RESEARCH ARTICLE
Lexical phylogene ics o he Tupı
´
-Gua anı
´
amily: Language, a chaeology, and he
p oblem o ch onology
Fab ı
´cio Fe az Ge a di
1
*, Tiago T esoldiID
2
, Ca olina Coelho A agon
3
,
S anisla Reiche
1
, Jonas G ego io de Souza
4
, F ancisco Sil a NoelliID
5
1S S, Ebe ha d Ka ls Uni e si a
¨ Tu¨bingen, Tu¨bingen, Ge many, 2Depa men o Linguis ics and Philology,
Uppsala Uni e si e , Uppsala, Sweden, 3DLPL, Uni e sidade Fede al da Pa aı
´ba, João Pessoa, B azil,
4Depa men o Humani ies, Uni e si a Pompeu Fab a, Ba celona, Spain, 5Cen o de A queologia,
Uni e sidade de Lisboa, Lisboa, Po ugal
* ab icio.ge a di@uni- uebingen.de
Abs ac
Tupı
´-Gua anı
´is one o he la ges b anches o he Tupı
´an language amily, bu despi e i s
ele ance he e is no consensus abou i s o igins in e ms o age, homeland, and expansion.
Linguis ic classi ica ions a y signi ican ly, wi h a chaeological s udies sugges ing incompa -
ible da e anges while e hnog aphic li e a u e con i ms he close simila i ies as a esul o
con inuous in e - amily con ac . To in es iga e his issue, we use a linguis ic da abase o
cogna e da a, employing Bayesian phylogene ic me hods o in e a da ed ee and o build a
phylogeog aphic expansion model. Resul s sugges ha he b anch o igina ed a ound 2500
BP in he a ea o he uppe cou se o he Tapajo
´s-Xingu basins, wi h a spli be ween Sou h-
e n and No he n a ie ies beginning a ound 1750 BP. We analyse he di icul ies in econcil-
ing a chaeological and linguis ic da a o his g oup, s essing he impo ance o de eloping
an in e disciplina y uni ied model ha inco po a es e idence om bo h disciplines.
1 In oduc ion
The p oblem o es ablishing he in e nal ela ions and ch onology o he Tupı
´-Gua anı
´lan-
guage amily (hence o h TG) has been a long-s anding one. Ideally, he e should be a uni ied
model explaining he language expansion and inco po a ing da a om bo h linguis ics and
a chaeology [1]. The conside a ion o a chaeological da a is c ucial o es ablishing he p e-
colonial geog aphy o TG popula ions, which would be e y incomple e i based only on his-
o ical eco ds, as shown by Fig 1. To achie e i , we began by e ising a gumen s buil wi hou
conside ing he a chaeological da a, especially hose de eloped be o e he 1960s [2–6], in
o de o con as hem wi h o he e idence o build ou models.
As a as linguis ic classi ica ions a e conce ned, he in e nal ela ions o he TG b anch o
he Tupı
´an amily ha e ecei ed much schola ly a en ion, wi h di e en app oaches employed
o es ablish hem om linguis ic da a alone. Phonological c i e ia ha e been pu o use along-
side g amma ical p ope ies and lexical cognacy, bo h in “ adi ional” [7–12] and “quan i a i e”
app oaches [13–17]. Al hough hese s udies ag ee o a la ge ex en on he opology o he
PLOS ONE
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 1 / 25
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Ci a ion: Fe az Ge a di F, T esoldi T, Coelho
A agon C, Reiche S, de Souza JG, Sil a Noelli F
(2023) Lexical phylogene ics o he Tupı
´-Gua anı
´
amily: Language, a chaeology, and he p oblem o
ch onology. PLoS ONE 18(6): e0272226. h ps://
doi.o g/10.1371/jou nal.pone.0272226
Edi o : Sø en Wichmann, Kiel Uni e si y,
GERMANY
Recei ed: Oc obe 29, 2021
Accep ed: July 14, 2022
Published: June 15, 2023
Pee Re iew His o y: PLOS ecognizes he
bene i s o anspa ency in he pee e iew
p ocess; he e o e, we enable he publica ion o
all o he con en o pee e iew and au ho
esponses alongside inal, published a icles. The
edi o ial his o y o his a icle is a ailable he e:
h ps://doi.o g/10.1371/jou nal.pone.0272226
Copy igh : ©2023 Fe az Ge a di e al. This is an
open access a icle dis ibu ed unde he e ms o
he C ea i e Commons A ibu ion License, which
pe mi s un es ic ed use, dis ibu ion, and
ep oduc ion in any medium, p o ided he o iginal
au ho and sou ce a e c edi ed.
Da a A ailabili y S a emen : The supplemen a y
ma e ial is a ailable in an anonymous online
eposi o y hos ed a OpenScienceF amewo k a he
add ess: h ps://os .io/a syk.
shallowe spli s, he e a e s ill i econcilable di e ences in e ms o he deepes ones, and much
disag eemen abou hei da ing. P e ious s udies using educed da ase s no designed o phy-
logene ic analysis [8–10] a e s ill he mos commonly e e enced ones. The o he wise ho ough
s udies by [11,12] con ained e o s in he da a ha may ha e in luenced he esul s. [15] is he
i s Bayesian phylogene ic classi ica ion, bu nei he he unde lying da a no he model a e
public. [18] has se e al issues, such as low pos e io suppo in b anches o well-known cases
o ela ionship (e.g., be ween Yuki and Si iono, o among Apiaka and Kawahi languages),
analyses including pa ame e s wi h e y low co e age, and he posi ion o some languages (e.g.,
Kamaju a
´), besides e o s in cognacy judgmen . In his s udy, we make all ou da a and models
a ailable, ollowing he p inciples o FAIR da a [19], and p epa e mul iple phylogene ic models.
Besides p o iding a phylogeny based on open da a, ou esul s a e he i s o o e a da ing o
he spli s h ough elaxed molecula clocks. Conside ing how he ques ion o he oo age and
he o de o spli s is a di iding poin among specialis s, he p ospec o building a uni ied in e -
disciplina y heo y in ol ing linguis ic, his o ical, gene ic, a chaeological, and e hnog aphic
e idence is conside ed in he discussion while p esen ing new g oupings.
2 Tupı
´-Gua anı
´languages and he ela ed a chaeology
2.1 Languages
TG is he la ges b anch o he Tupı
´an linguis ic amily [14,20,21], wi h abou 40 li ing lan-
guages (he e excluding Pi ipku a [22]) and a leas 9 ex inc ones [16]. The numbe o speake s
anges om less han a hund ed (e.g., Amondawa and Juma) o o e 6 million (Pa aguayan
Gua anı
´) [23]. The geog aphic dis ibu ion, wi h mos TG subg oups ound in Sou heas e n
Amazon, poin s o an o igin in his a ea due o i s g ea e linguis ic di e si y [24–26]. Such
hypo hesis con as s wi h common in e p e a ions o he a chaeological eco ds (poin ing o
an o igin close o he a ea be ween he uppe Tapajo
´s and Xingu i e s, u he o he wes
[6]), e hnog aphic sou ces, and indigenous cul u al epe oi e. A clea example o he la e a e
he ounda ion my hs and legends o he Ka’apo , ca ying a ious hin s ha hey we e once
loca ed o he wes o hei p esen e i o y [27,28].
No ma e he loca ion o he homeland, he expansion o TG is among he la ges in he
wo ld, sp eading ac oss o e 4000 km in bo h la i ude and longi ude [33] (see Fig 1), wi h i s
d i ing o ces a ma e o in ense deba e [6,13,21,34–42]. A chaeological esea ch sugges s
ha demog aphic g ow h was p opelled by he ise o ag icul u e, coupled wi h a s ong sense
o e i o iali y suppo ed by long- ange poli ical ne wo ks and by an expansionis wa like ide-
ology [6]. An inc easing a ea o o es ed landscape ha could be used o ag icul u e migh
ha e con ibu ed o his expansion [33,43]. Due o subs an ial simila i ies and a ini ies, ma e-
ial e idence sugges s a di e en scena io and a ch onology in line wi h wha one would expec
based on linguis ic and e hnog aphic g ounds. This is illus a ed by he a es o sha ed cog-
na es, as shown in Fig 2 (also in Fig 7 in Appendix C o S1 File), which a e ela i ely high when
compa ed wi h hose obse ed in o he g oups wi h supposedly compa able da es o hei
mos ecen common ances o , such as U alic a 43% and Romance a 93% [44]. A chaeolog-
ical da es conside ed oo ancien ha e o en been disca ded, based on he iew ha he TG dis-
pe sal is a ecen one. Howe e , o e ime he accumula ion o da es close o ca. 2000 BP in
di e en egions led o a ques ioning o his p emise. Glo och onological es ima es o ca. 2500
BP o he ini ial spli o he TG languages [45] ha e been used o suppo he a chaeological
da es. None heless, he disc epancy be ween such an ea ly ch onology and he ob ious p ox-
imi y be ween he TG languages was ne e le unno iced [3,4].
Any model seeking o explain he e olu ion o TG needs o accoun o hese ac s when
p oposing language phylogenies [43]. O iginally, wo such models we e pu o wa d. The i s
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 2 / 25
Funding: FFG and SR we e suppo ed by he by
Eu opean Resea ch Council (ERC) unde he
Eu opean Union’s Ho izon 2020 esea ch and
inno a ion p og amme (G an ag eemen No.
834050). TT was suppo ed by Cul u al E olu ion
o Tex s p ojec , wi h unding om he
Riksbankens Jubileums ond (g an ag eemen ID:
MXM19-1087:1).
Compe ing in e es s: The au ho s ha e decla ed
ha no compe ing in e es s exis .
[34] sees he lu ial ne wo k as he main enable o a apid expansion, an idea u he de el-
oped by [37,46], in which he causes o he dispe sal a e ela ed o clima ic ac o s. The o he
model inds he key d i e in popula ion inc ease, wi h he g owing need o mo e cul i a ion
a eas ( loodplain ag icul u e) and slowe mo emen s o expansion [6,35,38,47,48].
