the sole survivors, was cited by Peter Schrijver to indicate a local lexical and typological reminiscence in western Europe pointing to a possible Neolithic substratum.

PIE had a free pitch accent, which could appear on any syllable and whose position often varied among different members of a paradigm (e.g. between singular and plural of a verbal paradigm, or between nominative/accusative and oblique cases of a nominal paradigm). The location of the pitch accent is closely associated with ablaut variations, especially between normal-grade vowels (/e/ and /o/) and zero-grade vowels (i.e. lack of a vowel). The accent is best preserved in Vedic Sanskrit and (in the case of nouns) Ancient Greek, and indirectly attested in a number of phenomena in other PIE languages.

PIE was an inflected language, in which the grammatical relationships between words were signaled through inflectional morphemes (usually endings). The roots of PIE are basicmorphemes carrying a lexical meaning. By addition of suffixes, they form stems, and by addition of desinences (usually endings), these form grammatically inflected words (nouns or verbs). PIE roots are understood to be predominantly monosyllabic with a basic shape CvC©. This basic root shape is often altered by ablaut. Roots which appear to be vowel initial are believed by many scholars to have originally begun with a set of consonants, later lost in all but the Anatolian branch, called laryngeals (usually indicated *H, and often specified with a subscript number *h?, *h?, *h?). Thus a verb form such as the one reflected in Latin agunt, Greek? гпхуй (бgousi), Sanskrit ajanti would be reconstructed as*h2e?-onti, with the element *h?e? — constituting the root per se.

An important component of PIE morphophonology is the variation in vowels commonly termed ablaut, which occurred both within inflectional morphology (among different members of a nominal or verbal paradigm) and derivational morphology (between, for example, a verb and an associated verbal noun). Ablaut in PIE was closely associated with the position of the accent; for example, the alternation found in Latin est, sunt reflects PIE *h1йs-ti, *h1s-уnti. However, it is not possible to derive either one directly from the other. The primary ablaut variation was between normal grade or full grade (*/e/ and */o/), lengthened grade (*/з/ and */ф/), and zero grade (lack of a vowel, which triggered the vocalic allophones of nearby resonants). The normal grade is often characterized as e-grade or o-grade depending on the particular vowel involved. Ablaut occurred both in the root and the ending. Originally, morphological categories were distinguished both by ablaut variations and different endings, but the decay of endings has led some languages to use ablaut alone to distinguish grammatical categories, as in the Modern English words sing, sang, sung, originally reflecting a pre-Proto-Germanic sequence *sengw-, *songw-, *sngwSome scholars believe that the inflectional affixes of Indo-European reflect ablaut variants, usually zero-grade, of older PIE roots. Often the zero-grade appears where the word’s accent has shifted from the root to one of the affixes.

Proto-Indo-European nouns were declined for eight or nine cases (nominative, accusative, genitive, dative, instrumental, ablative, locative, vocative, and possibly a directive or allative). There were three genders: masculine, feminine, and neuter.

There are two major types of declension, thematic and athematic. Thematic nominal stems are formed with a suffix * - o — (in vocative * - e) and the stem does not undergo ablaut. The athematic stems are more archaic, and they are classified further by their ablaut behaviour (acrostatic, proterokinetic, hysterokinetic and amphikinetic, after the positioning of the early PIE accent in the paradigm).

PIE pronouns are difficult to reconstruct owing to their variety in later languages. This is especially the case for demonstrative pronouns. PIE had personal pronouns in the first and second person, but not the third person, where demonstratives were used instead. The personal pronouns had their own unique forms and endings, and some had two distinct stems; this is most obvious in the first person singular, where the two stems are still preserved in English I and me. According to Beekes, there were also two varieties for the accusative, genitive and dative cases, a stressed and an enclitic form.

The Indo-European verb system is complex and, like the noun, exhibits a system of ablaut. The most basic categorization for the Indo-European verb was grammatical aspect. Verbs were classed as stative (verbs that depict a state of being), imperfective (verbs depicting ongoing, habitual or repeated action) or perfective (verbs depicting a completed action or actions viewed as an entire process). Verbs have at least four moods (indicative, imperative, subjunctive and optative, as well as possibly the injunctive, reconstructible from Vedic Sanskrit), two voices (active and mediopassive), as well as three persons (first, second and third) and three numbers (singular, dual and plural). Verbs were also marked by a highly developed system of participles, one for each combination of tense and mood, and an assorted array of verbal nouns and adjectival formations.

