.Il presente volume in onore della prof.sa Annarita Puglielli
è stato stampato con il contribuito del
DIPARTIMENTO DI LINGUISTICA
DELL’UNIVERSITÀ DEGLI STUDI DI ROMA TRE
Sempre in onore della prof.sa Annarita Puglielli, è inoltre pubblicato
presso L’Harmattan Italia / L’Harmattan (Paris) il volume A Country
Called Somalia: Culture, Language and Society of a Vanishing State,
curato da Mara Frascarelli
© L’Harmattan Italia srl, 2011STRUCTURES A D MEA I GS:
Dipartimento di Linguistica
Università degli Studi Roma Tre
L’Harmattan Italia L’Harmattan
via Degli Artisti 15 5-7 rue de L’École Polytechnique
10124 Torino 75005 Paris.Table of contents
Mara Frascarelli 7
Section I: Language change and language variation
Tullio De Mauro 11
LAGUAGE, LAGUAGES: FORMAL CODITIOS
OF COVERGECE AD DIVERGECE
Chiara Fedriani, Gianguido Manzelli, Paolo Ramat 21
EXPERIETIAL COSTRUCTIOS I
THE CIRCUM-MEDITERRAEA AREA
Edoardo Lombardi Vallauri 38
SEMATIC AD SYTACTIC VARIATIO WITH ITALIA
IFIITIVES ITRODUCED BY “DI” AD “A”:
A PRELIMIARY RECOGITIO
Franca Orletti 61
FORMULATIG IDETITIES I MULTICULTURAL
Section II: The structures of meaning
Guglielmo Cinque 75
GREEBERG’S UIVERSAL 23 AD SVO LAGUAGES
Roland Hinterhölzl 81
HOW PROSODY RESTRICTS THE SYTAX
Lunella Mereu 100
OMIAL PREDICATIO AD
THE SYTAX-SEMATIC ITERFACE
Anna Pompei 124
RELATIVE CLAUSES BETWEE COORDIATIO
5Francesca Ramaglia 142
ATTRIBUTIVE ADJECTIVES AT THE SYTAX-SEMATICS
ITERFACE: COMPARIG ITALIA AD EGLISH
Section III: Applied linguistics and language teaching
Serena Ambroso 165
LAGUAGE RESEARCH AD TRAIIG PROJECTS:
AARITA PUGLIELLI’S COTRIBUTIO
Elisabetta Bonvino 178
FROM CROSS-LIGUISTIC COMPARISO
TO COSCIOUS LEARIG: A OVEL PERSPECTIVE
Biancamaria Tedeschini Lalli 192
AARITA’S BEGIIGS AS A GEERAL
AD APPLIED LIGUIST
Carla Vergaro 198
LIGUISTICS AS A WIDOW O COTRASTIVE RHETORIC
AD RHETORICAL TYPOLOGY
UNIVERSITY OF ROMA TRE
This volume is not simply a collection of papers that happen
to be published together on the occasion of a Festschrift. This
work is the outcome of a real desire: the desire to celebrate
Annarita Puglielli for what she means as a linguist, as a proj-
ect leader and, most of all, as a person. For years I have nur-
tured this plan in my heart, and finally the book is here, exact-
ly as I wanted it to be.
It was a pleasure for me to picture it in my mind: its title, sec-
tions, format, cover, colours, everything. In this respect, it was
a privilege to work with Elisa Pelizzari and the L’Harmattan
publishing company. She liked my project, supported it from
the very beginning and helped me with her invaluable sugges-
tions. My greatest thanks go to her.
And it was a pleasure to contact the (prospect) contributors,
a joy to receive their enthusiastic acceptance, despite the haste
and hurry we all experience and the hundreds of jobs we have
to carry out. They have all been friendly, collaborative and
(mostly!) punctual in submitting their papers. I thank them all!
