IN the case of this Upanishad we must first of all attempt to settle its right title. Professor Cowell, in his edition and translation of it, calls it Maitrî or Maitrâyanîya-upanishad, and states that it belongs to the Maitrâyanîya-sâkhâ of the Black Yagur-veda, and that it formed the concluding portion of a lost Brâhmana of that Sâkhâ, being preceded by the sacrificial (karma) portion, which consisted of four books.
In his MSS. the title varied between Maitry-upanishad and Maitrî-sâkhâ-upanishad. A Poona MS. calls it Maitrâyanîya-sâkhâ-upanishad, and a MS. copied for Baron von Eckstein, Maitrâyanîyopanishad. I myself in the Alphabetical List of the Upanishads, published in the journal of the German Oriental Society, called it, No. 104, Maitrâyana or Maitrî-upanishad, i.e. either the Upanishad of the Maitriyanas, or the Upanishad of Maitrî, the principal teacher.
In a MS. which I received from Dr. Burnell, the title of our Upanishad is Maitriyani-brâhmana-upanishad, varying with Maitriyani-brâhmana-upanishad, and Srîyagussâkhâyâm Maitrâyanîya-brâhmana-upanishad.
The next question is by what name this Upanishad is quoted by native authorities. Vidyâranya, in his Sarvopanishad-arthânubhûtiprakâsa, v. 1, speaks of the Maitrâyanîyanâmnî yâgushî sâkhâ, and he mentions Maitra (not Maitrî) as the author of that Sâkhâ. (vv. 55,150).
In the Muktikâ-upanishad we meet with the name of Maitrâyanî as the twenty-fourth Upanishad, with the name of Maitreyî as the twenty-ninth; and again, in the list of the sixteen Upanishads of the Sâma-veda, we find Maitrâyanâ and Maitreyî as the fourth and fifth.
Looking at all this evidence, I think we should come to the conclusion that our Upanishad derives its name from the Sâkhâ of the Maitrâyanas, and may therefore be called Maitrâyana-upanishad or Maitrâyanî Upanishad. Maitrâyana-brâhmana-upanishad seems likewise correct, and Maitriyani-brilimana-upanishad, like Kaushîtaki-brâhmana-upanishad and Vâgasaneyi-samhitopanishad, might be defended, if Maitrâyanin were known as a further derivative of Maitrâyana. If the name is formed from the teacher Maitrî or Maitra, the title of Maitrî-upanishad would also be correct, but I doubt whether Maitrî-upanishad would admit of any grammatical justification.
Besides this Maitrâyana-brâhmana-upanishad, however, I possess a MS. of what is called the Maitreyopanishad, sent to me likewise by the late Dr. Burnell. It is very short, and contains no more than the substance of the first Prapâthaka of the Maitrâyana-brâhmana-upanishad. I give the text of it, as far as it can be restored from the one MS. in my possession:
Harih Om. Brihadratho vai nâma râgâ vairâgye putram nidhâpayitvedam asâsvatam manyamânah sarîram vairâgyam upeto 'ranyam nirgagâma. Sa tatra paramam tapa âdityam udîkshamâna ûrdhvas tislithaty. Ante sahasrasya muner antikam âgagâma. Atha Brihadratho brahmavitpravaram munîndram sampûgya stutvâ bahusah pranâmam akarot. So 'bravîd agnir ivâdhûmakas tegasâ nirdahann ivâtmavid Bhagavâñ khâkâyanya, uttishthottishtha varam vrinîshveti râgânam abravît. Sa tasmai punar namaskrityovâka, Bhagavan nâ(ha)mâtmavit tvam tattvavik khusrumo vayam; sa tvam no brûhity etad vratam purastâd asakyam mâ prikkha prasñam Aikshvâkânyân kâmân vrinîshveti Sâkâyanyah. Sarîrasya sarîre (sic) karanâv abhimrisyamâno râgemâm gâthâm gagâda.
Bhagavann, asthikarmasnâyumaggâmâmsasuklasonitasreshmâsrudashikâvinmûtrapittakaphasamghâte durgandhe nihsâre 'smiñ kharire kim kâmabhogaih.
