As of June 2025, Eliotweb includes the unproofread texts of the following:
In addition to Eliot's and Mayhew's Bible translations, transcripts of the copious colonial-era secondary religious literature—referred to somewhat whimsically as "Mishnaic Massachusett"—are being added at irregular intervals. As of June 2025, Eliotweb also includes the following texts:
The mishnaic texts not by Mayhew all have the shorthand α in the text search for the time being; those by Mayhew have the shorthand M.
Many characters in the transcripts are bracketed, as w[o]sketomp (this can be toggled on and off in the text browser). This may mean a number of things, such as a suspected typo, missing diacritic or letter of uncertain identity. [a] and [o], in particular, most often indicate that there is a diacritic or suspected diacritic whose identity is not entirely certain. In many cases brackets are little more than a shorthand to myself to go back and triple-check the letter during future proofreading.
The "double-o" ligature ꝏ is displayed with a macron, as ꝏ̄, although no such macron is used in the texts (I have found no token of the ligature with any diacritic whatsoever and, given the constraints of 17th-century printing, it seems almost certain that none exists). This orthographic choice was originally motivated by my own misrecollection of its usage in the 1829 published version of Josiah Cotton's Massachusett Vocabulary. Upon taking another look when writing this page, it became clear that it is only used sparingly (and with unclear meaning). Nevertheless, the macron makes it easier to distinguish the ligature from double oo and, in my experience, makes the text easier on the eyes, so I have decided to keep it.
It is worth repeating that the texts have not been proofread; for the time being, Eliotweb should be used as a finding aid rather than an index of record. However, while typos are very common (I seem to find new ones every time I browse the texts), the transcripts should be accurate enough to look for interesting hapaxes or a general sense of word frequency. If you need to cite a source I have copied from scans or pictures that are not otherwise available, please feel free to email me at campbell dot nilsen at gmail dot com and I will double-check the text for you.
The texts included on Eliotweb are written, to varying degrees of competence, in Massachusett, an Eastern Algonquian language spoken until the mid-19th century across the present-day Boston metro area, Cape Cod, Martha's Vineyard and Nantucket. A sketch of the phonology and grammar of the language is envisioned for the website in the nearish future; the description in Part II of Native Writings in Massachusett (Goddard and Bragdon, 1988; hereafter NWiM) is excellent. For an overview of the historical phonology of Massachusett and its place within the larger Algonquian family, see Ives Goddard's 1981 article Massachusett Phonology: a Preliminary Look and David Costa's 2007 article The Dialectology of Southern New England Algonquian.
All the documents on Eliotweb (for the time being) are religious in nature, and none have indigenous authors, though one included author (Experience Mayhew) spoke the language natively due to his upbringing as the son of the white minister in an indigenous village on Martha's Vineyard. This is due in large part to the scholarly space of undone work: most of the easily-accessible documents written by Native authors have already been collected and analyzed in NWiM, though there are probably quite a few overlooked examples still lurking in archives and libraries in eastern New England. However, because the majority of documents in NWiM are legal documents such as deeds, wills and petitions, the range of vocabulary found in them is somewhat limited and a great many words and some grammatical forms are only known to us from the religious literature, such as uppasq 'leech' or nuttꝏ̄unnussog '(my) kidneys'.
Except for Mayhew's translations, most if not all of the documents were written with the help of indigenous assistants, most of whose names are unfortunately lost to us. Dialectal differences, most notably between a mainland dialect (spoken in the Boston area and Cape Cod) and an island dialect spoken on Nantucket and the Vineyard, are occasionally visible to the trained eye. For example, the mainland dialect distinguishes a palatalized coronal stop /tʸ/ (usually written t or teV before another vowel, and deriving from Proto-Algonquian *k before long *e· or metrically weakened short *i) from a postalveolar affricate /č/, consistently written ch, deriving from PA *č. The two undergo a merger or near-merger in the Island dialect; compare Eliot's translation of Nehemiah 13:17 with Mayhew's citation of the same verse in his translation of Cotton Mather's Lord's Day sermon:
More exhaustive analysis may turn up other dialectal distinctions or even, especially in Eliot's translation of the Bible, the detectable influence of multiple informants (we know he worked with at least four native speakers). However, where dialectal and lexical differences are visible, they are often subtle and rarely clean-cut, with different layers appearing from verse to verse and even within the same verse. One curious example is found in the numeral 'seven'. The usual mainland word for 'seven' was nesausuk /ni·sã·sək/, from PA *nyi·šwa·šiki, but an etymologically obscure form enadtuh (spelled variously; phonemically perhaps /i·na·tã·(h)/), attested in several native documents from Martha's Vineyard and Nantucket (but never in Mayhew), is found in a few places in Eliot's Ezekiel, Jeremiah and Zechariah. Zechariah 4:2 shows both nesausuk and enadtuh:
| First Edition (α) | Second Edition (β) | KJV |
|---|---|---|
| Kah nuttinuk, teagwas kenaum? kah nussip, nunnatauwompip, kah kusseh golde wequananteganuhtug, kah ꝏ̄nauag ohteau woskeche, kah nesausuk tahshinash lampsash naut, kah nesausuk tahshinash papómpuhkagish en enadtuh tahshe lampsut, nish ohtagish woskeche. | Kah nuttinuk, teagwas kenaum? kah nussin, nunnattauwompip, kah kusseh golde wequananteganuhtug, kah ꝏ̄nauag ohteau woskeche, kah nesausuk tahshinash lambsash naut, kah nesausuk tahshinash papómpuhkágish en enadtuh tahshe lambsut, nish ohtagish woskeche. | And said unto me, What seest thou? And I said, I have looked, and behold a candlestick all of gold, with a bowl upon the top of it, and his seven lamps thereon, and seven pipes to the seven lamps, which are upon the top thereof: |
What motivated Eliot's use of two forms within a single verse is unknown (it cannot be an attempt to calque the Hebrew, which has שִׁבְעָה throughout) and probably unknowable. Inconsistencies appear in other authors as well: Mayhew's 1707 translation of Cotton Mather's Lord's Day sermon generally shows the Island-dialect merger of /tʸ/ and /č/ as ch, but his 1709 edition of the Psalms and Gospel of John (the Massachusett Psalter) distinguishes them much more consistently. The maintenance of the distinction in the latter is probably due to the influence of Eliot's translation of the Bible, which despite its awkwardness seems to have been taken as 'standard Massachusett', as is often the case with Bible translations in societies with little or no prior written tradition. Eliot's influence can be seen even in some of the native documents collated in NWiM and is also visible in certain other publications: for example, in Eliot's text, there are about three tokens of chaguas 'what' for every seven of teaguas. We accordingly find chagua consistently used to translate 'what' in Grindal Rawson's Milk for Babes, though he otherwise does not display the merger and would not be expected to (he was the minister in Mendon, about thirty miles southwest of Boston, and as far as we know would not have picked up Island dialect forms organically).
There are almost certainly grammatical differences among the various texts that reflect real dialectal variation, though these are obscured by the fact that none of the texts in question were written by native speakers, except for Mayhew, and even Mayhew is prone to influence from English. There is at least one aspect in which Eliot is probably a more faithful recorder of actual usage than Mayhew: he uses the preterite in -p- sparingly, as the native documents do, while Mayhew's translation of the Psalms and Gospel of John nearly always uses the Massachusett preterite to translate an English past tense, as at John 6:2.
| First Edition (α) | Second Edition (β) | Mayhew (M) | KJV |
|---|---|---|---|
| Kah muttaanukeg wutasukauouh, newutche naumwog monchanatamꝏ̄ongash nish asitcheh naut machinanutcheh. | Kah muttaanukeg wutasukauouh, newutche naumwog monchanátamꝏ̄ongash nish asitcheh naut mahchinnanutcheh. | Kah Muttaanwog assuhkunkup, newutche namuppaneg ummuhchantamꝏ̄onkash, nish asehpash ut mahchinanitcheh. | And a great multitude followed him, because they saw his miracles which he did on them that were diseased. |
There are additional discrepencies and outright oddities in the texts. Cotton Mather's sermon Family Religion Excited and Assisted was published with a facing-page translation in 1714 at Bartholomew Green's printing house in Boston. The translator is unknown, but Thomas James Holmes postulates that it was Mayhew. The published version is consistent (at least at a skim) in distinguishing /tʸ/ from /č/, but the translator uses the first-person plural inclusive forms fairly competently and has a solid grasp of incorporated medials in the verbal complex, so Holmes's conjecture seems plausible. However, a handwritten version of the first part of the sermon dated to the 1720s found in the Experience Mayhew Papers at the Massachusett Historical Society shows a partial switch to exclusive forms, as in the first sentence of section I.3:
| Published Edition (1714) | Mayhew, handwritten (1720s) | English |
|---|---|---|
| OOwequaiyeum wussuhquohhosuongash ne piuhsukke koowotumonkquinnan kunnooswetamooonganun yeu ut. | Oowequaiyeum Wussuhquohhosuongash ne piuhsuhke noowotumonkquinnan nunnooswetamooonganun yeu ut | The Light of Scripture doth expresly inform us of our Duty in this matter. |
The handwriting is clearly Mayhew's, and the handwritten sermon otherwise follows the published version fairly closely, but the switch to exclusive forms is difficult to explain, especially as it is not consistent (contrast handwritten 2.1 Unnuppomantamooongane Wequai ne kukkuhkootumongqunan 'the very Light of Nature doth teach us', unchanged from the published version save for minor spelling differences). Moreover, the published version does not always show the inclusive where we would expect it; compare I.3:7:
| Published Edition (1714) | Mayhew, handwritten (1720s) | English |
|---|---|---|
| Wussuhquohwhosuongash nuttukquinnanonash, nash Teashshinninneongash matcheseae anakausitcheg uttuh matta wequttumauonkik Godoh. | Wussuhquohwhosuongash nuttukquinnanonash teashshinninneongash matcheseae, anakausitcheg uttuh matta wehuttumauongik Godoh | The Scriptures do tell us, That those Families are the Workers of Iniquity which Call not upon God. |
Further investigation of such details is warranted but may not necessarily give satisfying answers.