Mo e ecen ly, a comp omise has been ound by explici ly es ing demog aphic models
agains simula ed clima e change scena ios o he la e Holocene [43]. These models show ha
a combina ion o demic-di usion p ocesses and he p e e ence o a pa icula en i onmen al
niche ( opical mois o es s) bes explains he a chaeological ch onology and he gene al
Fig 1. The Tupı
´-Gua anı
´languages used in his s udy (in g een) and he Tupı
´an (non-TG) Awe ı
´(in blue), and Mawe
´(in ed), along wi h he
dis ibu ion o he TG a chaeological eco d (black do s). P epa ed by he au ho s wi h QGIS 3 [29], based on based on public domain da a and
as e images om “Na u al Ea h”, including da a om [30–32] and an unpublished da abase by Co êa and Noelli.
h ps://doi.o g/10.1371/jou nal.pone.0272226.g001
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 3 / 25
econs uc ion o his o ical linguis ics: a long s asis in he Amazon, wi h he eme gence and
de elopmen o he main Tupı
´b anches, ollowed by a apid expansion o o he pa s o Sou h
Ame ica (co esponding o he TG expansion) [43]. [34] concludes ha he mos likely cen e
o he dispe sion o he TG is he Uppe Tapajo
´s. [49] p oposes a sou hwes e n Amazonian
homeland o P o o-Tupı
´-Gua anı
´(PTG), lying nea he A inos and uppe Ju uena i e
basins. Using he Linguis ics Mig a ion Theo y and mo i a ed by he classi ica ion in [15,50]
posi s he homeland o PTG in he lowe Xingu.
The loca ion o he cen e o expansion is, as expec ed, dependen o he opology o he
amily. The Mawe
´-Awe ı
´-Tupı
´-Gua anı
´hypo hesis [21,51–56] s a es ha a single ances o o
hese h ee g oups b anched o om he es o he Tupı
´an amily [14] (see Fig 3). The spli o
he b anch oday composed by Mawe
´would ha e been ollowed by ha o he ances o o
Awe ı
´and PTG. Cogna es sha ed by Awe ı
´and TG languages bu no p esen in Mawe
´suppo
his hypo hesis, showing ha Mawe
´had al eady b anched o while he ances o s o Awe ı
´and
Fig 2. Cognacy o each language pai used in he main analysis, anging om 0 ( ull di e ence) o 1 (iden i y). The cognacy di e si y [44] o all
languages is 15%. I Mawe
´and Awe ı
´a e excluded, he alue is 14%.
h ps://doi.o g/10.1371/jou nal.pone.0272226.g002
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 4 / 25
PTG o med single g oup which hea ily bo owed lexical ma e ial o Ca iban o igin [49,57].
We p esen some o hese cases in Table 1, wi h cognacy judgmen based on [16].
2.2 A chaeology
The dispe sal o TG languages has a clea ma e ial co ela e in he sp ead h oughou eas e n
Sou h Ame ica o a package ha includes a pa icula ype o ce amics, plan managemen ,
Fig 3. The Tupı
´an languages wi h he sub-b anches o he Mawe
´-Awe ı
´-Tupı
´-Gua anı
´b anch emphasized.
Adap ed om [21].
h ps://doi.o g/10.1371/jou nal.pone.0272226.g003
Table 1. Cogna es sha ed by Mawe
´and Awe ı
´no p esen in TG (in yellow) and cogna es sha ed by Awe ı
´and TG (in blue) no p esen in Mawe
´.Tupinamba
´is aken
as a ep esen a i e o he PTG descendan s. The numbe s in he las column e e o TG languages whose concep s a e cogna es wi h he Tupinamba
´wo d p o ided, illus-
a ing cogna es in o he b anches o TG: A a
´-Canoei o (1), Wayampi (2), Guajaja a (3), Pa akanã(4), Asu inı
´Xingu (5), Kamaju a
´(6).
Mawe
´Awe ı
´Tupinamba
´
Leg ʔup ʔup e ɨmã(1,2,3,5,6)
Sing mepɨ epɨ ɲeʔeɲga (2,4)
Come back aipok ʔajpog jeβi (1,2,3,5,6)
Hoplias (genus) (n)ipiu a piu a
´ aeʔia (1,2,3,4)
Fly (insec ) win in meu (2,3,4,5,6)
Jagua awɨa o awa jawa (1,2,3,4,5,6)
An ea e a ihı
˜ amajua amandwa (1,2,3,4,6)
G and a he aseʔi amu
˜j amı
˜ja (2,3,4,5,6)
Wound pihi peʐep peɾeb –
Tapi wewa o apiʔi apiʔiɾ(1,2,3,4,5,6)
Sky a ipɨ ɨwak ɨβak (1,2,3,4,5,6)
Genipa wããhop ẽ ɨpap janɨpab (2,3,4,5,6)
Ba hakiʔi a iʔa anɨ a (1,2,3,4,5)
Sie e ( ool) panane ku upem u upem (2,4,5,6)
Bu n (some hing) wuk apɨapɨ(1,2,3,4,5,6)
Bow mo eawa ʒapa ɨβɨɾapaɾ(1,2,3,4,5,6)
S a wajki u a ɨaɁɨ jasɨ a a (1,2,3,4,5,6)
h ps://doi.o g/10.1371/jou nal.pone.0272226. 001
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 5 / 25

and cul i a ion o a a ie y o c ops [6,38], as shown in Fig 1. This is o en ci ed as one o he
ew cases whe e an ob ious co ela ion exis s be ween an a chaeological cul u e and a language
amily, o he poin whe e he name “Tupigua ani” (no hyphen) was applied o he a chaeolog-
ical adi ion ( o a c i icism o his concep see [47,58]). Admi edly, co ela ing a ma e ial
cul u e s yle wi h he speake s o a single language o language amily is in mos cases a p ob-
lema ic, i no naï e, app oach. Simila ly, ma e ial cul u e changes may p ecede o pos da e
ela ed changes in socie y and language [59–61]. Ne e heless, he e is o e whelming e idence
o suppo he associa ion be ween he ce amics con en ionally called “Tupigua ani” and he
sp ead o he Tupı
´-Gua anı
´language amily. O pa icula in e es is he no able homogenei y
o he ma e ial cul u e h oughou he Tupı
´-Gua anı
´ e i o y [31]. This conse a ism is seen
e en in a eas his o ically occupied by linguis ically dis inc g oups such as he Tupiniquim and
Tupinamba
´[62]. The high s anda diza ion in ce amic s yles ac oss ime and space—accompa-
nied by he main enance o a speci ic ocabula y o desc ibe essel shapes [63]—is a es imony
o he conse a ism ound in o he sphe es o he Tupı
´-Gua anı
´cul u es [64]. Ul ima ely, he
ce amics ecognized as “Tupigua ani” by a chaeologis s can be aced back o he Tupı
´an
homeland in sou hwes e n Amazon, whe e i s s ylis ic componen s, such as polych ome pain -
ing, can be ound among o he ce amic adi ions [65].
In wha ollows, we summa ize he ea lies adioca bon da es a ailable o Tupigua ani
si es. The da es a e di ided acco ding o i e egions: Eas e n Amazon, Boli ia, A lan ic Coas ,
No he n B azil, and he Pa ana
´Basin. All da es a e calib a ed wi h he sou he n hemisphe e
cu e [66] and epo ed in he 2-sigma in e al.
Eas e n Amazon An ea ly p esence in he Xingu-Tocan ins in e lu e is suppo ed by he
a ailable adioca bon da es. A da e o 2430 ±20 BP (cal BP 2680–2340) om a si e in he
Tocan ins-A aguaia con luence is s ill seen wi h cau ion, as i is conside ably olde han all
o he da es om he same egion [67,68]. The accep ed Tupigua ani ch onology o he eas -
e n Amazon s a s a 1670 ±80 BP (cal BP 1700–1350) be ween he Tapajo
´s and Tocan ins i -
e s [68].
Boli ia The ea lies po en ial TG si e in p e-Andean Boli ia has a da e o 1680 ±90 BP (cal
BP 1730–1320, UA-10240), which, i con i med, would imply an a i al o he Gua anı
´-speak-
ing Gua ayo and Chi iguano in he egion ea lie han commonly hough [69].
A lan ic coas In he egion his o ically occupied by he Tupinamba
´, a con o e sial ea ly
ch onology has been p oposed by Scheel-Ybe e al. [70], based on da es eaching 2920 ±70
BP (cal BP 3220–2790, Gi -11045) om si es in he s a e o Rio de Janei o. These p eda e he
TG expansion ou o he Amazon by any es ima e. Excluding hose ou lie s, he ea lies da e
o he A lan ic o es is o 1740 ±90 BP (cal BP 1825–1380, Be a-84333) [70], which is in line
wi h he ch onology o o he pa s o he TG e i o y. Mos da es a e conside ably mo e
ecen , la e han 1055 ±80 BP (cal BP 1060–740, SI-828) [71].
No heas e n B azil Few da es a e a ailable o no heas e n B azil. In he semi-a id hin-
e land, a da e o 1690 ±110 BP (cal BP 1810–1315, GIF-3225) is some imes a ibu ed o a TG
occupa ion, bu he cul u al a ilia ion o he da ed si e is no a consensus [38,72]. Discoun ing
da es wi h excessi ely la ge s anda d de ia ions [73], he occupa ion o he coas possibly
ex ends back o 1880 ±60 BP (cal BP 1920–1590, Be a-118818) [31], wi h mos da es being
conside ably la e .
Pa ana
´Basin The sou he nmos egion o Tupı
´-Gua anı
´occupa ion, whe e Gua anı
´and
ela ed languages we e dominan , has he mos comple e and eliable ch onology [32]. The
ea lies da e, 2010 ±75 BP (cal BP 2090–1740, SI-5028), comes om he middle Pa ana
´ i e
[74]. Be ween ha da e and he second millennium, mul iple si es a e a es ed in he São
Paulo highlands, sou he nmos B azil, and he Pa ana
´-U uguay in e lu e in A gen ina
[74].
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 6 / 25
3 Ma e ials and me hods
3.1 Da a
We ollowed he cu en bes p ac ices o linguis ic phylogene ics (“phylolinguis ics”), whe e
cogna e gain and loss in basic ocabula y a e he e olu iona y cha ac e s used o in e a da ed
ee [75–77]. The comple e da ase used in his s udy is de i ed om [16] and is publicly a ail-
able, along wi h he phylogene ic models, a h ps://os .io/a syk. Fo be e in eg a ion wi h
o he linguis ic esou ces, we s anda dized he da a ollowing he o ma s and ca alogues o
[78]. Cogna e se assignmen , ollowing he p inciple o oo -meaning ai s [79], was i s pe -
o med wi h he au oma ic me hods implemen ed in LingPy [44,77,80,81] and la e manually
e iewed by expe s in i s en i e y. Table 2 shows a sample o cognacy assignmen om [16].
The da a in ou s udy comp ises lexica om 40 “doculec s” [82] (i.e., language a ie ies).
