[back to list of Latin materials]
Householder (1971) [ Note 1] notes that "a morphemic analysis seems to depend very strongly on how the raw data is chopped up. The same paradigmatic set of forms may often be sliced in three different ways, one which makes the stem invariant and the affixes variable, one which makes the affixes firm and the stems variable, and a third which inserts a variable zero-element or process in between two non-varying morphemes." He gives an example of these three "styles" from Latin verbs:
================= Conjugation ===================================
Style | I | II | III |
=================|===============|================|==============|
(a) | port-o: | mon-eo: | reg-o: |
Variant Affix | port-a:s | mon-e:s | reg-is |
(Conjugational | port-at | mon-et | reg-it |
or Morphological)| | | |
-----------------|---------------|----------------|--------------|
(b) | port- o: | mone- o: | reg- o: |
Variant Stem | porta:-s | mone:-s | regi-s |
| porta- t | mone- t | regi-t |
-----------------|---------------|----------------|--------------|
(c) | port- -o: | mon-e -o: | reg- -o: |
Invariant Stem | port-a:-s | mon-e:-s | reg-i-s |
& Suffix | port-a -t | mon-e -t | reg-i-t |
(Phonological | | | |
or Process) | | | |
------------------------------------------------------------------
For some of the later exercises in this series, we will need an analysis that has invariant suffixes, which we get from either Style (b) or (c). [Note 2] Initially, however, we want to make only one cut in each word, between each lexical stem and its person-number suffixes. This then points to an analysis of Style (b).
Use as your data for Latin 1 the verbs displayed for each conjugation in A Student's Latin Grammar (page 28): I porto:, II doceo:, III traho:, and IV audio:. Use a postposed colon, rather than the superposed macron, to mark vowel length (as in these examples).
Solution to Latin 1. After you have cut off a set of invariable suffixes, you will be left with two or three variable stems for each verb. Present these stems and the suffixes in a table with the same format as is used for the "Variant Stem Solution" given in Latin Verb Inflection. Answer the following question: Of what possible value is a Variant Stem Solution? Cite any advantages or disadvantages. In addition (optional), experiment with permutations of the rows and columns in your matrix, if you wish, and comment on any optimal arrangement that you find.
[back to list of Latin materials]
Starting with the variant stems you found in Latin 1, decide on one of the variants that can best be used as a citation stem for each verb. The form you decide on should contain phonological substance that points to the conjugation class of the verb without ambiguity. [Note 3] It should also come from the same person-number combination for each verb, so that simple instructions can be given for finding the stem from a full verb-form. Such an instruction might say, for example, "To find the citation stem for any verb, drop the -t suffix from its 3s form." (This, incidentally, would not be a good choice. Can you explain why?)
Solution to Latin 2. As your solution, present your four citation stems, the set of invariant suffixes, and any other necessary instructions for using them to reconstitute the four paradigms, in a set of simple prose statements (use tables to save words where possible). Do not include the original data above, or any other worksheets with your solution. Answer the following questions: (1) Why would dropping the -t from 3s forms not be a good way to arrive at citation stems? (2) Why is a solution like this sometimes referred to as a "phonological solution?" (3) Of what possible value is such a solution? Cite any advantages or disadvantages.
[back to list of Latin materials]
Again, use as your data the verbs displayed for each conjugation in A Student's Latin Grammar (page 28): I porto:, II doceo:, III traho:, and IV audio:. The solution wanted here is one of Style (a), one that has invariant stems and variant suffixes.
Solution to Latin 3. Present your solution succinctly in table form. Of what possible value is such a solution? Cite any advantages or disadvantages.
[back to top of Latin 1] [back to list of essays]
Subject the perfect active indicative forms of the same four verbs used in Latin 1 to the same style of analysis. Take your data from A Student's Latin Grammar (p. 29). Note that this proves to be a simpler task, since no stem variants need result. That is to say, it is possible to analyze the perfect forms in such a way that both invariant stems and invariant suffixes result.
Compare these perfect stems with the citation
stems decided upon for each verb in Latin 1. Does there seem to
be a simple and straightforward way of predicting the perfect
from the present?