Thirteen contributors at the end – of necessity a strict selec-
tion, to be honest. In fact, to delineate the structure and inter-
nal composition of this volume was not easy at all, I must con-
fess. Annarita’s intense professional life (as well as her heart!)
has been always divided between two major interests: linguis-
tic investigation for theoretical and applied research on the one
hand and a passionate engagement in everything pertaining the
cultural, social and (of course) linguistic aspects of the Somali
nation on the other. The two fields would often overlap and the
people involved might also be the same. However, I realized
that it was necessary to keep them separate in order to obtain
an internally coherent volume that might qualify not only as a
7‘gift’, but also (and importantly) as a valid scientific contribu-
tion. Hence, I decided to dedicate STRUCTURES A'D MEA'I'GS:
CROSS-LI'GUISTIC PERSPECTIVES to linguistic analysis exclusive-
ly, and leave the ‘Somali project’ for a distinct work.
Different subjects have been approached and discussed by
the authors of this volume. In particular, Section I is concerned
with «Language change and language variation», both from a
diachronic and a synchronic perspective; Tullio De Mauro,
Paolo Ramat (with Chiara Fedriani and Gianguido Manzelli),
Edoardo Lombardi Vallauri and Franca Orletti have con-
tributed to this part. The authors of Section II («The structure
of meanings»), on the other hand, investigate the connection
between structure and meaning, especially focusing on inter-
face analysis (Roland Hinterhölzl, Lunella Mereu, Francesca
Ramaglia) and cross-linguistic comparison (Guglielmo
Cinque, Anna Pompei). Finally, Section III is dedicated to
applied research in linguistics and, in particular, to «Applied
linguistics and language teaching»; here the authors present the
results of recent investigations (Elisabetta Bonvino, Claudia
Vergaro), also making specific reference to Annarita’s activity
in this field (Serena Ambroso, Bianca Maria Tedeschini Lalli).
Other people also responded promptly to my request for col-
laboration on this volume. I’m referring here to Peter Douglas
and Giorgio Testa, who checked the English of authors who
asked me to, and to Francesca Ramaglia who, besides writing
her own paper, read all the others checking for typos, cross-ref-
erences and formatting. Their help was invaluable to me.
Thanks, my friends!
A special mention in this foreword must go to the Rector of
Roma Tre, Professor Guido Fabiani, who sponsored this vol-
ume. It was delightful to have his appreciation of my initiative
and obtain his immediate positive answer to my request. I truly
and deeply thank him (and the Institution he represents).
Finally, I wish to thank Annarita for having filled my heart
with joy as I write this foreword. Meeting Annarita was a turn-
ing point in my life. She started my interest in linguistics,
8opened my mind to new ways of thinking and taught me more
than I can ever realize. With her I learnt how to investigate phe-
nomena, how to search for data, how to be rigorous and strict
in analysis. And, at the same time, I enjoyed the pleasure of
infinite discussions, the cheerfulness of our talks, the releasing
power of laughter. Indeed, I have always admired her extraor-
dinary capacity to combine scholarship with a special gift for
human relationships. Her tireless organizing capacity despite
(too) often adverse conditions. Her caring and protective atti-
tude, without being sugary or over-permissive. Trusting, but
attentive. A tough mother – this is the way I feel about her.
So many things done together, so many still to be done…
Come on Annarita: what’s next?
With greatest love,
Language change and
Tullio De Mauro
LA GUAGE, LA G UAGES:
FORMAL CO DITI O S OF CO VERGE CE
A D DIVERGE CE
UNIVERSITY OF ROME «LA SAPIENZA»
Introduced by Charles Sanders Peirce and Ferdinand de
Saussure in the second half of the nineteenth century, semiotic
studies (or semiotics or semiology) have been neglected by a
good number of linguists – whether historical, descriptive
(general/theoretical) or applied – as well as by most philoso-
phers and hermeneuticians. Even less attention has been given
by linguists to two key areas of semiotic studies: zoosemiotics
(or animal communication) which burgeoned during the sec-
ond half of the twentieth century, and the study of sign lan-
guages for the deaf, which flourished only in the last part of the
twentieth century, despite the recommendations of the great
American linguist William D. Whitney a century before.