Mawe
´and Awe ı
´we e included in he analyses, wi h he spli o he Mawe
´ances o se ing as
he oo and e lec ing he a o emen ioned and well es ablished Mawe
´-Awe ı
´-TG hypo hesis.
We also included Omagua and Kokama due o a high po ion o hei lexicon being o TG o i-
gin, despi e hei non-TG o igin [83,84], a hypo hesis ejec ed by [85]. Some TG languages
a ailable in ou sou ce we e excluded om he analyses due o an excessi ely low co e age.
The lis o concep s is p o ided in Appendix A o S1 File, along wi h he co esponding
Concep icon cogna e se ids and glosses [86] when a ailable. The choice o concep s elied on
he ollowing c i e ia: concep s om he Swadesh [87] and Leipzig-Jaka a [88] lis s, he Swa-
desh lis ex ended by [89], and cul u ally ele an TG concep s aken om [90] and expanded
by he au ho s. The concep co e age o each language is gi en in Table 3. We used 415 con-
cep s om an upcoming e sion o [16]. We assessed he deg ee o ee-likeliness by compu -
ing he concep s’ TIGER sco es [91,92] wi h he implemen a ion by [93], ob aining a mean
sco e o 0.14 (±0.14) (indi idual sco es a e epo ed in Appendix K o S1 File). This alue sug-
ges s a compa a i ely high le el o non- e ical ansmission, being lowe han he lowes sco e
epo ed in [93] o 0.20 o D a idian, and suppo s he quali a i e assessmen ha “ he e is an
o e all absence o well-delimi ed lexical clus e s inside [TG]” [13].
3.2 Phylogene ic econs uc ion and da ing
Da a was p epa ed wi h he s a e-o - he-a so wa e ools o compu e -assis ed pipelines in
compu a ional his o ical linguis ics [80,94] and expo ed in he ex ended NEXUS o ma [95].
The iles p oduced by his pipeline we e p ocessed and no malized wi h Py hon sc ip s de el-
oped o his esea ch.
Since he e olu iona y his o y o he TG languages is no comple ely ee-like, as pe [13]
and measu es in Sec ion 3.1, we i s gene a ed a dis ance ma ix o build a Neighbo Ne ne -
wo k using Spli sT ee e sion 4.17.1 [96] o isualize he con lic ing signal and calcula e he Q-
esiduals and he δ-sco es.
Di e en phylogene ic models we e hen explo ed in e ms o subse s o concep s, lan-
guages, molecula clocks, calib a ion da es, subs i u ion models, a e a ia ion, and
Table 2. Cognacy sample om ou da abase.
Language Concep Phone ic o m Cogna e se
Tupinamba
´BAT anɨ a 171
Wayampi BAT anɨla 171
Gua anı
´BAT mopi 172
Kaiowa
´BAT
m
bopi i 172
Mawe
´BAT hakiɁi 4513
h ps://doi.o g/10.1371/jou nal.pone.0272226. 002
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 7 / 25
monophyle ic cons ain s. We decided in a o o he simples and mos common p ac ices
whene e possible and sensible, ollowing he p inciple ha we should begin wi h mo e
app oachable s udies be o e en u ing in o mo e complex scena ios. The ini ial explo a ion,
pa ly published in [97], was ele an o he au ho s o discuss he concep s ha we e deemed
less eliable, and he p oblems ha could a ise om he analysis. These s udies also se ed o
e alua e he easibili y and obus ness o ou app oach.
Table 3. Concep co e age o he languages used in his s udy om [16].
Language Glo ocode ISO 639–3 Code Co e age
Ache ache1246 guq 80%
Amondawa amun1246 adw 74%
Anambe anam1249 aan 49%
Apiaka apia1248 api 65%
A awe e a aw1273 aw 71%
Asu inı
´Tocan ins oca1235 asu 84%
Asu inı
´Xingu xing1248 asn 63%
A a-Canoei o a ac1239 a 79%
Awe i awe 1244 awe 93%
Chi iguano eas 2555 gui 90%
Guaja guaj1256 g j 80%
Guajaja a guaj1255 gub 97%
Gua anı
´pa a1311 gnn 99%
Gua ayo gua 1292 gy 89%
Ka’apo u ub1250 u b 95%
Kaiowa kaiw1246 kgk 51%
Kamaju a kama1373 kay 68%
Kayabi kaya1329 kyz 63%
Kokama coca1259 cod 82%
Mawe sa e1243 ma 88%
Mbya mbya1239 gun 84%
Nheenga u nhen1239 y l 91%
Old Gua anı
´oldp1258 g n 83%
Omagua omag1248 omg 80%
Pa akanãpa a1312 pak 83%
Pa in in in enh1241 pah 96%
Si iono si i1273 s q 94%
Su ui-Aikewa a su u1262 mdz 83%
Tapie e api1253 pj 77%
Tapi ape api1254 a 68%
Teko eme 1243 eme 96%
Tembe emb1276 qb 93%
Tenha im nucl1663 pah 76%
Tupinamba upi1273 pw 99%
U ueuwauwau u ue1240 u z 60%
Wa azu paus1244 psm 85%
Wayampi waya1270 oym 99%
Xe a xe a1241 xe 61%
Yuki yuqu1240 yuq 64%
Zo’e zoee1240 p o 82%
h ps://doi.o g/10.1371/jou nal.pone.0272226. 003
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 8 / 25
We s uc u ed he esea ch in o wo ounds, he i s one designed o ob ain summa y ees
gi en di e en scena ios o analysis and he second one using hese esul s o pe o m a phylo-
geog aphic s udy. The i s ound was composed o wo s udies ha di e in he subse o con-
cep s used: a “ ull” s udy, wi h all concep s desc ibed abo e il e ed o ensu e hey we e
missing a mos in 20% o he languages, and a “swadesh” s udy using he lis o [87] (see
Appendix C in S1 File) as close as possible, il e ed o ensu e hey we e missing in a mos 30%
o he cases. Such h esholds we e necessa y due o he high le el o spa si y o he da a. In
bo h cases concep s we e g ouped in wo equal-sized pa i ions based on he o e all numbe o
cogna es in each. Besides simple s ic -clock models, which a e compa able o glo och ono-
logical app oaches, all analyses also used unco ela ed elaxed-clock models sampled om a
logno mal dis ibu ion [98,99]. In he la e , each b anch o a ee has i s own clock a e, wi h
pa ame e s ha a e independen om hose o he mo he and sis e b anches, allowing ab up
changes in e olu iona y a es. These a e conside ed compa ible wi h bo h he e olu ion o TG,
gi en i s ela i ely ecen and apid expansion, and Sou h Ame ican languages in gene al, pa -
icula ly due o he impac o Eu opean coloniza ion in e ms o popula ion size, displacemen ,
and eplacemen [100,101].
We pe o med phylogene ic econs uc ion using BEAST2 e sion 2.6.6 [102], i ing di e -
en bina y co a ion models [103], whe e he ansi ion be ween “p esence” and “absence” o a
cogna e in a language is assumed o be symme ic and equally p obable, along wi h a la en
a iable modeling whe he each cogna e swi ches be ween p esence and absence a a “ as ” o
“slow” a e. Asce ainmen co ec ion was pe o med acco ding o p ac ices desc ibed in [76].
Conside ing how ou da a only o e s wo his o ical languages ha could be used o empo al
calib a ion (Tupinamba
´and Old Gua anı
´), bo h o which a e o some ex en composed om
mul iple sou ces di e ging in p o enance and da e (each spanning o e mo e han a hund ed
yea s), we decided o guide he in e ence only by se ing a uni o m dis ibu ion o he oo , in
ag eemen wi h all sensible a chaeological and linguis ic hypo heses, and by es ablishing
monophyle ic g oups accep ed by i ually all expe s, also adjus ing ip da es o languages
collec ed mo e han 50 yea s ago (de ailed in Appendix D o S1 File). We used a Bi h-Dea h
model [104], pe o ming 25
7
MCMC i e a ions, sampling ees om he pos e io dis ibu ion
o ob ain a maximum clade c edibili y ee (MCC) based on common ances o heigh s a e a
50% bu n-in, using T eeAnno a o e sion 2.6.4 [105]. We plo ed ees wi h [106] and Fig-
T ee e sion 1.4.4; he ees, including o he supplemen a y models, a e p esen ed in Figs
8–11, all in Appendix E o S1 File.
The esul s in Sec ion 4 a e hose o he “ ull” s udy using a elaxed clock. The decision in
a o o his model as ou main esul is based on he se o concep s i in ol es, which,
despi e a highe e icula ion signal, includes amily-speci ic concep s ha we e deemed ele-
an o s udying he e ical ansmission. I is necessa y o no e ha he logma ginal likeli-
hood (see Appendix L in S1 File), compu ed wi h nes ed samples [107], no only a o ed he
“swadesh” da ase , as expec ed in ace o i s lowe da a complexi y, bu i also yielded a be e
sco e o he s ic molecula clock in he case o he “ ull” da ase . Ou decision in a o o
he elaxed clock model was due o an expe analysis o he esul ing opology and da es, as
i was a mo e compa ible wi h he li e a u e, and by he ac ha mos unexpec ed esul s,
such as he posi ion o he Anambe
´-A awe e
´clade o he b anch leng h o Tupinamba
´, can
be explained by di e ences in concep co e age. The comple e s udies a e p esen ed in he
supplemen a y ma e ial and should guide u u e esea ch and e inemen s o cogna e
judgmen s.
The phylogeog aphic s udy used he opology o he MCC ee o he “ ull” s udy as a se o
monophyle ic cons ain s whe e e we had ob ained a pos e io suppo o a leas 0.70, along
wi h he 95% heigh ange o each such spli , ocusing on ha ing he model sea ch o da es
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 9 / 25
6 Concluding ema ks
Mos cases o lowe pos e io suppo in ou ee can be explained, a leas in pa , by missing
da a. The low con idence in some o he spli s wi hin indi idual g oups, such as among he
Gua anı
´languages, migh be due o bo h echnical aspec s, like an unequal le el o sampling,
and he ac ual his o y o he languages, in ol ing dialec chains, admix u e bo h a he linguis-
ic and gene ic le els, e c. These issues a e heigh ened by he lack o calib a ion da a o empo-
al and geog aphic ma e ha can be applied di ec ly, as well as by ou decision o begin his
esea ch pa h by using models which a e simple o unde s and and less suscep ible o p io
hypo heses speci ied by us. The opology and he pos e io suppo a e expec ed o imp o e as
we ex end he da a, employ mo e complex models (which end o in ol e di e en ypes o cal-
ib a ions), and, po en ially, he di ec o indi ec usage o addi ional linguis ic e idence o
allow he a p io i de ini ion o monophyle ic g oups, aiming o mo e p ecise pa ame e s o
local e olu ion. The his o ical-an h opological su ey wo k p esen ed in he p e ious sec ions,
in pa icula , may p o e o be ex emely aluable in u u e esea ch, p o ided ha i is used
wi h he due cau ion (see [159]).