Present Perfect
porta:- porta:v-
etc.
This problem is completed when you have determined the four perfect stems, and the invariant set of PERFECT ACTIVE INDICATIVE suffixes that go with them. This will be one of our sets of suffixes to be used in Latin 7 below. [back to top of Latin 4]
[back to list of Latin materials]
Add to the list of pairs of present and perfect stems developed in Latin 4 other pairs, including especially the variety to be found in Conjugation III. These can be taken from any reference grammar, as for example pp. 109-115 of A Student's Latin Grammar. [Note 4] As you add to the list, begin to classify the pairs according to the formal differences between the members of the pair.
For example, the pair porta:-porta:v- would be classed as "Add v", trahi-tra:cs as "Drop stem-vowel, add s, lengthen root vowel," and carpi-carps- as "Drop stem-vowel, add s."
Note that to classify trahi-tra:cs in this way, it is necessary to posit a phonological rule, informally: hs > cs. By doing so, we can end up with fewer classes. Keep a list of all such rules you find it necessary to posit. Do not posit a rule unless it seems to be quite natural, something that we would expect to happen based on what happens generally in natural languages.
Proceed to work your way through the whole list of stem-pairs available to you, classifying each one and refining your classification as you proceed. The aim will be to sort the entire list of pairs so as to bring together each type. This a place where the sorting feature of a word processor can be of help. Familiarize yourself with the sorting capabilities of the word processor available to you (if any). [Note 5] As you refine your classification, you may find that it falls logically into a scheme that can be summarized by an outline. One level might concern whether or not the stem vowel is retained, another whether -s , -v/u, or no suffix is added. This second level might also include the option of reduplication, since it and suffixation appear to be mutually exclusive. (Features that stand in a disjoint, "or" relation can be part of the same sort.) Sorting by the two levels just mentioned would already separate the stem-pairs into major groupings. Additional sorting levels should be added to further sort out other sub-varieties of perfect stem formation, including vowel length, vowel modification, metathesis of r, and so forth. The aim is to sort the entire list of pairs so as to bring together each type, those that form their perfect stems with exactly the same combination of modifications.
Make special note of any verbs for which there is more than one perfect stem. Do we find any examples of the converse, verbs with just one perfect stem, but which have more than one present stem? What might this indicate about the relation between the two, as to whether the perfect was formed from the present, or vice versa?
As a result of our work on Latin up to this point, we have pairs of stems. From one of the pair we can form all the present tense inflections; from the other the perfect. Presumably the learner of Latin would need learn only the pair and their meaning, and from them would be able to inflect fully for the present and the perfect. From what you have seen in this exercise, do you think it would be possible to simplify the learning task further by developing rules for forming one member of the pair from the other? Which member of the pair should be considered basic? (Consider in this connection your answers to the questions in the last paragraph.) What would such rules look like? What would the basic forms look like? What would the learner have to know or learn in addition to the basic form and its meaning in order to determine the full present and perfect paradigms?
After you have achieved a perfect sort, the
remaining task is to describe it and its results in some
insightful way. Can you identify the weak verbs? [Note 6] Can you identify major and minor
patterns of strong verbs? Are there ways to tell which pattern is
followed by looking at the present stem?--which is to ask, how
much of perfect stem formation (if any) is phonologically
conditioned? [back to
top of Latin 5] [back to list of essays]
[back to list of Latin materials]
Fit IIIa capio: into your final matrix of Latin 1 in a column between III and IV. Use equals signs to the right and/or left of the variant stem forms of capio: in the matrix to show where the stem endings are identical to those of either traho: or audio:, or both.
