*septm̥

The Luwian numeral šap(pa)tam(m)- can be recovered from various derivatives and its association with Indo-European *septm̥ is obvious. All derivatives show an -m(m)-, which is likely to represent the outcome of the (vocalized) nasal in this position; cf. CLuw. tummant- n. ‘ear’ /tḿ̥mant-/ or /túmmant-/ (cf. Melchert 1994a:74). Contra Melchert 1994a:277, there is no lenition of the second stop in šappatammimma/i- (belonging to a denominative šappatammi-^(ti) equating Hitt. šiptamiye/a-^(mi) (< PAnat. *septm̥-i̯é/ó-), whose spelling simply expresses the sequence /sapta-/ (or /sapp_əta-/); see Sasseville 2021a:261f. for new evidence against Melchert’s putative sound law of secondary lenition after anaptyxis. Thus, the Proto-Anatolian numeral for ‘seven’ can be reconstructed based on Hittite šiptamiye/a-^(mi) The root vowel -i- in Hittite is perfectly regular as it stands for an unaccented *e, i.e. *septm̥-i̯é/ó-; see Melchert 1994a:139, contra Eichner 1992a:84f.and Luwian evidence, i.e. *septm̥. The ending of šaptame/inzu can be interpreted as an original locative plural in -su (with epenthesis after n), i.e. /-ntsu/ < */-n-su/ (cf. PLuw. *trisu). The -n- likely comes from the locative plural of an n-stem, which would have been reanalyzed as part of the ending. Therefore, šaptame/inzu can be reconstructed as *septm̥-nsu (cf. 8-in-zu). The -e/i- would thus simply stand for an epenthetic vowel. However, for the complexity of the context in which šaptame/inzu is attested and its possible substantivized status, see under lemma.

[D.S.]

PIE *septḿ̥ means ‘seven’ (cf. IEW:909, EIEC:402, Watkins 2011a:78).

The PIE cardinal number for ‘seven’ can be reconstructed as *septḿ̥ (IEW:909, EIEC:402, Watkins 2011a:78). Direct continuants of this form are PIIr. *saptá (> Ved. saptá, YAv. hapta), Gk. ἑπτά, Arm. ewtʻn, Lat. septem, and PCelt. *sextam (> OIr. secht, MW seith, etc.); cf. also PAnat. *septm̥, which lives on in Luw. šap(pa)tam(m)° and Hitt. šiptam-iye/a-^(mi) (see sub PAnat. Reconstruction). The forms of the other IE languages either reflect some kind of analogical reshaping (e.g., TochB ṣukt ‘7’, presumably reshaped after TochB okt ‘8’) or continue extended stem forms (e.g., Alb. shtatë < *septḿ̥-tā-, as per Matzinger 2006a:115). For further information, cf. the literature cited in the IE Transmission above.

[T.S.]

Ved. saptá ‘seven’ (RV+); YAv. hapta ‘seven’Cf., e.g., EWAia II:700.

Gk. ἑπτά ‘seven’ (Il.+)Cf., e.g., GEW I:545, EDG:448.

Arm. ewtʻn ‘seven’ (Bible+)Cf., e.g., EDAIL:270f.

Lat. septem ‘seven’Cf., e.g., EDLIL:555.

OIr. secht, MW seith, etc.Cf., e.g., EDPC:332.

Goth. sibun ‘seven’, ON sjau ‘id.’, OE seofon, syfon ‘id.’, OS siƀun, siƀon ‘id.’, OHG sibun ‘id.’, etc.Cf., e.g., EDPG:429.

Lith. septynì nom. pl. ‘seven’, Latv. septiņi nom. pl. ‘id.’Cf., e.g., ALEW II:902f., EDBIL:393f.

OCS sedmĭ ‘seven’, Russ. sem’ ‘id.’, Cz. sedm ‘id.’, SCr. sȅdam ‘id.’, etc.Cf., e.g., EDSIL:443.

TochA ṣpät ‘seven’, TochB ṣukt ‘id.’Cf., e.g., DTB²:720.

Alb. shtatë ‘seven’Cf., e.g., Matzinger 2006a:115.