LITHUANIAN QUARTERLY JOURNAL OF ARTS AND SCIENCES
Volume 29, No.3 - Fall 1983
Editor of this issue: Antanas Klimas
Copyright © 1983 LITUANUS Foundation, Inc.
AN ETYMOLOGY OF THE WORD 'TO FEAR'
IN INDIC, BALTIC, AND SLAVIC
WILLIAM R. SCHMALSTIEG
The Pennsylvania State University
The purpose of this paper is to suggest a historical connection between the Indo-European root for 'being' (*bhû-) and the widely represented Baltic, Slavic and Indic root for fear *(bhoy-). In 1973, 108, and 1980, 131-134, I suggested that the earliest form of the root denoting 'being' was a simple *bhe/o (or merely the consonant *bh- plus the ablauting vowel). The simple root form *bhe- (or *bho-) is retained in the second element of such Old Indic compounds as garda-bhá-, râsa-bha- 'donkey,' vrsa-bhá-'bull.' Mayrhofer, 1956, I, 327, notes that the initial elements gard- and râsa- probably have something to do with 'crying out.' Thus the donkey could be considered the 'crying, shouting being.' The initial element vrsa- denotes 'male' so that the bull could be considered the 'male being' (Mayrhofer, 1956, III, 251).
In my 1973 and 1980 works mentioned above I have also proposed that within Indo-European there were various monophthongizations such that *-ew-C passed to *-û-C, and *-ey-C passed to *-î-C at the same time that the disyllabic sequences *-éw-eC and *-éy-eC passed to *-ew-C and *-ey-C respectively and *-ew-éC and *-ey-éC passed to *-we-C and *-y-eC respectively. Since the sequence of vowel plus resonant and resonant plus vowel were both possible new sequences *-ew-eC and *-ey-eC could be easily analogically recreated.
*-ew-C > *-û-C
*-ey-C > *-î-C
*-éw-eC > *-ew-C
*-éy-eC > *-ey-C
*-ew-éC > *-w-eC
*-ey-éC > *-y-eC
Analogically restored sequences or 'contaminated' forms are *-ew-eC and *-ey-eC Thus the root form *bhe-w-to is represented by Indic bhû-tá and Baltic bû-ta. A restored form is represented by the Indic thematic present conjugation (3rd sg.) bhav-a-ti 'becomes, exists.'
A few of the words commonly brought together to illustrate the etymological root for 'to fear' in Indo-lranian, Baltic and Slavic are the following: Old Indic (3rd sg. present middle) bháy-a-te 'he fears' Avestan bayente, byente 'sie setzen in Furcht,' OCS bojati se, Lith. bijóti 'to fear.' Probably the apparent Avestan cognate should be dropped because Lommel, 1940, 11, compares it with Slavic biti 'to beat' and says that it can hardly have the meaning which we have given in the previous sentence. Mayrhofer, 1956, II, 471, supports Lommel's view. The Old Indic 3rd sg. perfect of this root is bibhaya from which a new Old Indic 3rd sg. active present bibhéti has been created, so that the latter form cannot be directly compared, as is so often done, with Old High German bibçt 'bebt, shivers, quakes' (Wackernagel, 1907, 305; Mayrhofer, 1956, II, 431).
In Old Indic the root is attested in various forms as far back as the Vedic hymns, e.g., (Book 1,154, 2) where Vishnu is compared with a 'dread beast' mrgó ná bhîmáh (Macdonell, 1917, 32). The 3rd pl. present tense of the verb 'to fear' is encountered, e.g., (book 1, 85, 8):
fear all creatures Maruts . . .
'All creatures fear the Maruts' (Macdonell, 1917, 27); (Book 2, 12, 13):
cid asya párvatâ
vehemence even his mountains are afraid . . .
'before his vehemence even the mountains are afraid' (Macdonell, 1917, 54).