The resultant matrix should have equals signs to both the right and the left of the 3S form of capio, indicating that all three forms (the 3s of traho:, capio:, and audio:) end in short i-, and that therefore the traditional conjugation membership of verbs such as these cannot be discerned from looking at their 3s forms:
III IIIa IV
trahi- = capi- = audi-
Indo-European and Latin scholars tell us that IIIa verbs were originally members of IV (see for example the note at the bottom of page 99 in New Latin Grammar). [Note 7] This means that capio: and other IIIa verbs as well earlier had variable stems and were inflected just like audio: (and the other IV verbs). Can you posit a scenario whereby they came to be partially confused with (and inflected like) the verbs of III? Hint: use the concept of "analogy," and give examples of the analogies involved in your scenario with the format used by Greenberg in his chapter on "Linguistic Change." [Note 8]
Your answer to the above should include two possible analogies faced by the Latin speaker attempting to inflect a Latin IV verb for some person-number combination other than 3s, a correct one, and an "incorrect" one. For example, consider facio: 'make, do', originally a member of IV. Let's say the speaker wants to say the equivalent of 'we make', to inflect it for 1p. The correct analogy will involve another member of IV, such as audio:; the incorrect analogy will involve a member of III, such as traho::
3s : 1p = 3s : 1p
CORRECT: audit : audi:mus = facit : x (faci:mus)
INCORRECT: trahit : trahimus = facit : x (facimus)
If the speaker (and others like him/her) makes correct analogies, facio: continues to be a member of IV, but if incorrect analogies are made, a MORPHOLOGICAL CHANGE takes place (as actually happened historically).
For Latin 4, give a clear and concise explanation as to how verbs of conjugation IIIa arose, replete with several such analogies, spelled out. [back to top of Latin 6] [back to list of essays]
[back to list of Latin materials]
Our task in Latin 7 will be to describe Latin verb inflection, which consists of a large paradigm containing 90 cells for each verb. [Note 9] Our data are given in Chapter 7 of A Student's Latin Grammar. A total of 20 subparadigms are identified in the overview on page 27. We will ignore five of these, since they are accomplished periphrastically rather than morphologically. They are at the intersection of PERFECT and PASSIVE; see if you can identify them (both in the overview on page 27 and in the pages of data that follow) and tag them somehow for omission from your work on this problem. [Note 10]
The chart "Schematic of Latin Verb Inflection" is furnished to help you get your bearings (and keep them!) in this massive paradigm, and to simplify your task by giving you the beginnings of a solution. Together with the five present-perfect pairs of stems for the exemplar verbs, the boxes in the chart make up the basic Items for an IP solution. Your task is to describe as economically, exhaustively, and elegantly as possible the rules for combining the items so as to accurately populate the 90 paradigm cells for each exemplar verb. Refer to the verbs by their present stem endings, rather than by their traditional conjugation numbers. Although phonological conditioning is generally preferable to morphological conditioning, it may be necessary to settle for morphological conditioning in some instances. [Note 11] When a given suffix always seems to have the same effect, it may be more rightfully a characteristic of the suffix as suffix, rather than as assembly of certain phones.
One other suggestion to point you toward insight and elegance. Two of the increments as hypothesized consist not of any substance (any phonemes as such), but of their effect on the preceding stem vowels. They are termed "Vowel Shift" and "Future Vowel Shift," and their details (as you will see) involve a certain switching between back and front vowels (for which you should give the details). These vowel shifts constitute the seeds of a phenomenon that has come to fruition in Spanish, identified by Matthews [Note 12] as "vowel reversal," where the switching between indicative and subjunctive can be viewed as a toggling between e and a, depending on the verb class, rather than either vowel being identified with either mood across the board. [back to top of Latin 7]
[back to list of Latin materials]
Read the first two sections of Word and Paradigm at this web site. Develop paradigmatic strategies for each of the 15 "corners" of the inflectionall paradigm of a regular verb from one of the four conjugations (as indicated by your instructor), using the same format.
[back to list of Latin materials]
Develop paradigmatic strategies for each of the 6 "corridors" of the inflectional paradigm of a regular verb from one of the four conjugations (as indicated by your instructor), using the same format.