I is essen ial o emphasize ha he classi ica ion p esen ed he e is exclusi ely based on lexi-
cal changes, al hough o mos o he clades he e is a signi ican ag eemen wi h Rod igues’
axonomy based on phonology [10], and e en wi h he cul u al classi ica ion in [42]. A ca ea
is necessa y he e: linguis ic classi ica ions based exclusi ely on phonological changes, such as
he one by [10], a e gene ally conside ed o be mo e suscep ible o common independen
inno a ions, ha is, cases in which he same cha ac e (a sound change) independen ly a ises
mo e han once in di e en b anches, leading o “homoplasy”. This is one o he main easons
o he sugges ion ha mos phylolinguis ic s udies should in ol e exclusi ely o majo ly cha -
ac e s based on lexical inno a ions, which can be assumed o be independen and a ise only
once. Likewise, when conside ing di e ences wi h a chaeological da ings, i is wo h no ing
ha such phylolinguis ic models conside , and by ex ension da e, spli s as he momen when
he i s disjunc i e lexical inno a ion in he basic ocabula y akes place. Such e en does no
necessa ily imply a deg ada ion o mu ual in elligibili y, no can i be au oma ically associa ed
wi h ei he popula ion displacemen o changes in a cheological packages.
In e ms o ou es o expansion, we belie e ha he ances o s o Tupinamba
´ ook di e en
di ec ions, a eling eas wa ds, while he es o he g oup a eled wes wa ds i s and hen
no hwa ds. Pa alleling Rod igues’ G oup VIII (which includes Guaja
´), he g oup con aining
Ka’apo , Teko
´, Wayampi, and Zo’e is clea ly suppo ed by phonological and lexical inno a-
ions alike, despi e he p esence o he Teko
´in he F ench Guiana al eady in he 1500s [160].
Since Zo’e is phonologically close o Teko
´ han o Wayampi [161], i is possible ha he ances-
o s o he Zo’e, as hose o he Teko
´, had al eady sepa a ed om he ances o s o Wayampi,
whose mig a ion no hwa ds om he lowe Xingu i e only began in he ea ly 1600s [154,
155,162]. The la e spli o Wayampi and Teko
´in ou ee is p obably caused by inno a ions
common o bo h g oups and bo owings o Ca iban o igin no p esen in Zo’e, exempli ied in
Table 5. I is unknown whe he he Tupı
´an g oup e e ed o as “Apama” [144,163] and
desc ibed in 1691 be ween he Cu ua and Maicu u a e ances o s o he Zo’e, who in 1600 we e
s ill loca ed in Lowe Xingu. I he iden i ica ion o his g oup wi h he Zo’e is co ec , we
could in e hei mo emen s based on addi ional, non-linguis ic e idence.
The mig a ion o mul iple g oups owa ds Rondo
ˆnia du ing he colonial pe iod is no only
acknowledged by mul iple sou ces [123,139,140,143,164], bu can also be demons a ed lin-
guis ically on he basis o he abo emen ioned Ca ib loans in PTG [124], no ound in Mawe
´
o Awe ı
´. These loans a e also ound in he Kawahi languages. O all conjec u es ega ding
how he Kamaju a
´ eached hei cu en loca ion [126,132,165], an a ac i e one is old by
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 16 / 25

he Kuiku o, acco ding o whom hey came om he No h, passing ia he A aguaia i e
h ough he Ka aja
´ e i o ies, en e ing he Xingu basin ia he Suya
´-Missu
´ i e [166]. How-
e e , he e is no a chaeological sign o such an en ance o a TG g oup in he Uppe Xingu.
In conclusion, a ho ough his o y o he o ma ion and de elopmen o TG languages,
including he dis inc ion be ween e ical, in- amily, and ou - amily ho izon al ansmission,
is ye o be w i en, e iewing e e y hing ha has been p oposed so a . A uni ied in e disci-
plina y heo y mus gi e weigh o da a om linguis ics, a chaeology, e hnology, as well as
gene ics and he app oach p oposed in his a icle collabo a es owa ds such an en e p ise. We
mus also conside ha he p esence o ma e ial ha is no e ically ansmi ed does no
mean only ha a ee will be dis o ed: i also means ha e en a “pe ec ” ee, one co ec ly
cap u ing all ela ionships o descen , will mi o only a pa o he his o y, especially i he
sp ead o “ho izon al” inno a ions was much as e han ha o he “ e ical” descen . A ee
o lexical inno a ions is no a na a i e o he his o y o he languages in ol ed, bu a means o
ell one.
Al hough a c i ical e iew o he en i e adioca bon eco d associa ed wi h he TG dispe sal
is beyond he scope o his wo k, he quick assessmen o he ea lies egional da es summa-
ized in his pape illus a es he di icul y o concilia ing a chaeological and linguis ic da a. In
he u u e, s ic c i e ia o ch onome ic hygiene should be applied o he published TG ch o-
nology o asce ain he eliabili y o each da e [167]. Fo now, e en i he long ch onology p o-
posed o some egions is disca ded [70], he ch onology a ailable o he Pa ana
´basin makes
i di icul o a gue o a ecen a i al in he egion. Nume ous si es in sou he n B azil and
A gen ina p eda e he second millennium [32], which is impossible o concilia e wi h an es i-
ma e o a ound 1750 yea s BP o he beginning o he TG dispe sal o hose egions.
In e ms o phylogene ic s udies based on linguis ic da a, besides inco po a ing expedien-
cies om a chaeology as p io s, u u e wo k migh in es iga e combining non-pa ial cognacy
da a wi h o he ea u es, such as pa ial cognacy se s, mo phology, and phonology. Fo exam-
ple, due o he s ong composi e cha ac e o TG lexicon, we decided no o use in o ma ion
on pa ial cognacy, despi e i s limi ed a ailabili y in [16]. Despi e he sou ce da a ca ying
in o ma ion on pa ial cognacy, we decided o employ exclusi ely simple cogna es, also in con-
side a ion o how he subs i u ion models a ailable in Bayesian amewo ks s ill demand non-
s anda d con igu a ions o use hem in an adequa e way [81,168].
Also dese ing mo e conside a ion a e TG p ac ices ha esul ed in he conse ancy o pa
o he language and i s meanings obse ed in hei ma e ial cul u e and en i onmen al man-
agemen . These a e ac s o en his o ically, e hnog aphically, linguis ically, and a chaeologi-
cally eco ded in di e en imes and places by people wi h di e en expe ise and objec i es
Table 5. Lexical inno a ions in ou G oup Ib (Wayampi, Teko
´, and Zo’e). Some, no sha ed by Zo’e, ook place when Wayampi and Teko
´we e al eady in he F ench
Guiana, as he sou ce o he bo owings indica es. The wo d o ‘ imbo liana’ and he plu al ma ke a e exclusi e o hese languages.
Concep Wayampi Teko
´Zo’e Bo owed om
Timbo
´liana ɨmeku beku meku
˜Wayana
Plu al ma ke ku
˜kom kãCa iban language
Pan pa u pa u apimãF om Po uguese h ough Wayana
Milk ile di e ɨC eole
Mi o wa ua wa uwa po oesake Lı
´ngua Ge al
Kni e ma ija m
b
a iʤe boke Wayana
Sal sau u sau u jukɨ F om English h ough Wayana
3
d
pl. kupa kupa – Ca iban language
Hen masakala masakala ɲa ı
˜Wayana
h ps://doi.o g/10.1371/jou nal.pone.0272226. 005
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 17 / 25
who pe cei ed a ious “empi ical” and “ heo e ical” aspec s o he TG peoples, as shown in
[169]. Bo h ways lead o unde s anding he ela ions be ween he TG and o he cul u es,
which included he app op ia ion and ans o ma ion o people, objec s and language [170,
171], in p ocesses cha ac e ized by “changes wi hin con inui ies”. The answe o hese ques-
ions can be said o be he holy g ail o TG his o iog aphy.
Suppo ing in o ma ion
S1 File.
(PDF)
Acknowledgmen s
We hank he anonymous e iewe s o hei commen s and sugges ions ha g ea ly helped us
in imp o ing ou models, esul s, and manusc ip . We hank Ta iana Me zhe ich o helping
in c ea ing he igu es in his pape .
Au ho Con ibu ions
Da a cu a ion: Fab ı
´cio Fe az Ge a di, Ca olina Coelho A agon, S anisla Reiche .
Fo mal analysis: Fab ı
´cio Fe az Ge a di, Tiago T esoldi.
Supe ision: Fab ı
´cio Fe az Ge a di, Tiago T esoldi.
W i ing – o iginal d a : Fab ı
´cio Fe az Ge a di, Tiago T esoldi, Ca olina Coelho A agon,
S anisla Reiche , Jonas G ego io de Souza, F ancisco Sil a Noelli.
W i ing – e iew & edi ing: Fab ı
´cio Fe az Ge a di, Tiago T esoldi, Ca olina Coelho A agon,
S anisla Reiche , Jonas G ego io de Souza, F ancisco Sil a Noelli.
Re e ences
1. An hony DW. The ho se, he wheel, and language: how B onze-Age ide s om he Eu asian s eppes
shaped he mode n wo ld. P ince on Uni e si y P ess; 2009.
2. Noelli FS. As hipo
´ eses sob e o cen o de o igem e o as de expansão dos Tupı
´. Re is a de An opolo-
gia. 1996; 39(2):7–53. h ps://doi.o g/10.11606/2179-0892. a.1996.111642
3. Vi ei os de Cas o E. Comen a
´ io ao a igo de F ancisco Noelli. Re is a de An opologia. 1996; 39
(2):55–60. h ps://doi.o g/10.11606/2179-0892. a.1996.111643
4. U ban G. On he geog aphical o igins and dispe sion o Tupı
´an languages. Re is a de An opologia.
1996; 39(2):61–104. h ps://doi.o g/10.11606/2179-0892. a.1996.111644
5. Noelli FS. Respos a a Edua do Vi ei os de Cas o e G eg U ban. Re is a de An opologia. 1996; p.
105–118. h ps://doi.o g/10.11606/2179-0892. a.1996.111645
6. Noelli FS. The Tupı
´: Explaining o igin and expansions in e ms o a chaeology and o his o ical linguis-
ics. An iqui y. 1998; 72(277):648–663. h ps://doi.o g/10.1017/S0003598X00087068
7. Lemle M. In e nal classi ica ion o he Tupı
´-Gua anı
´linguis ic amily. Tupı
´s udies I. 1971; 29:107–129.