Kâmakrodhalobhamohabhayavishâdersheshtaviyogânishtasamprayogakshutpipâsâgarâmrityurogasokâdyair abhihate 'smiñ kharire kim kâmabhogaih.
Sarvam kedam kshayishnu pasyâmo yatheme damsamasakâdayas trinavan  nasyata yodbhûtapradhvamsinah.
Atha kim etair vâ pare 'nye dhamartharâs (sic) kakravartinah Sudyumnabhûridyumnakuvalayâsvayauvanâsvavaddhriyâsvâsvapatih sasabindur hariskandro 'mbarisho nanukastvayâtir yayâtir anaranyokshasenâdayo marutabharataprabhritayo râgâno mishato bandhuvargasya mahatîm sriyam tyaktvâsmâl lokâd amum lokam prayânti.
Atha kim etair vâ pare 'nye gandharvâsurayaksharâkshasabhûtaganapisâkoragrahâdinâm nirodhanam pasyâmah.
Atha kim etair vânyanâm soshanam mahârnavânâm
sikharinâm prapatanam dhruvasya prakalanam vâtarûnâm nimagganam prithivyâh sthânâpasaranam surânâm. So 'ham ity etadvidhe 'smin samsâre kim kâmopabhogair yair evâsritasya sakrid âvartanam drisyata ity uddhartum arhasi tyandodapânabheka ivâham asmin sam Bhagavas tvam gatis tvam no gatir iti.
Ayam agnir vaisvânaro yo 'yam antah purushe yenedam annam pakyate yad idam adyate tasyaisha ghosho bhavati yam etat karnâv apidhâya srinoti, sa yadotkramishyan bhavati nainam ghosham srinoti.
Sa sivah so 'nte vaisvânaro bhûtvâ sa dagdhvâ sarvâni bhûtâni prithivyapsu pralîyate, âpas tegasi lîyante, tego vâyau pralîyate, vâyur âkâse vilîyate, âkâsam indriyeshv, indriyâni tanmâtreshu, tanmâtrâni bhûtâdau vilîyante, bhûtâdi mahati vilîyate, mahân avyakte vilîyate, avyaktam akshare vilîyate, aksharam tamasi viliyate, tama ekibhavati parasmin, parastân na san nâsan na sad ityetan nirvânam anusâsanam iti vedânusâsanam.
We should distinguish therefore between the large Maitrâyana-brâhmana-upanishad and the smaller Maitreyopanishad. The title of Maitreyî-brâhmana has, of course, a totally different origin, and simply means the Brâhmana which tells the story of Maitreyî.
As Professor Cowell, in the Preface to his edition and translation of the Maitrâyana-brâhmana-upanishad, has discussed its peculiar character, I have little to add on that subject. I agree with him in thinking that this Upanishad has grown, and contains several accretions. The Sanskrit commentator himself declares the sixth and seventh chapters to be Khilas or supplementary. Possibly the Maitreya-upanishad, as printed above, contains the earliest framework. Then we have traces of various recensions. Professor Cowell (Preface, p. vi) mentions a MS., copied for Baron Eckstein, apparently from a Telugu original, which contains the first five chapters only, numbered as four. The verses given in VI, 34 (p. 177), beginning with 'atreme slokâ bhavanti, are placed after IV, 3. In my own MS. these verses are inserted at the beginning of the fifth chapter. Then follows in Baron Eckstein's MS. as IV, 5, what is given in the printed text as V, 1, 2 (pp. 69-76). In my own MS., which likewise comes from the South, the Upanishad does not go beyond VI, 8, which is called the sixth chapter and the end of the Upanishad.
We have in fact in our Upanishad the first specimen of that peculiar Indian style, so common in the later fables and stories, which delights in enclosing one story within another. The kernel of our Upanishad is really the dialogue between the Vâlakhilyas and Pragâpati Kratu. This is called by the commentator (see p. 331, note) a Vyâkhyâna, i.e. a fuller explanation of the Sûtra which comes before, and which expresses in the few words, “He is the Self, this is the immortal, the fearless, this is Brahman,” the gist of the whole Upanishad.