[back to list of Latin materials]
It is an instructive exercise to minimize the
morphological conditioning and maximize the phonological
conditioning in the inflection of Latin nouns. The traditional
declensions cut off suffixes immediately after the root, so that
the different theme vowels create five different sets of
suffixes-the five declensional endings-that differ in some
measure just in these vowels. Phonological conditioning can be
maximized by cutting further to the right, so that theme vowels
remain with the roots to form stems that end in each of the five
vowels-plus one type that has no vowel, which ends in a
consonant. Then one no longer needs to learn the declension of a
noun; its type is visible in the final phoneme of its stem. Stems
are best based on the genitive singular, the case that is
traditionally cited along with the nominative. Nominative stems
often differ from genitive stems; sometimes the nominative stem
can be predicted, but for some nouns a second, variant stem must
also be given. Create 1) a list of noun stems, plus variant stems
as necessary, 2) a table of suffixes that includes a column for
each of the six stem types, and 3) a set of sandhi rules and any
other rules that prove necessary. Permute the rows and columns of
your suffix table to juxtapose instances of syncretism and other
similarities. Include all the nouns whose paradigms are given in A
Student's Latin Grammar plus the additional 2nd and 3rd
declension paradigms listed below, taken from Allen and
Greenough's New Latin Grammar.
puella
Fem.
'girl'
Nom. sg. puella
Gen. sg. puellae
Dat. sg. puellae
Acc. sg. puellam
Abl. sg. puella:
Nom. pl. puellae
Gen. pl. puella:rum
Dat. pl. puelli:s
Acc. pl. puella:s
Abl. pl. puelli:s
servus deus vir puer magister
Masc. Masc. Masc. Masc. Masc.
'slave' 'god' 'man' 'boy' 'teacher'
Nom. sg. servus (-os) deus vir puer magister
Gen. sg. servi: dei: viri: pueri: magisteri:
Dat. sg. servo: deo: viro: puero: magistero:
Acc. sg. servum (-om) deum virum puerum magisterum
Abl. sg. servo: deo: viro: puero: magistero:
Voc. Sg. serve
Nom. pl. servi: dei:/di: viri: pueri: magisteri:
Gen. pl. servo:rum deo:rum/deum viro:rum/virum puero:rum magistero:rum
Dat. pl. servi:s dei:s/di:s viri:s pueri:s magisteri:s
Acc. pl. servo:s deo:s viro:s puero:s magistero:s
Abl. pl. servi:s dei:s/di:s viri:s pueri:s magisteri:s
ager Other 2nd decl. nouns: Other 2nd decl. nouns:
Masc. Those that do not have e Those that have e
'field' in stem; like ager: in stem; like puer:
Nom. sg. ager adulter 'adulterer' arbiter 'judge'
Gen. sg. agri: socer 'father-in-law' cancer 'crab'
Dat. sg. agro: gener 'son-in-law' caper 'goat'
Acc. sg. agrum vesper 'evening' coluber 'snake'
Abl. sg. agro: culter 'knife'
Nom. pl. agri: faber 'smith'
Gen. pl. agro:rum liber 'book'
Dat. pl. agri:s minister 'servant'
Acc. pl. agro:s
Abl. pl. agri:s
templum bellum
Neut. Neut.
'temple' 'war'
Nom. sg. templum bellum
Gen. sg. templi: belli:
Dat. sg. templo: bello:
Acc. sg. templum bellum
Abl. sg. templo: bello:
Nom. pl. templa bella
Gen. pl. templo:rum bello:rum
Dat. pl. templi:s belli:s
Acc. pl. templa bella
Abl. pl. templi:s belli:s
pri:nceps ra:dics mi:les leo: merca:tor
Com. Fem. Masc. Masc. Masc.