8. Rod igues AD. Relac¸ões in e nas na amı
´lia lingu¨ı
´s ica Tupı
´-Gua anı
´. Re is a de An opologia. 1984;
27/28:33–53.
9. Die ich W. Mo e e idence o an in e nal classi ica ion o Tupı
´-Gua anı
´languages. Be lin: Geb .
Mann; 1990.
10. Rod igues AD, Cab al ASAC. Re endo a classi icac¸ão in e na da amı
´lia Tupı
´-Gua anı
´; 2002. Lı
´nguas
Indı
´genas B asilei as. Fonologia, G ama
´ ica e His o
´ ia, A as do I Encon o In e nacional do GTLI da
ANPOLL.
11. Mello AAS. Es udo his o
´ ico da amı
´lia linguı
´s ica Tupı
´-Gua anı
´: Aspec os onolo
´gicos e lexicais [PhD
hesis]. Uni e sidade Fede al de San a Ca a ina. Flo iano
´polis; 2000.
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 18 / 25
12. Mello AAS. E idências onolo
´gicas e lexicais pa a o sub-ag upamen o in e no Tupı
´-Gua anı
´; 2002.
Lı
´nguas Indı
´genas B asilei as. Fonologia, G ama
´ ica e His o
´ ia. A as do I Encon o In e nacional do
GTLI da ANPOLL.
13. E iksen L, Galucio AV. The Tupian expansion. In: O’Conno L, Muysken P, edi o s. The na i e lan-
guages o Sou h Ame ica. Camb idge; 2014. p. 177–199.
14. Galucio AV, Mei a S, Bi chall J, Moo e D, Gabas Ju
´nio N, D ude S, e al. Genealogical ela ions and
lexical dis ances wi hin he Tupian linguis ic amily. Bole im do Museu Pa aense Emı
´lio Goeldi Ciên-
cias Humanas. 2015; 10(2):229–274. h ps://doi.o g/10.1590/1981-81222015000200004
15. Michael LD, Chousou-Polydou i N, Ba olomei K, Donnelly E, Wau e s V, Mei a S, e al. A Bayesian
phylogene ic classi ica ion o Tupı
´-Gua anı
´. LIAMES. 2015; 15(2):193–221.
16. Ge a di FF, Reiche S, A agon C, Wien zek T, Lis JM, Fo kel R. TuLeD: Tupı
´an lexical da abase
( 0.12); 2022. A ailable om: h ps://doi.o g/10.5281/zenodo.6572576.
17. Ja¨ge G. phylogene icIn e enceASJP19; 2021. A ailable om: h ps://os .io/a97sz.
18. Ge a di F, Reiche S. The Tupı
´-Gua anı
´language amily: A phylogene ic classi ica ion. Diach onica.
2021; 38(2):151–188. h ps://doi.o g/10.1075/dia.18032. e
19. Wilkinson MD, Dumon ie M, Aalbe sbe g IJ, Apple on G, Ax on M, Baak A, e al. The FAIR Guiding
P inciples o scien i ic da a managemen and s ewa dship. Scien i ic Da a. 2016; 3(1):1–9. h ps://doi.
o g/10.1038/sda a.2016.18 PMID: 26978244
20. Rod igues AD. Tupı
´. In: Dixon RMW, Aikhen ald A, edi o s. The Amazonian Languages. Camb idge
Uni e si y P ess; 1999. p. 107–124.
21. Rod igues AD, Cab al A. Tupı
´an. In: Campbell L, G ondona V, edi o s. The Indigenous Languages o
Sou h Ame ica. Mou on de G uy e , Be lin/Bos on; 2012. p. 495–574.
22. Ju
´nio TdSS, Cando JC. Uso de ecu sos na u ais pelos I
´ndios Pi ipku a no No oes e de Ma o
G osso: uma ana
´lise do Conhecimen o Ecolo
´gico T adicional no con ex o da polı
´ ica expansionis a do
B asil na Amazo
ˆnia Me idional. Re is a b asilei a de lı
´nguı
´s ica an polo
´gica. 2016; 8(2):73–104.
23. Ebe ha d DM, Simons GF, Fennig CD. E hnologue: Languages o he Wo ld. Twen y- ou h edi ion.
ol. 16. Dallas, TX: SIL in e na ional; 2021. A ailable om: h p://www.e hnologue.com.
24. Ca alli-S o za LL, Menozzi P, Piazza A. The his o y and geog aphy o human genes. P ince on, NJ:
P ince on Uni e si y P ess; 1994.
25. Ne le D. Linguis ic di e si y o he Ame icas can be econciled wi h a ecen coloniza ion. P oceedings
o he Na ional Academy o Sciences. 1999; 96(6):3325–3329. h ps://doi.o g/10.1073/pnas.96.6.3325
PMID: 10077683
26. Henn BM, Ca alli-S o za LL, Feldman MW. The g ea human expansion. P oceedings o he Na ional
Academy o Sciences. 2012; 109(44):17758–17764. h ps://doi.o g/10.1073/pnas.1212380109 PMID:
23077256
27. Ribei o D. Dia
´ ios ı
´ndios: Os U ubus-Kaapo . São Paulo: Edi o a Companhia das Le as; 1996.
28. Bale
´e WL, e al. Foo p in s o he o es : Ka’apo e hnobo any— he his o ical ecology o plan u iliza ion
by an Amazonian people. New Yo k, NY: Columbia Uni e si y P ess; 1994.
29. QGIS De elopmen Team. QGIS Geog aphic In o ma ion Sys em; 2009. A ailable om: h p://qgis.o g.
30. Na ional Ins i u e o His o ic and A is ic He i age (IPHAN). Na ional Regis e o A chaeological Si es
(CNSA). B ası
´lia: IPHAN; 2018.
31. Co êa A
ˆA. Pindo ama de mboı
ˆa e ı
ˆaka e
´: con inuidade e mudanc¸a na aje o
´ ia das populac¸ões Tupi;
2014.
32. Bonomo M, Ang izani RC, Apolinai e E, Noelli FS. A model o he Gua anı
´expansion in he La Pla a
Basin and li o al zone o sou he n B azil. Qua e na y In e na ional. 2015; 356:54–73. h ps://doi.o g/
10.1016/j.quain .2014.10.050
33. I ia e J, Smi h RJ, G ego io de Souza J, Mayle FE, Whi ney BS, Ca
´ denas ML, e al. Ou o Amazonia:
La e-Holocene clima e change and he Tupi–Gua ani ans-con inen al expansion. The Holocene.
2017; 27(7):967–975. h ps://doi.o g/10.1177/0959683616678461
34. Me
´ aux A. Mig a ions his o iques des Tupi-Gua ani. Jou nal de la Socie
´ e
´des Ame
´ icanis es. 1927;
19(1):1–45.
35. La h ap DW. The uppe Amazon. Thames & Hudson Sou hamp on; 1970.
36. Heckenbe ge MJ, Ne es EG, Pe e sen JB. De onde su gem os modelos?: Conside ac¸ões sob e a
o igem e expansão dos Tupi. Re is a de An opologia. 1998; 41(1):69–96. h ps://doi.o g/10.1590/
S0034-77011998000100003
37. Megge s BJ. Clima ic oscilla ion as a ac o in he p ehis o y o Amazonia. Ame ican An iqui y. 1979;
44(2):252–266. h ps://doi.o g/10.2307/279075
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 19 / 25
38. B ochado JJJP. An ecological model o he sp ead o po e y and ag icul u e in o Eas e n Sou h Ame -
ica [PhD hesis]. Uni e si y o Illinois a U bana-Champaign; 1984.
39. Mille ET. A Cul u a Ce a
ˆmica do T onco Tupı
´no al o Ji-Pa ana
´, Rondo
ˆnia, B asil: Algumas e lexões
eo
´ icas, hipo e
´ icas e conclusi as. Re is a B asilei a de Linguı
´s ica An opolo
´gica. 2009; 1(1):35–
136. h ps://doi.o g/10.26512/ bla. 1i1.12288
40. Ne es EG. A chaeological cul u es and pas iden i ies in he p e-colonial Cen al Amazon. In: A H, Hill
J, edi o s. E hnici y in ancien Amazonian: econs uc ing pas iden i ies om A chaeology, Linguis ic
and E nohis o y. Boulde : Uni e si y P ess o Colo ado; 2011. p. 1–27.
41. Ne es WA, Be na do DV, Okumu a M, Fe ei a de Almeida T, S auss AM. O igin and dispe sion o
he Tupigua ani: Wha does c anial mo phology say? Bole im do Museu Pa aense Emilio Goeldi: Ciên-
cias Humanas. 2011; 6(1):95–122. h ps://doi.o g/10.1590/S1981-81222011000100007
42. Walke RS, Wichmann S, Mailund T, A kisson CJ. Cul u al phylogene ics o he Tupı
´language amily
in lowland Sou h Ame ica. PLOS One. 2012; 7(4):e35025. h ps://doi.o g/10.1371/jou nal.pone.
0035025 PMID: 22506065
43. G ego io de Souza J, Noelli F, Madella M. Reassessing he ole o clima e change in he Tupi expan-
sion (Sou h Ame ica, 5000-500 BP). Jou nal o he Royal Socie y. 2021; 18(183):20210499. h ps://
doi.o g/10.1098/ si .2021.0499 PMID: 34610263
44. Lis JM, G eenhill SJ, G ay RD. The po en ial o au oma ic wo d compa ison o his o ical linguis-
ics. PLOS One. 2017; 12(1):e0170046. h ps://doi.o g/10.1371/jou nal.pone.0170046 PMID:
28129337
45. Rod igues AD. A classi icac¸ão do onco lingu¨ı
´s ico upi. Re is a de An opologia. 1964; 12(1/2):99–
104. h ps://doi.o g/10.11606/2179-0892. a.1964.110739
46. Megge s BJ. A cheological and e hnog aphic e idence compa ible wi h he model o o es agmen a-
ion. In: P ance G, edi o . Di e si ica ion in he opics. New Yo k: Columbia Uni e si y P ess; 1982. p.
483–496.
47. Noelli FS. Jose
´P oenza B ochado, ida acadêmica e a A queologia Tupi. In: P ous AP, And ade Lima
T, edi o s. Os ce amis as Tupigua ani. Volume I—Sı
´n eses Regionais. Belo Ho izon e: Sigma; 2008.
p. 17–47.