This dialogue, or at all events the doctrine which it was meant to illustrate, was communicated by Maitrî (or Maitra) to Sâkâyanya, and by Sâkâyanya to King Brihadratha Aikshvâka, also called Marut (II, 1; VI, 30). This dialogue might seem to come to an end in VI, 29, and likewise the dialogue between Sâkâyanya and Brihadratha; but it is carried on again to the end of VI, 30, and followed afterwards by a number of paragraphs which may probably be considered as later additions.
But though admitting all this, I cannot bring myself to follow Professor Cowell in considering, as he does, even the earlier portion of the Upanishad as dating from a late period, while the latter portions are called by him comparatively modern, on account of frequent Vaishnava quotations. What imparts to this Upanishad, according to my opinion, an exceptionally genuine and ancient character, is the preservation in it of that peculiar Sandhi which, thanks to the labours of Dr. von Schroeder, we now know to be characteristic of the Maitrâyana-sâkhâ. In that Sâkhâ final unaccented as and e are changed into â, if the next word begins with an accented vowel, except a. Before initial a, however, e remains unchanged, and as becomes o, and the initial a is sometimes elided, sometimes not. Some of these rules, it must be remembered, run counter to Pânini, and we may safely conclude therefore that texts in which they are observed, date from the time before Pânini. In some MSS., as, for instance, in my own MS. of the Maitrâyanabrâhmana-upanishad, these rules are not observed, but this makes their strict observation in other MSS. all the more important. Besides, though to Dr. von Schroeder belongs, no doubt, the credit of having, in his edition of the Maitrâyanî Samhitâ, first pointed out these phonetic peculiarities, they were known as such to the commentators, who expressly point out these irregular Sandhis as distinctive of the Maitrâyanî sâkhâ. Thus we read Maitr. Up. II, 3 (p. 18), that tigmategasâ ûrdhvaretaso, instead of tigmategasa, is evamvidha etakkhâkhâsanketapâthas khândasah sarvatra, i.e. is throughout theVedic reading indicatory of that particular Sâkhâ, namely the Maitrâyanî.
A still stranger peculiarity of our Sâkhâ is the change of a final t before initial s into ñ. This also occurs in our Upanishad. In VI, 8, we read svâñ sarîrâd; in VI, 2 7, yañ sarîrasya. Such a change seems phonetically so unnatural, that the tradition must have been very strong to perpetuate it among the Maitrâyanas.
Now what is important for our purposes is this, that these phonetic peculiarities run through all the seven chapters of our Upanishad. This will be seen from the following list:
Final as changed into â before initial vowel:
II, 3, tigmategasâ ûrdhvaretaso (Comm. etakkhâkhâsanketapâthas khândasah sarvatra). II, 5, vibodhâ evam. II, 7, avasthitâ iti. III, 5, etair abhibhûtâ îti. IV, i, vidyatâ iti. VI, 4, pranavâ iti; bhâmyâdayâ eko. VI, 6, âdityl iti; âhavanîyâ iti; sûryâ iti; ahankârâ iti; vyânâ iti. VI, 7, bhargâ iti. VI, 7, sannivishtâ iti. VI, 23, devâ onkâro. VI, 30, prâyâtâ iti. VI, 30, vinirgatâ iti.
Final e before initial vowels becomes â. For instance:
I, 4, drisyatâ iti. II, 2, nishpadyatâ iti. III, 2, âpadyatâ iti. III, 2, pushkarâ iti. IV, i, vidyatâ iti. VI, 10, bhunktâ iti. VI, 20, asnutâ iti. VI, 30, ekâ âhur.
Even pragrihya e is changed to â in:
VI, 23, etâ upâsita, i.e. ete uktalakshane brahmanî.
In VI, 31, instead of te etasya, the commentator seems to have read te vâ etasya.