As PIE was conjectured to be spoken by a prehistoric society, no genuine sample texts are available, but since the 19th century modern scholars have made various attempts to compose example texts for purposes of illustration. These texts are educated guesses at best; Calvert Watkins in 1969 observes that in spite of its 150 years' history, comparative linguistics is not in the position to reconstruct a single well-formed sentence in PIE. Because of this and other similar objections based on Pratishakyas, such texts are of limited use in getting an impression of what a coherent utterance in PIE might have sounded like.

1.2 Proto-Indo-European phonology

The descriptions of PIE phonology found in our handbooks date in great part from the nineteenth century. When we maintain these descriptions we accept the foundations on which they are based. At present our data have been increased by Hittite, and our linguistic methodology refined. In this monograph I shall examine the new data and some well-known but unexplained data in accordance with current linguistic methodology and propose a revised description of PIE phonology.

The phonology of PIE was first described and greatly developed in the course of the nineteenth century. Grimm, Rask, and Bopp did not even attempt to reconstruct the language from which the dialects developed, assuming it to be much like Skt. Schleicher, trying to escape such undue emphasis on Skt., reconstructed PIE on the basis of all the dialects. Yet with the nineteenth century emphasis on phonetics he and other Indo-Europeanists dealt primarily with the sounds of IE, not with their patterning. Thus the chief contribution of nineteenth century Indo-Europeanists consisted in determining the phonetic system of PIE. Only a few concerned themselves with its phonemic structure. Today this study rather than the phonetic description of PIE needs further development.

The chief difficulties removed between 1860 and 1880 were those solved in the formulations known as Grassmann’s Law, Verner’s Law, and the Law of Palatals.

Grassmann in 1863 demonstrated that in Ind.-Ir. and Greek the first of two aspirates beginning successive syllables or a syllable that ends in an aspirate lost its aspiration: e.g. Skt. dadhвмti, Gk. ФЯизфй `places', from PIE dhe-dhз-; Skt. Bandh — `bind', Gk. Рениесьт `step-father', but Mod. Eng. bind, from PIE bhendhThe deaspiration occurred independently in the two dialects. After Grassmann’s discovery the PIE aspirates could be reconstructed from reflexes in Ind.-Ir. and Gk. as well as from Gmc. and, less easily, some of the other dialects.

Verner in 1877 demonstrated that in Gmc., voiced spirants developed from voiceless spirants if the preceding syllable did not have the chief word accent in PIE: e.g. Goth. frawaнrюan `spoil', frawardjan `ruin' from PIE wert — (compare Skt. Vбrtate, vartбyati); OHG ziohan `pull', pret. Part. gizogan, from PIE deukHis demonstration indicated that the movable pitch accent of PIE had survived into early Proto-Germanic.

About the time of Verner’s discovery a number of linguists independently demonstrated that in PInd.-Ir. velar consonants were palatalized before i о y e з: e.g. Skt. Ugrб `strong', уjоya `stronger'; cakбra `I have made' from PIE kekora. After the origin of these Ind.-Ir. palatals was demonstrated, the group of PIE consonants was even more clearly defined.

The difficulties that remained involved only a small proportion of the vocabulary; chief of these were the status of the voiceless aspirates, that of the four proposed spirants [ю юh р рh], and the relationship between velars, labio-velars, and palatals in PIE. Evidence for reconstructing PIE with voiceless aspirates and spirants other than s was drawn primarily from Ind.-Ir. and Gk.; the materials could be variously interpreted, and no interpretation has yet been widely accepted. The problem of the relationship between velars, labio-velars, and palatals is one of analysis. Evidence for the three groups of sounds can be found in the dialects. Interpretation of the evidence has varied with the methods employed in analyzing the data; if the data are analyzed phonetically, three groups of sounds must be assumed; if they are analyzed phonemically, one may set up two groups.

With the solution of the chief problems in the PIE group of consonants went the assumption that the PIE accent was a variable pitch accent like that of Vedic Sanskrit. Verner demonstrated that if one assumed that the PIE (Vedic) accent had survived into PGmc., one could account for the development of voiced from voiceless spirants.

We may note incidentally that Gmc., with all of its changes from PIE, has been involved in the clarification of many of the important problems of IE phonology; PGmc. maintained the PIE accentual system and consonantal system much longer than did IE dialects like Italic and Celtic, and even in some respects than Gk. For the apparently great number of changes described in Grimm’s Law involved little change in the PIE consonant system. The PIE three-stop system was maintained in Gmc.; PIE p b bh > Pgmc. f p b, PIE t d dh > Pgmc. ю t р, PIE k g gh (k? g? g? h) > Pgmc. ч k? (чw kw? w). In other dialects the PIE system was broken down; in Celtic a two-stop consonant system developed, e.g. PIE p b bh > Pcelt. p b. A two-stop system developed also in Slavic and Baltic. Moreover from the evidence we have about the time of the Gmc. accent change we may assume that Gmc. maintained a variable pitch accent like that of PIE until the fourth century B.C.; at this time the Italic and Celtic dialects had an initial stress accent, Gk. a fixed pitch accent. Consequently PGmc. must be considered one of the most conservative IE dialects; investigation of PGmc. may therefore yield important information for a solution of some of the remaining phonological problems of PIE.