48. Sil a FA, Noelli FS. A queologia e Linguı
´s ica: cons uindo as aje o
´ ias his o
´ ico-cul u ais dos po os
Tupı
´. C ı
´ ica e Sociedade: Re is a de Cul u a Polı
´ ica. 2017; 7(1):55–87.
49. Rod igues AD. Hipo
´ ese sob e as mig ac¸ões dos ês subconjun os me idionais da amı
´lia Tupi-Gua-
ani. In: A as do II Cong esso Nacional da ABRALIN. Flo iano
´polis: Associac¸ão B asilei a de Lingu¨ı
´s-
ica; 2000. p. 1596–1605.
50. O’Hagan Z, Chousou-Polydou i N, Michael L. Phylogene ic classi ica ion suppo s a No heas e n
Amazonian P o o-Tupı
´-Gua anı
´homeland. LIAMES: Lı
´nguas Indı
´genas Ame icanas. 2019; 19:
e019018–e019018.
51. Rod igues AD, Die ich W. On he linguis ic ela ionship be ween Mawe
´and Tupı
´-Gua anı
´. Diach o-
nica. 1997; 14(2):265–304. h ps://doi.o g/10.1075/dia.14.2.04 od
52. Co êa-da Sil a BC. Hipo
´ eses sob e a His o
´ ia Lingu¨ı
´s ica ka’apo
´ ; 2000. II Cong esso da Associac¸ão
B asilei a de Lingu¨ı´s ica e XIV Ins i u o Lingu¨ı´s ico.
53. D ude S. On he posi ion o he Awe ı
´language in he Tupı
´ amily. In: Gua ani y “Mawe i-Tupı
´-Gua-
anı
´”. Es udios his o icos y desc ip i os sob e una amilia lingu¨is ica de Ame ica del Su . Be lin: Li Ve -
lag; 2006. p. 11–45.
54. Co êa-da Sil a BC. Mais undamen os pa a a hipo
´ ese de Rod igues (1984/1985) de um P o o-
Awe ı
´-Tupı
´-Gua anı
´. In: RODRIGUES ASAC AD; Cab al, edi o . Lı
´nguas e Cul u as Tupı
´. ol. 1. Cam-
pinas, SP: Nimuendaju
´; 2007. p. 219–240.
55. Co êa-da Sil a BC. Mawe
´/Awe ı
´/Tupı
´-Gua anı
´: Relac¸ões Linguı
´s icas e Implicac¸ões His o
´ icas [PhD
hesis]. Uni e sidade de B ası
´lia; 2011.
56. Mei a S, D ude S. A summa y econs uc ion o P o o-Mawe ı
´-Gua anı
´segmen al phonology. Bole im
do Museu Pa aense Emı
´lio Goeldi Ciências Humanas. 2015; 10(2):275–296. h ps://doi.o g/10.1590/
1981-81222015000200005.
57. Rod igues AD. E idence o Tupi-Ca ib ela ionships. Sou h Ame ican Indian languages: e ospec
and p ospec . 1985; 371:404.
58. Noelli FS. A ocupac¸ão humana na Região Sul do B asil: A queologia, deba es e pe spec i as-1872-
2000. Re is a USP. 1999; 44:218–269.
59. Smi h ME. The Expansion o he Az ec Empi e: A Case S udy in he Co ela ion o Diach onic A chaeo-
logical and E hnohis o ical Da a. Ame ican An iqui y. 1987; 52(1):37–54. h ps://doi.o g/10.2307/
281059
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 20 / 25
60. Ma sh EJ, Kidd R, Ogbu n D, Du a
´n V. Da ing he Expansion o he Inca Empi e: Bayesian Models
om Ecuado and A gen ina. Radioca bon. 2017; 59(1):117–140. h ps://doi.o g/10.1017/RDC.2016.
118
61. Pa
¨ ssinen M, Sii ia
¨inen A. Inka-S yle Ce amics and Thei Ch onological Rela ionship o he Inka
Expansion in he Sou he n Lake Ti icaca A ea (Boli ia). La in Ame ican An iqui y. 1997; 8(3):255–271.
h ps://doi.o g/10.2307/971655
62. Noelli F, Sallum M. Po uma his o
´ ia da linguagem da Ce a
ˆmica Paulis a. Re is a B asilei a de Linguı
´s-
ica An opolo
´gica. 2021; 13:367–396. h ps://doi.o g/10.26512/ bla. 13i01.40664
63. Noelli FS, B ochado JP,Co êa A
ˆA. A linguagem da ce a
ˆmica Gua anı
´: sob e a pe sis ência das p a
´ i-
cas e ma e ialidade (pa e 1). Re is a B asilei a De Linguı
´s ica An opolo
´gica. 2018; 10(2):167–200.
h ps://doi.o g/10.26512/ bla. 10i2.20935
64. De Cas o EV. A awe e
´—os deuses canibais. Jo ge Zaha ; 2012.
65. Pa¨ ssinen M. Tequinho Geoglyph Si e and Ea ly Polych ome Ho izon BC 500/300—AD 300/500 in he
B azilian S a e o Ac e. Amazo
ˆnica—Re is a de An opologia. 2021; 13(1):177–220.
66. Hogg AG, Hea on TJ, Hua Q, Palme JG, Tu ney CSM, Sou hon J, e al. SHCal20 Sou he n Hemi-
sphe e Calib a ion, 0–55,000 yea s cal BP. Radioca bon. 2020; p. 1–20. h ps://doi.o g/10.1017/RDC.
2020.59
67. de Almeida FO. O complexo Tupi da Amazo
ˆnia O ien al [MA hesis]. Uni e si y o São Paulo. São
Paulo; 2008.
68. de Almeida FO, Ne es EG. E idências a queolo
´gicas pa a a o igem dos Tupi-Gua ani no les e da
Amazo
ˆnia. Mana. 2015; 21:499–525. h ps://doi.o g/10.1590/0104-93132015 21n3p499
69. Pa
¨ ssinen M. Quando comec¸ou, ealmen e, a expansão gua ani em di ec¸ãoàs Se as Andinas O ien-
ais? Re is a de A queologia. 2005; 18:51–66. h ps://doi.o g/10.24885/sab. 18i1.204
70. Scheel-Ybe R, Maca io K, Bua que A, Anjos RM, Beauclai M. A new age o an old si e: he ea lies
Tupigua ani se lemen in Rio de Janei o S a e? Anais da Academia B asilei a de Ciências. 2008;
80:763–770. h ps://doi.o g/10.1590/S0001-37652008000400015 PMID: 19039497
71. Pe o a C. As da ac¸ões do C-14 dos sı
´ ios a queolo
´gicos do Espı
´ i o San o. Re is a da Cul u a UFES.
1975; 4(6).
72. Ma in G. P e
´-His o
´ ia do No des e do B asil. Reci e: Ed. UFPE; 1997.
73. Albuque que M. Recipien es ce a
ˆmicos de g upos Tupi, no No des e B asilei o. In: P ous A, Lima TA,
edi o s. Os Ce amis as Tupigua ani. Belo Ho izon e: Sigma; 2008. p. 55–78.
74. Chmyz I. Se
´ imo ela o
´ io do P oje o A queolo
´gico I aipu. Cu i iba; 1983.
75. G eenhill SJ, Hegga y P, G ay RD. Bayesian phylolinguis ics. The Handbook o His o ical Linguis ics.
2020; 2:226–253. h ps://doi.o g/10.1002/9781118732168.ch11
76. Ho mann K, Bouckae R, G eenhill SJ, Ku¨hne D. Bayesian phylogene ic analysis o linguis ic da a
using BEAST. Jou nal o Language E olu ion. 2021; h ps://doi.o g/10.1093/jole/lzab005
77. Lis JM, Walwo h M, G eenhill SJ, T esoldi T, Fo kel R. Sequence compa ison in compu a ional his o -
ical linguis ics. Jou nal o Language E olu ion. 2018; 3(2):130–144. h ps://doi.o g/10.1093/jole/lzy006
78. Fo kel R, Lis JM, G eenhill SJ, Rzymski C, Bank S, Cysouw M, e al. C oss-Linguis ic Da a Fo ma s,
ad ancing da a sha ing and e-use in compa a i e linguis ics. Scien i ic Da a. 2018; 5(180205):1–10.
h ps://doi.o g/10.1038/sda a.2018.205 PMID: 30325347
79. Chang W, Ca hca C, Hall D, Ga e A. Ances y-cons ained phylogene ic analysis suppo s he
Indo-Eu opean s eppe hypo hesis. Language. 2015; 91(1):194–244. h ps://doi.o g/10.1353/lan.2015.
0007
80. Lis JM, G eenhill SJ, T esoldi T, Fo kel R. LingPy. A Py hon lib a y o quan i a i e asks in his o ical
linguis ics; 2019. A ailable om: h ps://doi.o g/10.5281/zenodo.3554103.
81. Lis JM, Lopez P, Bap es e E. Using sequence simila i y ne wo ks o iden i y pa ial cogna es in mul i-
lingual wo dlis s. P oceedings o he 54 h Annual Mee ing o he Associa ion o Compu a ional Lin-
guis ics (Volume 2: Sho Pape s). 2016; 2:599–605. h ps://doi.o g/10.18653/ 1/P16-2097
82. Good J, Cysouw M. Languoid, doculec , and glossonym: Fo malizing he no ion ‘language’. Language
documen a ion & conse a ion. 2013; 7.
83. Cab al ASAC. Con ac -Induced Language Change in he Wes e n Amazon: The Non-Gene ic O igin
o he Kokama Language [PhD hesis]. Uni e si y o Pi sbu gh; 1995.
84. Epps P. Language classi ica ion, language con ac , and Amazonian p ehis o y. Language and Linguis-
ics Compass. 2009; 3(2):581–606. h ps://doi.o g/10.1111/j.1749-818X.2009.00126.x
85. Michael L. On he P e-Columbian o igin o P o o-Omagua-Kokama. Jou nal o Language Con ac .
2014; 7(2):309–344. h ps://doi.o g/10.1163/19552629-00702004
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 21 / 25

86. Lis JM, Rzymski C, G eenhill S, Schweikha d N, Pianykh K, Tjuka A, e al.. Concep icon 2.5.0; 2021.
A ailable om: h ps://concep icon.clld.o g/.
87. Swadesh M. The o igin and di e si ica ion o language: Edi ed pos mo em by Joel She ze . Chicago:
Aldine; 1971.