Final as before â, u, and au becomes a, and is then contracted. For instance:
I, 4, vanaspatayodbhûta, instead of vanaspataya, udbhûta. (Comm. Sandhis khândaso vâ, ukâro vâtra lupto drashtavyah.) II, 6, devaushnyam, instead of deva aushnyam. (Comm. Sandhis khândasah.) VI, 24, atamâvishtam, instead of atama-âvishtam (Comm. Sandhis khândasah); cf. Khând. Up. VI, 8, 3, asanâyeti (Comm. visarganîyalopah).
Final e before i becomes a, and is then contracted. For instance:
VI, 7, itmâ ganîted for ganita iti. (Comm. gânite, gânâti.) VI, 28, avataiva for avata iva. (Comm. Sandhivriddhi khândase.)
Final au before initial vowels becomes â. For instance:
II, 6, yena vâ etâ anugrihitâ iti. VI, 22, asâ abhidhyâtâ.
On abhibhûyamânay iva, see p. 295, note 2.
V, 2, asâ âtmâ (var. lect. asâv âtmâ).
Final o of atho produces elision of initial short a. For instance:
III, 2, atho 'bhibhûatvât. (Comm. Sandhis khândasah.) Various reading, ato 'bhibhûtatvât. VI, 1, so antar is explained as sa u.
VI, 7, âpo pyâyanât, explained by pyâyanât and âpyâyanât. Might it be, âpo 'py ayanât? VI, 7, âtmano tmâ netâ. II, 6, so tmânam abhidhyâtvâ. VI, 35, dvidharmondharn for dvidharmândham. (Comm. khândasa.) VI, 35, tegasendham, i. c. tegasâ-iddhan. (In explaining other irregular compounds, too, as in I, 4, the commentator has recourse to a khândasa or prâmâdika licence.) VI, 1, hiranyavasthât for hiranyâvasthât. Here the dropping of a in avasthât is explained by a reference to Bhâguri (vashti Bhâgurir allopam avâpyor upasargayoh). See Vopadeva III, 171.
VII, 2, brahmadhiyâlambana. (Comm. vislishtapâthas khândasah.) VI, 35, apyay ankurâ for apy ankurâ. (Comm. yakârah pramâdapathitah.)
On the contrary VI, 35, vliyânte for viliyante.
If on the grounds which we have hitherto. examined there seems good reason to ascribe the Maitrâyana-brâhmana-upanishad to an early rather than to a late period, possibly to an ante-Pâninean period, we shall hardly be persuaded to change this opinion on account of supposed references to Vaishnava or to Bauddha doctrines which some scholars have tried to discover in it.
As to the worship of Vishnu, as one of the many manifestations of the Highest Spirit, we have seen it alluded to in other Upanishads, and we know from the Brâhmanas that the name of Vishnu was connected with many of the earliest Vedic sacrifices.
As to Bauddha doctrines, including the very name of Nirvâna (p. xlvi, 1. 19), we must remember, as I have often remarked, that there were Bauddhas before Buddha. Brihaspati, who is frequently quoted in later philosophical writings as the author of an heretical philosophy, denying the authority of the Vedas, is mentioned by name in our Upanishad (VII, 9), but we are told that this Brihaspati, having become Sukra, promulgated his erroneous doctrines in order to mislead the Asuras, and thus to insure the safety of Indra, i.e. of the old faith.