Semiotics has revealed how many of the aspects that lin-
guists consider to be characteristic of natural human languages
– their use, their structures, their formal (generative) function-
ing – are in fact present in other semiotic codes and languages
that go to make up mankind’s semiotic universe. Indeed, they
are present in non-human and non-natural codes and lan-
guages, including programs regulating mechanical and auto-
11The present paper aims at shedding light on a phenomenon
that seems to be highly characteristic of human verbal lan-
1guages: their great multiplicity. It would seem that this phe-
nomenon – as well as the multiplicity of usages of a given
mother tongue over time in the same masse parlante (i.e., in
the same community of speakers who converge in their usages)
– is not shared by other natural languages. How can we recon-
cile the immense multiplicity of human verbal languages on
our planet with the biological naturalness and apparently
innate character of language in the human species?
This paper will contribute to identifying, insofar as possible,
those characteristics or formal properties of human verbal lan-
guage/languages which explain the immense variety of mother
tongues in the world and their intrinsic changeability. These
characteristics may then shed light on the intimate relationship
between human languages and human cultures and on the
development of the latter.
To that end, section 2 will review the pertinent constitutive
properties of all languages – and, indeed, of the universe of lin-
guistic codes lying far beyond strictly verbal language. These
properties will be enunciated in the form of statements, framed
in such a way as to facilitate falsification.
Section 3 will list – concisely – the formal properties that
languages present from a semiotic standpoint.
Section 4 will show how these properties, taken together,
make it inevitable that languages (mother tongues) vary enor-
mously and that each particular language (mother tongue) will
vary to some extent – in time and in space – within the com-
munity that adopts it; moreover, these variances will be inti-
mately connected with the diversity of human cultures.
In the cosmos every recognizable entity acts on others and is
subjected to their actions; this provokes changes in the state of
rest (or of movement) of the entities.
12This reciprocal action, while known for centuries, received a
name only in the nineteenth century: “interaction” (1832) in
English, then interaction in French (1876), and so on.
Interactions are, in general, regulated by causality in the
atomic, molecular, macrocosmic and subatomic worlds.
Living matter is not immune to chemical-physical causal
interactions. But with a difference: living entities intervene in
order to orient the interaction in such a way as to maintain an
equilibrium (homeostasis), thus conserving the entities’ exis-
tence. In other words, their survival requires constant interac-
tion aimed at homeostasis. They must “take advantage” of
causal circumstances to achieve that end.
The resulting interactions carry out various functions, such
as sensing, moving, feeding, reproducing.
In various living species, the individual entities achieve
many of these functions by cooperating with others of their
species; this produces aggregations of various kinds and com-
plexity: for example, swarms, herds, flocks, and societies reg-
ulated by constriction or cooperation.
Zoosemiotics and applied animal ethology have shown that in
an extremely large number of living species (from unicellular
organisms to insects, reptiles, birds, superior mammals), what
we may call “semiotic” interactions can be detected. They occur
especially in social and cooperative species.
The semiotic function coordinates sensorial input and move-
ment in living entities in order to produce and receive objects
called signals, generally exchanged among members of a species.
While the production and reception of objects functioning as
signals are causal events, they can have (and often do have)
effects that go beyond mere causality. The signal function
enhances causal interaction. Take, for example, the camouflage
that a virus assumes to ward off antibodies; or the cry signal-
ing danger made by a bird that averts the flock of an approach-
ing predator; or the crying of an infant in a nursery, calling for
attention; or the cry “Fire!” in a crowded theater. These are all
examples of simple signals that produce complex reactions:
13antibodies that begin to behave differently, scores of birds that
suddenly change their path, a chorus of cries from the other
babies in the nursery (the imitation reflex described by Piaget),
the rush to exit the theater made by hundreds of spectators.