88. Tadmo U, Haspelma h M, Taylo B. Bo owabili y and he no ion o basic ocabula y. Diach onica.
2010; 27(2):226–246. h ps://doi.o g/10.1075/dia.27.2.04 ad
89. Ca ling G, La sson F, C onhamn S, Fa en R, Aliye E, Johansson N, e al. Diach onic A las o Com-
pa a i e Linguis ics Online. Lund Uni e si y; 2017. A ailable om: h ps://diacl.h .lu.se/Con en /
documen s/DiACL-lexicology.pd .
90. Rod igues AD. Linguis ic Recons uc ion o Elemen s o P ehis o ic Tupi Cul u e. In: Linguis ics and
A chaeology in he Ame icas. B ill; 2010. p. 1–10.
91. Cummins CA, McIne ney JO. A Me hod o In e ing he Ra e o E olu ion o Homologous Cha ac e s
ha Can Po en ially Imp o e Phylogene ic In e ence, Resol e Deep Di e gence and Co ec Sys em-
a ic Biases. Sys ema ic Biology. 2011; 60(6):833–844. h ps://doi.o g/10.1093/sysbio/sy 064 PMID:
21804093
92. Sy ja
¨nen K, Mau i s L, Leino U, Honkola T, Ro a J, Vesakoski O. C ouching TIGER, hidden s uc u e:
Explo ing he na u e o linguis ic da a using TIGER alues. Jou nal o Language E olu ion. 2021; 6
(2):99–118. h ps://doi.o g/10.1093/jole/lzab004
93. Lis JM. Co ec ing a bias in TIGER a es esul ing om high amoun s o in a ian and single on cog-
na e se s. Jou nal o Language E olu ion. 2022; h ps://doi.o g/10.1093/jole/lzab007
94. Lis JM. A web-based in e ac i e ool o c ea ing, inspec ing, edi ing, and publishing e ymological
da ase s. In: o Compu a ional Linguis ics (EACL) A, edi o . P oceedings o he 15. EACL 2017 So -
wa e Demons a ions, Valencia, Spain, Ap il 3-7 2017; 2017. p. 9–12.
95. Maddison DR, Swo o d DL, Maddison WP. NEXUS: an ex ensible ile o ma o sys ema ic in o ma-
ion. Sys ema ic biology. 1997; 46(4):590–621. h ps://doi.o g/10.1093/sysbio/46.4.590 PMID:
11975335
96. Huson DH, B yan D. Applica ion o phylogene ic ne wo ks in e olu iona y s udies. Molecula Biology
and E olu ion. 2006; 23(2):254–267. h ps://doi.o g/10.1093/molbe /msj030 PMID: 16221896
97. Ge a di FF, T esoldi T. Lexical Phylogene ics o he Tupı
´-Gua anı
´Family Linguis wee s Con e ence
(ABRALIN). 2021.
98. D ummond AJ, Ho SYW, Phillips MJ, Rambau A. Relaxed Phylogene ics and Da ing wi h Con idence.
PLOS Biology. 2006; 4(5):null. h ps://doi.o g/10.1371/jou nal.pbio.0040088 PMID: 16683862
99. Zhang R, D ummond A. Imp o ing he pe o mance o Bayesian phylogene ic in e ence unde elaxed
clock models. BMC e olu iona y biology. 2020; 20:1–28. h ps://doi.o g/10.1186/s12862-020-01609-4
PMID: 32410614
100. Hemming J. Red Gold: The Conques o he B azilian Indians. McMillan; 1979.
101. Koch A, B ie ley C, Maslin MM, Lewis SL. Ea h sys em impac s o he Eu opean a i al and G ea
Dying in he Ame icas a e 1492. Qua e na y Science Re iews. 2019; 207:13–36. h ps://doi.o g/10.
1016/j.quasci e .2018.12.004
102. Bouckae R, Vaughan TG, Ba ido-So ani J, Duchêne S, Fou men M, Ga yushkina A, e al. BEAST
2.5: An ad anced so wa e pla o m o Bayesian e olu iona y analysis. PLOS Compu a ional Biology.
2019; 15(4):1–28. h ps://doi.o g/10.1371/jou nal.pcbi.1006650 PMID: 30958812
103. Huelsenbeck JP. Tes ing a Co a io ide Model o DNA Subs i u ion. Molecula Biology and E olu ion.
2002; 19(5):698–707. h ps://doi.o g/10.1093/ox o djou nals.molbe .a004128 PMID: 11961103
104. Ge nha d T. The condi ioned econs uc ed p ocess. Jou nal o heo e ical biology. 2008; 253(4):769–
778. h ps://doi.o g/10.1016/j.j bi.2008.04.005 PMID: 18538793
105. Heled J, Bouckae RR. Looking o ees in he o es : summa y ee om pos e io samples. BMC
e olu iona y biology. 2013; 13(1):1–11. h ps://doi.o g/10.1186/1471-2148-13-221 PMID: 24093883
106. Le unic I, Bo k P. In e ac i e T ee O Li e (iTOL) 5: an online ool o phylogene ic ee display and
anno a ion. Nucleic Acids Resea ch. 2021; 49(W1):W293–W296. h ps://doi.o g/10.1093/na /gkab301
PMID: 33885785
107. Russel PM, B ewe BJ, Klae e S, Bouckae RR. Model Selec ion and Pa ame e In e ence in Phylo-
gene ics Using Nes ed Sampling. Sys ema ic Biology. 2018; 68(2):219–233. h ps://doi.o g/10.1093/
sysbio/syy050
108. Bouckae RR. Phylogeog aphy by Di usion on a Sphe e. bioRxi . 2015;
109. Bielejec F, Baele G, V ancken B, Sucha d MA, Rambau A, Lemey P. Sp eaD3: In e ac i e Visualiza-
ion o Spa io empo al His o y and T ai E olu iona y P ocesses. Molecula Biology and E olu ion.
2016; 33(8):2167–2169. h ps://doi.o g/10.1093/molbe /msw082 PMID: 27189542
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 22 / 25
110. Bouckae RR. DensiT ee: making sense o se s o phylogene ic ees. Bioin o ma ics. 2010; 26
(10):1372–1373. h ps://doi.o g/10.1093/bioin o ma ics/b q110 PMID: 20228129
111. Bouckae RR, Heled J. DensiT ee 2: Seeing T ees Th ough he Fo es . bioRxi . 2014;
112. Kolipakam V, Jo dan FM, Dunn M, G eenhill SJ, Bouckae R, G ay RD, e al. A Bayesian phylogene ic
s udy o he D a idian language amily. Royal Socie y Open Science. 2018; 5(3):1–17. h ps://doi.o g/
10.1098/ sos.171504 PMID: 29657761
113. Holland BR, Hube KT, D ess A, Moul on V. δplo s: A ool o analyzing phylogene ic dis ance da a.
Molecula Biology and E olu ion. 2002; 19(12):2051–2059. h ps://doi.o g/10.1093/ox o djou nals.
molbe .a004030 PMID: 12446797
114. G ay RD, B yan D, G eenhill SJ. On he shape and ab ic o human his o y. Philosophical T ansac-
ions o he Royal Socie y o London B: Biological Sciences. 2010; 365(1559):3923–3933. h ps://doi.
o g/10.1098/ s b.2010.0162 PMID: 21041216
115. G eenhill SJ, Wu CH, Hua X, Dunn M, Le inson SC, G ay RD. E olu iona y dynamics o language sys-
ems. P oceedings o he Na ional Academy o Sciences. 2017; 114(42):E8822–E8829. h ps://doi.
o g/10.1073/pnas.1700388114 PMID: 29073028
116. Jolkesky MPDV. Es udo a queo-ecolinguı
´s ico das e as opicais sul-ame icanas [PhD hesis]. Uni-
e sidade de B ası´lia; 2016.
117. Dixon RMW, Dixon RMW, Uni e si y C, P ess CU, Robe Malcolm Wa d D. The Rise and Fall o Lan-
guages. Camb idge Uni e si y P ess; 1997.
118. Wagley C, Gal ão E. The Tene eha a Indians o B azil. Columbia Uni e si y P ess; 1949.
119. Co mie L. Kinship wi h monkeys. Columbia Uni e si y P ess; 2003.
120. Ga cia UF. Ka awa a: a cac¸a e o mundo dos Awa
´-Guaja
´[PhD hesis]. Uni e sidade de São Paulo;
2010.
121. Bale
´e W. Cul u al o es s o he Amazon: a his o ical ecology o people and hei landscapes. Uni e -
si y o Alabama P ess; 2013.
122. Co êa-da Sil a BC. U ubu-Ka’apo –da G ama
´ ica àHis o
´ ia: a aje o
´ ia de um po o [M.A. hesis].
Uni e sidade de B ası
´lia; 1997.
123. Nimuendaju
´C. Os ı
´ndios Pa in in in do io Madei a. Jou nal de la Socie
´ e
´des Ame
´ icanis es. 1924;
16:201–278.
124. Die ich W. As lı
´nguas Tupi-Gua ani boli ianas e o conjun o Kawahiwa: no as hipo
´ eses sob e as o i-
gens. Con luência. 2021; p. 258–295.
125. Menezes Bas os RJd. Ri ual, his o
´ ia e polı
´ ica no Al o Xingu: obse ac¸ões a pa i dos Kamayu a
´e do
es udo da es a da jagua i ica (Jawa i). In: Heckenbe ge M, edi o . Os po os do Al o Xingu: his o
´ ia e
cul u a. Rio de Janei o: Edi o a da UFRJ; 2001. p. 335–357.
126. D ude S. Awe ı
´in ela ion wi h Kamayu a
´: The wo Tupian languages o he Uppe Xingu. In: Al o
Xingu. Uma sociedade mul ilı
´ngu¨e. Museu do Indio-FUNAI; 2011. p. 155–192.
127. Ba bosa AS, Schmi z PI, Ne o AT, Gomes H. O pia da ju i i pepena: na a i a ecolo
´gica da ocupac¸ão
humana do ce ado. Edi o a PUC Goia
´s; 2014.
128. Rod igues PdM. Os A a
´-Canoei o do A aguaia e o empo do ca i ei o. Anua
´ io An opolo
´gico. 2013;
I:83–137.
129. Eh en eich P. Ma e ialien zu Sp achenkunde B asiliens.(Fo se zung). Zei sch i u¨ E hnologie.
1895; 27:149–176.
130. Julião S, Risolê a M. Aspec s mo phosyn axiques de l’anambe
´[PhD hesis]. Uni e si e
´Toulouse-Le
Mi ail. Toulouse; 2005.