2. Comparison of modern languages of Proto-Indo-European origin

In the second chapter of our coursework we would like to compare some words of modern languages which have a common ancestor — the Proto-Indo-European language. As we mentioned above the Proto-Indo-European language is the reconstructed common ancestor of the Indo-European languages, spoken by the Proto-Indo-Europeans 6,000 years ago. So, all the words in Proto-Indo-European language that are given here, do not have the written evidence of their existence, they are just reconstructed words and they are still in the process of their reconstruction and are debated by different linguists. That is why the words in the first row of the following table are all marked with an asterisk (*).

This means that they are believed to be the P.I.E. roots that later became words in other languages, as shown in the other rows of the table.

These are all words about family, particularly family members:

language word indo european

Here are the words about counting:

Here are the words about animals:

Here are the words about plants:

As we see in these tables above many words have almost the same spelling, usually the same word in different languages starts from the same letter. Words strongly resemble each other, with similar sounds conveying similar meaning. And this proves that all these modern languages from Indo-European group have many things in common. They have almost the same word structures even if their pronunciation is different. All these languages came from the same source.

Conclusion

Nearly every language uses a similar grammatical structure, even though they may not be linked in vocabulary or origin. Common features, especially common words, shared by many of the languages used in Europe, India, and Asia, led scholars to believe that these languages may have developed from the same source.

That source language was never written down and is now extinct, but it has a name: it is called the Proto Indo-European language, and the family of languages believed to have developed from it is called Indo-European.

English is spoken practically all over the world. It is spoken as the mother tongue in Great Britain, the United States of America, Canada, Australia and New Zealand. A lot of people speak English in China, Japan, India, Africa and other countries. It is one of 6 official languages of the United Nations. It is studied as a foreign language in many schools. English is related to German and Dutch, and they are all part of the Indo-European family of languages. These also include Romance languages, such as French, Spanish and Italian, which come from Latin.

Languages are grouped into families that share a common ancestry. English is related to German and Dutch, and they are all part of the Indo-European family of languages. These also include Romance languages, such as French, Spanish and Italian, which come from Latin.

The object of our research is a common ancestral tongue of the Indo-European languages — the Proto-Indo-European language. This language is a mother tongue of many languages, that’s why it is a language of a profound interest for all linguists. This field is not loosing its popularity among them, because this protolanguage is still in the process of its reconstruction and wide research. Many claims are still debated, in their details by different scholars, so in our coursework we will try to do our own research on the field of historical linguistics and will give you a wide information about Proto-Indo-European language.

Bibliography

1. Barber, Elizabeth J.W. 1991. Prehistoric Textiles. Princeton University Press.

2. Beekes, Robert S.P. 1995. Comparative Indo-European Linguistics: An Introduction. Amsterdam, Philadelphia: Benjamins.

3. Benveniste, Emile. 1973. Indo-European language and society.

4. Buck, Carl Darling. 1949. A Dictionary of Selected Synonyms in the Principal Indo-European Languages.s.

5. Bynon, Theodora. 1977. Historical Linguistics. Cambridge: Cambridge University Press.

6. Gamkrelidze, Tomas and Vyacheslav Ivanov. 1995. Indo-European and the Indo-Europeans. Berlin, New York: Mouton de Gruyter.

7. Lehmann, Winfred P. 1993. Theoretical bases of Indo-European linguistics. London, New York: Routledge.

8. Mallory, J. P 1989. In Search of the Indo-Europeans. London: Thames and Hudson.

9. Mallory, J.P. and D.Q. Adams. 1997. Encyclopedia of Indo-European Culture. Chicago: Fitzroy Dearborn.

10. Polome, Edgar C. 1989. Essays on Germanic religion. Washington, D. C: Journal of Indo-European Studies Monograph 6.

11. Puhvel, Jan. 1987. Comparative mythology. Baltimore, London: Johns Hopkins University Press.

12. Szemerenyi, Oswald J.L. 1996. Introduction to Indo-European linguistics. Oxford.

13. Watkins, Calvert. 1985. The American Heritage Dictionary of Indo-European Roots. Boston: Houghton Mifflin.