The 3rd sg. perfect is encountered (book 5, 83, 2):
whole fears world him of the mighty weapon.
'The whole world fears him of the mighty weapon' (Macdonell, 1917, 105).
The 3rd pl. aorist is encountered (book 8, 48, 11):
off those have started ailments diseases
away have sped powers of darkness
ábhaisuh. have been affrighted.
'Those ailments have started off, diseases have sped away, the powers of darkness have been affrighted' (Macdonell, 1917, 161).
Several different ablaut grades of the root are represented in Indic, Baltic and Slavic. An etymological *bhey-C>*bhî-C is represented in Old Indic bhî-má- 'dreadful,' bhi-ra- 'terrifying,' bhî-rú 'timid, fearful,' bhî-ta- 'excessively terrified,' bhî-ti 'fear, alarm, dread,' bhî-'smá- 'frightful, dreadful,' Latvian bî-ties 'to be afraid,' 1st sg. present bî-stuos. The etymological *bhóy-e/oC = *bhoy- is represented in Baltic also by such nouns as Lith. bái-më, Latv. baî-me, 'fear, anxiety,' such adjectives as Lith. bai-lùs, Latv. baîls, baîlð 'timid,' such causative verbs as Lith. bai-dýti, Latv. baĩ-dìt 'to frighten.' Note also the adjectives Lith. bai-sùs, Latv. bai-ss 'frightful' which Fraenkel, 1955, 29, connects with Slavic bĕsb 'devil.'
When a plosive consonant plus resonant occurred in word-initial position in Indo-European it could have several possible sandhi developments. If in sentence sandhi the preceding syllable was short, i.e., it consisted of a single vowel, then the initial syllable of the following word would have the consonantal reflex of the resonant, thus, e.g., *-ĕ / bhye-; if the preceding syllable was long, i.e., it consisted of a long vowel or short vowel plus consonant, then the initial syllable of the following word would have the vocalic plus the consonantal reflex, e.g., *-ç (*-eC) / bhiye-. These doublets became phonemic and it is common for the prevocalic zero-grade of a root to be represented in the daughter languages by the vocalic plus the consonantal reflex of the resonant, thus, e.g., Lith. bij-óti 'to fear' (<*bhy-â-). Although the Old Indic nominative singular bhî-h could represent *bhey-s the gen. singular bhiy-áh may derive from *ghiy-és < *bhy-és. Another possibility is that the *-î- of the nom. sg. was analyzed phonologically as a sequence of *-ì- plus -y- in prevocalic position. I.e., *bhî-es was rendered phonetically as *bhîy-és. The form of the root showing the vocalic plus consonantal form of the resonant is commonly called the Sievers'-Law variant.
Discussing the various ablaut grades of this root Vasmer, 1953, 115, wrote: 'Die balt.-slav. Verbalformen sind zurück-zuführen auf Praesensst. *baie-, Praeteritalst. *biiâ-, Infinit. *bîtei mit verschiedenen Auslgeichungen.' In 1960 I proposed that the proto-Balto-Slavic form of the root had a zero-grade vocalism in the singular and an *-o-grade vocalism in the plural. One would compare the Old Indic 1st sg. active perfect véd-a 'I know' (<*void-) with the 1st pl. active perfect vid-má 'we know' (<*vid-). I no longer accept the schwa nor the laryngeal as I did at that time, but it would still be possible to posit an *-o-grade 1st sg. *bhoy-m > Slavic *bhoy-o (with loss of the old middle endings) vs. a 1st pl. *bhi-mé. In Slavic the *-o-grade vocalism was generalized throughout the paradigm and in Baltic the zero-grade vocalism was generalized. In addition in Baltic the suffix *-â- was added in order to reinforce the intransitive meaning so that *bhi-â-me replaced *bhi-me. Sievers' Law requires that an initial resonant be replaced by its vocalic plus its consonantal form, and the aspiration of the voiced aspirates was lost so that an original *bhi-â-me passed to *bij-â-me which is indeed represented in modern Lith. 1st pl. pres. bìj-o-me 'we fear.'