131. Cab al ASAC, Solano EdJB. Mais Fundamen os pa a a Hipo
´ ese de P oximidade Gene
´ ica do a a-
we e
´com lı
´nguas do sub- amo V da amı
´lia Tupı
´-Gua anı
´(Fu he Founda ions o he Hypo hesis o
Gene ic P oximi y o he A awe e
´Language o he Languages o sub-se V o he Tupı
´. Es udos da Lı
´n-
guagem. 2006; 4(1):41.
132. Baldus H. Tapi ape
´: ibo upı´no B asil Cen al. B asiliana. 1970;.
133. Coud eau HA. Voyage àI aboca e àl’I acayuna, Ie juille 1897-11 oc ob e 1897. Camb idge Uni e -
si y P ess; 1898[2009].
134. La aia RdB. I
´ndios e cas anhei os: a emp esa ex a i a e os ı
´ndios no Me
´dio Tocan ins. Paz e Te a;
1978.
135. Nimuendaju
´K. The Asu inı
´. In: S ewa d JH, edi o . Handbook o Sou h-Ame ican Indians. ol. 3.
Wes iew; 1948. p. 225–243.
136. de Ba os La aia R. Akua
´wa-Asu inı
´e Su uı
´= Ana
´lise de Dois G upos Tupi. Re is a do Ins i u o de
Es udos B asilei os. 1972; 12:7–30.
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 23 / 25
137. A naud E. Mudanc¸as en e G upos Indı
´genas Tupı
´da Região do Tocan ins-Xingu Bacia Amazo
ˆnica.
Bole im do Museu Pa aense Emı
´lio Goeldi. 1983; 84:1–50.
138. Rod igues AD, Cab al ASAC. Conside a ions on he concep s o language and dialec : a look on he
case o Asu inı
´o Tocan ins and Pa akana. Re is a Vi ual de Es udos da Linguagem—ReVEL. 2009;
3:1678–8931.
139. Sampaio WBdA. Es udo compa a i o sinc o
ˆnico en e o pa in in in ( enha im) e o u u-eu-uau-uau
(amonda a): Con ibuic¸ões pa a uma e isão na classi icac¸ão das lı
´nguas upı
´-kawahib [M.A. hesis].
Uni e sidade de Campinas; 1997.
140. Sampaio W. As lı
´nguas upı
´-kawahib: Um es udo sis ema
´ ico e ilogene
´ ico [PhD hesis]. Uni e sidade
de Rondo
ˆnia; 2001.
141. Aguila AMGC. Con ibuic¸ões pa a os es udos his o
´ ico-compa a i os sob e a di e si icac¸ão do sub-
amo VI da amı
´lia linguı
´s ica Tupı
´-Gua anı
´[PhD hesis]. Uni e sidade de B ası
´lia. B ası
´lia; 2015.
142. K acke WH, I aga’nga M. A posic¸ão his o
´ ica dos Pa in in in na e oluc¸ão das cul u as Tupı´-Gua anı´.
In: Cab al ASAC, Rod igues AD, edi o s. T abalho ap esen ado no Encon o In e nacional sob e as
Lı
´nguas e Cul u as dos Po os Tupi. Labo a o
´ io de Lı
´nguas Indı
´genas, Ins i u o Linguis ico, UNB. B a-
sı
´lia: UNB; 2005. p. 23–35.
143. Ma c¸oli O. Es udo compa a i o dos diale os da lı´ngua kawahib (Tupi-Gua ani) enha im, jiahui e amon-
dawa [M.A. hesis]. Uni e sidade de Campinas; 2018.
144. Mene
´ndez H. Uma con ibuic¸ão pa a a e no-his o
´ ia da a
´ ea Tapajo
´s-Madei a. Re is a do Museu Pau-
lis a São Paulo. 1981; 28:289–388.
145. Me
´ aux A. The na i e ibes o eas e n Boli ia and wes e n Ma o G osso. Bu eau o Ame ican e hnol-
ogy Bulle in; 1942.
146. Rod igues AD. Tupi languages in Rondo
ˆnia and in Eas e n Bolı
´ ia. In: We zels WL, edi o . Language
endange men and endange ed languages. Linguis ic and an h opological s udies wi h special empha-
sis on he languages and cul u es o he Andean-Amazonian bo de a ea. Indigenous Languages o
La in Ame ica. Indigenous Languages o La in Ame ica se ies (ILLA). Leiden: CNWS Publica ions;
2007. p. 355–363.
147. Bale
´e W. En i onmen , cul u e, and Si iono
´plan names. In: Ma i L, edi o . On biocul u al di e si y.
Smi hsonian Ins i u ion P ess; 2001. p. 298–310.
148. Roessle EM. Syn ac ic e ec s o in lec ional mo phology es uc u ing in Ache
´: on language change
and language con ac in Tupı
´-Gua anı
´subg oup-1 = E ei os sin a
´ icos da ees u u ac¸ão de mo ologia
lexional em Achê: um es udo de mudanc¸a linguı
´s ica e eno
ˆmenos de con a o no subg upo-1 da
amı
´lia Tupı
´-Gua anı
´[PhD hesis]. Uni e sidade de Campinas; 2018.
149. Alenca TCd. A he anc¸a da ala: iden idade e
´ nica e memo
´ ia documen al da lı
´ngua Xe a
´( upı
´-gua-
anı
´) [PhD hesis]. Uni e sidade de B ası
´lia; 2013.
150. Rami ez H, Vegini V, de F anc¸a MCV. O Wa a
´zu do Guapo e
´(Tupi-Gua anı´): P imei a desc ic¸ão lin-
guı
´s ica. LIAMES. 2017; 17(2):411–506.
151. Sne hlage MR, Sne hlage AM, Me e G, edi o s. Die Guapo e
´-Expedi ion (1933-1935). Ein Fo schung-
s agebuch. Bo
¨hlau; 2017.
152. Sil a MAC, Nunes K, Lemes RB, Mas-Sando al À, Amo im CEG, K iege JE, e al. Genomic insigh
in o he o igins and dispe sal o he B azilian coas al na i es. PNAS. 2020; 117(5):2372–2377. h ps://
doi.o g/10.1073/pnas.1909075117
153. on Ma ius CF. Bei a
¨ge zu E hnog aphie und Sp achenkunde Ame ikas zumal B asiliens. F ied ich
Fleische ; 1867.
154. Gallois DT. Con ibuic¸ão ao es udo do po oamen o indı
´gena da Guiana B asilei a: um caso especı
´-
ico, os Waiãpi [M.A. hesis]. Uni e sidade de São Paulo; 1980.
155. G enand P. Ainsi pa laien nos ancê es: Essai d’e hnohis oi e Wayãpi. O s om; 1982.
156. Gallois DT. Mig ac¸ão, gue a e come
´ ico: os waiãpi na Guiana. ol. 15. FFLCH-USP; 1986.
157. Rose F. G ammai e del L’E
´me
´ illon Teko, une langue Tupı
´-Gua anı
´de Guyane F anc¸aise. Pee e s;
2011.
158. Be endo JP. C o
ˆnica das missões does pad es da Companhia de Jesus no es ado do Ma anhão.
Edic¸ões do Senado Fede al; 2010.
159. Mau i s L, de Hee M, Honkola T, Dunn M, Vesakoski O. Bes p ac ices in jus i ying calib a ions o da -
ing language amilies. Jou nal o Language E olu ion. 2020; 5(1):17–38. h ps://doi.o g/10.1093/jole/
lzz009
160. Na e E. in oduc ion. In: Couchilli T, Mau el D, edi o s. Con es des indiens e
´me
´ illon. Conseil In e na-
ional de la Langue F anc¸aise; 1994. p. 1–11.
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 24 / 25
161. Rose F. Ele
´men s de phone
´ ique, phonologie e mo phophonologie de l’e
´me
´ illon (Teko) [M.A. hesis].
Uni e si e
´Lyon; 2000.
162. Nimuendaju
´C. T ibes o he lowe and middle Xingu i e . In: Handbook o Sou h Ame ican Indians.
ol. 3. Uni ed S a es Go e nmen P in ing O ice Washing on; 1948. p. 213–243.
163. Nimuendaju C. Mapa-e nohis o
´ ico do B asil e egiões adjacen es. Segunda Edic¸ão; 2021.
164. Mene
´ndez MA. A a
´ ea Madei a-Tapajo
´s: si uac¸ão de con a o e elac¸ões en e colonizado e indı
´-
genas. In: da Cunha MC, edi o . His o
´ ia dos ı
´ndios no B asil. Companhia das Le as São Paulo;
1992. p. 281–296.
165. Seki L. G ama
´ ica do kamaiu a
´: Lı´ngua upı´-gua anı´do al o Xingu. Edi o a da UNICAMP; 2000.
166. Heckenbe ge MJ. The ecology o powe : cul u e, place and pe sonhood in he sou he n Amazon, AD
1000–2000. Rou ledge; 2004.
167. Napoli ano MF, DiNapoli RJ, S one JH, Le in MJ, Jew NP, Lane BG, e al. Ree alua ing human coloni-
za ion o he Ca ibbean using ch onome ic hygiene and Bayesian modeling. Science Ad ances.
2021; 5(12):eaa 7806. h ps://doi.o g/10.1126/sciad .aa 7806
168. Lis JM. Beyond cognacy: His o ical ela ions be ween wo ds and hei implica ion o phylogene ic
econs uc ion. Jou nal o Language E olu ion. 2016; 1(2):119–136. h ps://doi.o g/10.1093/jole/
lzw006
169. Noelli FS, Vo e GC, San os MCP, Pa ei DD, Campos JB. N
˜ande eko: he undamen als o Gua anı
´
adi ional en i onmen al knowledge in sou he n B azil. Vege a ion His o y and A chaeobo any. 2021;
p. 1–17.
170. Sallum M, Noelli FS. Poli ics o Rega d and he Meaning o Things: The pe sis ence o ce amic and
ag o o es y p ac ices by women in São Paulo. In: Lee M Panich SLG, edi o . The Rou ledge Hand-
book o he A chaeology o Indigenous-Colonial In e ac ion in he Ame icas. Rou ledge; 2021. p. 338–
356.
171. da Col G, de Cas o EV. The p oblem o a ini y in Amazonia. HAU: Jou nal o E hnog aphic Theo y.
2018; 8(1-2):347–393. h ps://doi.o g/10.1086/698527
PLOS ONE
Lexical phylogene ics o he Tupı
´-Gua anı
´ amily
PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0272226 June 15, 2023 25 / 25