'chief' 'root' 'soldier' 'lion' 'merchant'
Nom. sg. pri:nceps ra:dics mi:les leo: merca:tor
Gen. sg. pri:ncipis ra:di:cis mi:litis leo:nis merca:to:ris
Dat. sg. pri:ncipi: ra:di:ci: mi:liti: leo:ni: merca:to:ri:
Acc. sg. pri:ncipem ra:di:cem mi:litem leo:nem merca:to:rem
Abl. sg. pri:ncipe ra:di:ce mi:lite leo:ne merca:to:re
Nom. pl. pri:ncipe:s ra:di:ce:s mi:lite:s leo:ne:s merca:to:re:s
Gen. pl. pri:ncipum ra:di:cum mi:litum leo:num merca:to:rum
Dat. pl. pri:ncipibus ra:di:cibus mi:litibus leo:nibus merca:to:ribus
Acc. pl. pri:ncipe:s ra:di:ce:s mi:lite:s leo:ne:s merca:to:re:s
Abl. pl. pri:ncipibus ra:di:cibus mi:litibus leo:nibus merca:to:ribus
cu:sto:s re:cs du:cs cor caput genus
Com. Masc. Masc. Neut. Neut. Neut.
'guard' 'king' 'leader' 'heart' 'head' 'race'
Nom. sg. cu:sto:s re:cs du:cs cor caput genus
Gen. sg. cu:stodis re:gis ducis cordis capitis generis
Dat. sg. cu:stodi: re:gi: duci: cordi: capiti: generi:
Acc. sg. cu:stodem re:gem ducem cor caput genus
Abl. sg. cu:stode re:ge duce corde capite genere
Nom. pl. cu:stode:s re:ge:s duce:s corda capita genera
Gen. pl. cu:stodum re:gum ducum -- capitum generum
Dat. pl. cu:stodibus re:gibus ducibus cordibus capitibus generibus
Acc. pl. cu:stode:s re:ge:s duce:s corda capita genera
Abl. pl. cu:stodibus re:gibus ducibus cordibus capitibus generibus
corpus aequor no:men tempus co:nsul
Neut. Neut. Neut. Neut. Masc.
'body' 'sea' 'name' 'time' 'consul'
Nom. sg. corpus aequor no:men tempus co:nsul
Gen. sg. corporis aequoris no:minis temporis co:nsulis
Dat. sg. corpori: aequori: no:mini: tempori: co:nsuli:
Acc. sg. corpus aequor no:men tempus co:nsulem
Abl. sg. corpore aequore no:mine tempore co:nsule
Nom. pl. corpora aequora no:mina tempora co:nsule:s
Gen. pl. corporum aequorum no:minum temporum co:nsulum
Dat. pl. corporibus aequoribus no:minibus temporibus co:nsulibus
Acc. pl. corpora aequor no:mina tempora co:nsule:s
Abl. pl. corporibus aequoribus no:minibus temporibus co:nsulibus
virgo: pater
Fem. Masc.
'maiden' 'father'
Nom. sg. virgo: pater
Gen. sg. Virginis patris
Dat. sg. virgini: patri:
Acc. sg. virginem patrem
Abl. sg. virgine patre
Nom. pl. virgine:s patre:s
Gen. pl. virginum patrum
Dat. pl. virginibus patribus
Acc. pl. virgine:s patre:s
Abl. pl. virginibus patribus
sitis turris i:gnis imber
Fem. Fem. Masc. Masc.
'thirst' 'tower' 'fire' 'rain'
Nom. sg. sitis turris i:gnis imber
Gen. sg. sitis turris i:gnis imbris
Dat. sg. siti: turri: i:gni: imbri:
Acc. sg. sitim turrim (-em) i:gnem imbrem
Abl. sg. siti: turri: (-e) i:gni: (-e) imbri: (-e)
Nom. pl. -- turre:s i:gne:s imbre:s
Gen. pl. -- turrium i:gnium imbrium
Dat. pl. -- turribus i:gnibus imbribus
Acc. pl. -- turri:s (-e:s) i:gni:s (-e:s) imbri:s (-e:s)
Abl. pl. -- turribus i:gnibus imbribus
sedi:le animal calcar mare
Neut. Neut. Neut. Neut.