Eliot's and Mayhew's translations are generally based on the King James Version, or Authorized Version (abbreviated on this site as KJV), so this is what is given as the parallel text; I have used Project Gutenberg's transcript, which may be found here. Somewhat surprisingly, this was by no means unusual in the context of Puritan New England, whose clergymen regularly used the KJV in addition to the nonconformist Geneva Bible and their own translations from Hebrew or Greek.
There are a number of places where Eliot and/or Mayhew have drawn from the Hebrew of Ben Hayyim's Masoretic Text or the Greek of the Textus Receptus, as at Leviticus 24:12:
| First Edition (α) | Second Edition (β) | KJV | Hebrew |
|---|---|---|---|
| Kah ukkupshagkinóuh, woh noh ꝏ̄wahteauwaheónat nashpe wuttꝏ̄n Jehovah. | Kah ukkupshagkinóuh, woh noh ꝏ̄wahteauwaheónat nashpe wuttꝏ̄n Jehovah. | And they put him in ward, that the mind of the LORD might be shewed them. | וַיַּנִּיחֻהוּ בַּמִּשְׁמָר לִפְרֹשׁ לָהֶם עַל־פִּי יְהוָה׃ |
Eliot in particular is somewhat notable for calques on Hebrew syntax and idiom in addition to vocabulary choice (I have yet to check Mayhew's Psalms thoroughly on this point); Genesis 2:17 provides an interesting example.
| Zeroth Edition (א) | First Edition (α) | Second Edition (β) | KJV | Hebrew |
|---|---|---|---|---|
| Qut ne mahtug woh wahteòmuk wanègik & machuk, matta pish kummeechꝏ̄un, newutche ne kesukok meechꝏ̄an nuppꝏ̄an pish kenup. | Qut ne mahtug nashpe woh wahteómuk wanégik kah machuk, matta woh kummeechꝏ̄un, newutche ne kesukok meechꝏ̄an nuppꝏ̄e pish kenup. | Qut ne mahtug nashpe woh wahteómuk wanegik kah machuk matta woh kummeechꝏ̄un, newutche ne kesukok meechꝏ̄an nuppꝏ̄e pish kenup | But of the tree of the knowledge of good and evil, thou shalt not eat of it: for in the day that thou eatest thereof thou shalt surely die. | וּמֵעֵץ, הַדַּעַת טוֹב וָרָע--לֹא תֹאכַל, מִמֶּנּוּ: כִּי, בְּיוֹם אֲכָלְךָ מִמֶּנּוּ--מוֹת תָּמוּת. |
John 1:43 provides an example of a verse where Eliot has translated from the KJV (with would left untranslated, as much modal and tense marking in English tends to be), but Mayhew has drawn on the Greek of the Textus Receptus:
| First Edition (α) | Second Edition (β) | Mayhew (M) | KJV | Greek |
|---|---|---|---|---|
| Na wonk kesukok, Jesus sohham en Galile, kah naméheau Philip, kah wuttinuh, asuhkah. | Na wonk kesukok, Jesus sohham en Galile, kah nameheau Phillip, kah wuttinuh, Asuhkah. | Na wonk kasukohk, Jesus kesanumꝏ̄p suhhomunnát en Galilee, kah nameheau Philipoh, kah wuttunuh ohsuhkah. | The day following Jesus would go forth into Galilee, and findeth Philip, and saith unto him, Follow me. | Τῇ ἐπαύριον ἠθέλησεν ὁ Ἰησοῦς ἐξελθεῖν εἰς τὴν Γαλιλαίαν· καὶ εὑρίσκει Φίλιππον, καὶ λέγει αὐτῷ, Ἀκολούθει μοι. |
As of June 2025, neither Greek nor Hebrew are included on Eliotweb for comparison. The relative obscurity of the Textus Receptus is something of a frustration for Greek (there are few text files or easily-scraped web versions available), while for Hebrew the main barrier is the need to convert verse numbers to the Christian ordering and handle qere/ketiv readings in the HTML.