According to Burrow, 1965, 296, the Old Indic 3rd sg. pres. bhayate 'he becomes afraid' is to be considered a process as opposed to the 3rd sg. perfect bibhâya 'he is afraid' which is to be considered a state. Nevertheless a very interesting syntactic parallel can be established. The verb 'to fear' is construed with the ablative case in Old Indic, e.g., bhayate vrkât (ablative sg.) 'he is, becomes afraid of the wolf.' (Note that in the Vedic examples given above the object feared is also in the ablative case.) In Baltic and Slavic the verb 'to fear' is construed with the genitive case, so we find Russian boitsja volka which corresponds exactly to Lith. bìjo vĩlko 'he is afraid of the wolf.' Interestingly enough, however, the Blato-Slavic *o-stem genitive singular ending *-â (= Lith. -o) apparently derives from the Indo-European ablative ending *-ât (with loss of final *-t in Balto-Slavic).
It can be noted then that with regard to the concept 'fear' psychologists frequently mention 'avoidance behavoir,' which can be characterized as 'flight, hiding or evasion.' (See Thomson, 1979, 8 and Gray, 1971, 34.) I propose then that etymologically the stem *bhe/o-y- was merely an alternative form of the stem *bhe/o-w- 'to be,' but that when construed with the ablative case it originally denoted 'to be away from, to stay away from.' Originally the syntactic construction *bhe/o-y- *vlkw-ât denoted the avoidance behavoir, viz. 'stays away from the wolf.' Later the concrete meaning of the construction which described the act was replaced by the abstract meaning which described the feeling which accompanied the act. Thus the construction became concentrated in the verb itself, so that the form of the verb 'to be' with the element *-y, viz. *bhe/o-y took on the meaning 'to fear.'1
Buck, Carl Darling. 1949. A dictionary of selected synonyms in the principal
Indo-European languages. Chicago and London, University of Chicago
Burrow, t. 1965. The Sanskrit language. 2nd ed. Glasgow, The University Press.
Fraenkel, Ernst. 1955 ff. Litauisches etymologisches Wörterbuch. Heidelberg, Carl Winter; Göttingen, Vandenhoeck and Ruprecht.
Gray, Jeffrey. 1971. The psychology of fear and stress. New York, McGraw Hill.
Lommel, H. 1940. Yasna 34. KZ 67.6-26.
Macdonell, Arthur A. 1917. A Vedic reader for students. 1970 Seventh impression of 1951 India reprint used here. John Brown, Oxford University Press, Madras.
Mayrhofer, M. 1956 ff. Kurzfgefasstes etymologisches Wörterbuch des Altindischen. Vol. l-lll. Heidelberg, Carl Winter.
Schmalstieg, William R. 1960. OCS bojati sæ and Lith. bijóti. The Slavic and East European Journal 4 (New series;: 44-45.
1973. New thoughts on Indo-European phonology. KZ 87.99-157.
1980. Indo-European linguistics: A new synthesis. University Park and London, Penn State University Press.
Thomson, Robert. 1979. The concept of fear. In Fear in animals and man, ed. by W. Sluckin. New York, Van Nostrand Reinhold.
Vasmer, Max. 1953 ff. Russisches etymologisches Wörterbuch. Heidelberg, Carl Winter.
Wackernagel, J. 1907. Indisches und Italisches. KZ 41.305-318.
1 Carl Darling Buck, 1949, 1153, writes that words for 'fear' and the stronger emotion 'fright, terror' are mostly based upon those physical actions expressive of fear such as 'tremble, shake, flee, be struck,' etc. Thus Gk. phóbos originally meant 'flight' (still the only sense in Homer), hence 'panic, fright' and eventually the most common word for 'fear.' Thus the concrete meaning is usually at the basis of the later abstract meaning.