'seat' 'animal' 'spur' 'sea'
Nom. sg. sedi:le animal calcar mare
Gen. sg. sedi:lis anima:lis calca:ris maris
Dat. sg. sedi:li: anima:li: calca:ri: mari:
Acc. sg. sedi:le animal calcar mare
Abl. sg. sedi:li: anima:li: calca:ri: mari:
Nom. pl. sedi:lia anima:lia calca:ria maria
Gen. pl. sedi:lium anima:lium calca:rium marium
Dat. pl. sedi:libus anima:libus calca:ribus maribus
Acc. pl. sedi:lia anima:lia calca:ria maria
Abl. pl. sedi:libus anima:libus calca:ribus maribus
nu:be:s ci:vis urbs nocs
Fem. Com. Fem. Fem.
'cloud' 'citizen' 'city' 'night'
Nom. sg. nu:be:s ci:vis urbs nocs
Gen. sg. nu:bis ci:vis urbis noctis
Dat. sg. nu:bi: ci:vi: urbi: nocti:
Acc. sg. nu:bem ci:vem urbem noctem
Abl. sg. nu:be ci:ve urbe nocte
Nom. pl. nu:be:s ci:ve:s urbe:s nocte:s
Gen. pl. nu:bium ci:vibus urbium noctium
Dat. pl. nu:bibus ci:vibus urbibus noctibus
Acc. pl. nu:bi:s (-e:s) ci:ve:s urbi:s (-e:s) nocti:s (-e:s)
Abl. pl. nu:bibus ci:vibus urbibus noctibus
clie:ns aeta:s
Masc. Fem.
'client' 'age'
Nom. sg. clie:ns aeta:s
Gen. sg. clientis aeta:tis
Dat. sg. clienti: aeta:ti:
Acc. sg. clientem aeta:tem
Abl. sg. cliente aeta:te
Nom. pl. cliente:s aeta:te:s
Gen. pl. clientium (-um) aeta:tum (-ium)
Dat. pl. clientibus aeta:tibus
Acc. pl. clienti:s (-e:s) aeta:ti:s (-e:s)
Abl. pl. clientibus aeta:tibus
iter nics senecs caro: bo:s
Neut. Fem. Masc. Fem. Com.
'journey' 'snow' 'old man' 'flesh' 'ox, cow'
Nom. sg. iter nics senecs caro: bo:s
Gen. sg. itineris nivis senis carnis bo:vis
Dat. sg. itineri: nivi: seni: carni: bovi:
Acc. sg. iter nivem senem carnem bovem
Abl. sg. itinere nive sene carne bove
Nom. pl. itinera nive:s sene:s carne:s bove:s
Gen. pl. itinerum nivium senum carnium boum
Dat. pl. itineribus nivibus senibus carnibus bo:bus (bu:bus)
Acc. pl. itinera nive:s sene:s carne:s bove:s
Abl. pl. itineribus nivibus senibus carnibus bo:bus (bu:bus)
os vi:s
Neut. Fem.
'bone' 'force; strength (in pl.)'
Nom. sg. os vi:s
Gen. sg. Ossis vi:s
Dat. sg. ossi: vi:
Acc. sg. os vim
Abl. sg. osse vi:
Nom. pl. ossa vi:res
Gen. pl. ossium vi:rium
Dat. pl. ossibus vi:ribus
Acc. pl. ossa vi:ris (-es)
Abl. pl. ossibus vi:ribus
manus domus genu:
Fem. Fem. Neut.
'hand' 'house'(irregular) 'knee'
Nom. sg. manus domus genu:
Gen. sg. manu:s domu:s genu:s
Dat. sg. manui: domui: genu:
Acc. sg. manum domum genu:
Abl. sg. manu: domo: genu:
Nom. pl. manu:s domu:s genua
Gen. pl. manuum domuum/domo:rum genuum
Dat. pl. manibus domibus genibus
Acc. pl. manu:s domu:s/domo:s genua
Abl. pl. manibus domibus genibus
die:s re:s
Masc. Fem.
'day' 'thing'
Nom. sg. die:s re:s
Gen. sg. die:i: rei:
Dat. sg. die:i: rei:
Acc. sg. diem rem
Abl. sg. die: re:
Nom. pl. die:s re:s
Gen. pl. die:rum re:rum
Dat. pl. die:bus re:bus
Acc. pl. die:s re:s
Abl. pl. die:bus re:bus
[back to list of Latin materials]