The fact that the teacher of King Briliadratha in our Upanishad is called Sâkâyanya, can never be used in support of the idea that, being a descendant of Sâka, he must have been, like Sâkyamuni, a teacher of Buddhist doctrines. He is the very opposite in our Upanishad, and warns his hearers against such doctrines as we should identify with the doctrines of Buddha. As I have pointed out on several occasions, the breaking through the law of the Âsramas is the chief complaint which orthodox Brâhmans make against Buddhists and their predecessors, and this is what Sâkâyanya condemns. A Brâhman may become a Sannyâsin, which is much the same as a Buddhist Bhikshu, if he has first passed through the three stages of a student, a householder, and a Vânaprastha. But to become a Bhikshu without that previous discipline, was heresy in the eyes of the Brâhmans, and it was exactly that heresy which the Bauddhas preached and practised. That this social laxity was gaining ground at the time when our Upanishad was written is clear (see VII, 8). We hear of people who wear red dresses (like the Buddhists) without having a right to them; we even hear of books, different from the Vedas, against which the true Brâhmans are warned. All this points to times when what we call Buddhism was in the air, say the sixth century B.C., the very time to which I have always assigned the origin of the genuine and classical Upanishads. The Upanishads are to my mind the germs of Buddhism, while Buddhism is in many respects the doctrine of the Upanishads carried out to its last consequences, and, what is important, employed as the foundation of a new social system. In doctrine the highest goal of the Vedânta, the knowledge of the true Self, is no more than the Buddhist Samyaksambodhi; in practice the Sannyâsin is the Bhikshu, the friar, only emancipated alike from the tedious discipline of the Brâhmanic student, the duties of the Brâhmanic householder, and the yoke of useless penances imposed on the Brâhmanic dweller in the forest. The spiritual freedom of the Sannyâsin becomes in Buddhism the common property of the Sangha, the Fraternity, and that Fraternity is open alike to the young and the old, to the Brâhman and the Sûdra, to the rich and the poor, to the wise and the foolish. In fact there is no break between the India of the Veda and the India of the Tripitaka, but there is an historical continuity between the two, and the connecting link between extremes that seem widely separated must be sought in the Upanishads.
F. Max Müller.
Oxford, February, 1884.
 See Cowell, Maitr: Up. pref. p. iv.
 Calcutta, 1791 (1869), p. 4; also as quoted in the Mahâvâkya-ratnâvalî, p.2b.
 Dr. Burnell, in his Tanjore Catalogue, mentions, p. 35a, a Maitrâyanî-brâhmanopanishad, which can hardly be a right title, and p. 36b a Maitrâyanîya and Maitreyîbrâhmana.
 One expects âsthâya.
 This seems better than the Maitrâyana text. He went near a Muni, viz. Sâkiyanya.
 This seems unnecessary.
 There may be an older reading hidden in this, from which arose the reading of the Maitrayana B. U. trinavanaspatayodbhûtapradhvamsinah, or yo bhûtapradhvainsinah.
 Maitr. Up. II, 6; p. 32.
 kramishyân, m.
 Yadhâ, m.
 Maitr. Up. VI, 34; p. 178.
 See Khand. Up. p. 623.
 See p. 303, note 1; p. 305. note 1; p. 312, note 1.
 I have left out the restriction as to the accent of the vowels, because they are disregarded in the Upanishad. It should be observed that this peculiar Sandhi occurs in the Upamishad chiefly before iti.
 Sâkâyanya means a grandson or further descendant of Sâka; see Gauaratnâvalî (Baroda, 1874), p. 57a.
 As there is room left on this page, I subjoin a passage from the Abhidharma-kosha-vyâkhyi, ascribed to the Bhagavat, but which, as far as style and thought are concerned, might be taken from an Upanishad: Uktam hi Bhagavatâ: Prithivî bho Gautama kutra pratishikitâ? Prithivî Brâhmana abmandale pratishthitâ. Abmandalam bho Gautama kva pratishthitam? Vâyau pratishthitam. Vâyur bho Gautama kva pratishthitah? Âkâse pratishthitah. Âkâsam bho Gautama kutra pratishthitam? Atisarasi Mahâbrâhmana, atisarasi Mahâbrâhmana. Âkâsam Brâhmanâpratishthitam, anâlambanam iti vistarah. Tasmâd asty âkâsam iti Vaibhâshikâh. (See Brihad-Âr. Up. III, 6, 1. Burnouf, Introduction à l'histoire du Buddhisme, p. 449.)
“For it is said by the Bhagavat: ‘O Gautama, on what does the earth rest?’ ‘The earth, O Brâhmana, rests on the sphere of water.’ ‘O Gautama, on what does the sphere of water rest?’ ‘It rests on the air." "O Gautama,on what does the air rest?’ ‘It rests on the ether (âkâsa)." "O Gautama,on what does the ether rest?’ ‘Thou goest too far, great Brâhmana; thou goest too far, great Brahmana. The ether, O Brâhmana, does not rest. It has no support.’ Therefore the Vaibhâshikas hold that there is an ether,” &c.