These rapid, almost instantaneous changes of large numbers of
living entities would be generally impossible to attain with
purely causal interactions or, if attainable, would require con-
siderable time and an enormous amount of energy. We may call
this phenomenon “semiotic interaction.”
Signals have two aspects: their expression, consisting in the
change of state of a physical medium called the “channel” (air,
luminosity, environment, etc.), and that which the signal trans-
mits (or which we presume that it transmits), i.e. its “sense” or
“signification” (in French, sens, signification) or, as it is more
frequently called, “meaning”.
Ordinary observation as well as (at least) two accredited the-
ories – the mathematical theory of information and the theory
of classes – show that a signal does not exist in isolation. It
exists as an alternative to at least one other potential signal: in
the simplest case, “no change,” i.e. the conservation of the state
of the channel with no modification of the physical state. Often
a signal exists along with a large number of other potential sig-
nals, or possible changes of state, in the given channel.
In order for a signal to be exist as such, it must be produced
and received by living entities as that particular signal. This
means that the entities must have, in their memories, a classi-
ficatory apparatus or device, capable of distinguishing that sig-
nal from other possible ones. This in turn means that the living
entities must have a classification-based device capable of pro-
ducing, as well as recognizing, that signal as different from
another or others.
If we consider that the essential function of the thinking
process (which we may call the “mind” or, in other languages,
pensée or esprit, mente or spirito, Intelligenz, etc.) is to identi-
fy, differentiate and classify, then we must admit that thinking
is not a uniquely human capacity. Other living species are, in
14fact, capable of those operations. As Karl Popper once said,
from the perspective of the entire universe, the amoeba is only
a step from Einstein.
A number of theoretical linguists have found it useful to con-
ceptualize and distinguish verbally two aspects of signals: their
concrete “expression” hic et nunc and their general “meaning”
(the abstract class each one belongs to, in opposition to other
classes that it does not belong to). In addition, these linguists
distinguish between the “signal” in itself (concrete, individual)
and the “sign,” (the abstract class with which a given given sig-
nal is or is not associated).
The classificatory device is called a “code.”
In order for perception/reception/comprehension to be
appropriate, a code must be shared – at least to some extent.
Minimal sharing is guaranteed by membership in the same
species; in social species, such sharing is augmented by social
Codes consist of expression forms and content forms. The
first are the set of expressions which the code is able to identi-
fy and differentiate. The second are the meanings that the code
is able to identify and differentiate.
A “noetic field” is the set of possible meanings recognized by
a given code or given codes.
Codes having similarities constitute families of codes or
The codes making up a language have noetic fields of limit-
ed scope: the meanings they call up are situated within a pre-
cisely defined area of possible experiences. Rare are the lan-
guages with open noetic fields: they are generally non-conven-
tional elaborations, such as human verbal languages (e.g.
Neapolitan sign language) but also certain religious iconolo-
gies (typically paleochristian).
By “open” is meant that these languages can evoke, through
their words, utterances and texts, potentially infinite meanings
From the description just given of the semiotic character of
signs, it is possible to discern general properties that are large-
ly present in every instance (although not absolutely so, since
they coexist with alternative properties). A few examples fol-
Signs in a language (words, sentences) are generally composed of units of
signifier+meaning called “morphs”, which in turn are generally composed of
distinctive, identifiable, asemantic units (syllables and phonemes). Note
however that signs functioning as interjections are often extraneous to pri-
mary and secondary articulation.
(b.1) The overall meaning of most signs can be pieced together from the
meanings of the constituent parts (e.g., words). Note however that a certain
number of signs are amalgamations (compound words, complex or multi-
word lexemes, idiomatic expressions) and are often opaque, i.e. the overall
meaning cannot be deduced from the constituent parts.
(b.2) What is more, the overall meaning of a complex sign can be deter-
mined, in speech, by supersegmental intonation patterns and in writing by
layout and graphical representation (characters used, punctuation, underlin-
(b.3) Lastly, the compositional nature of signs can also be obfuscated in the
realization of distinctive, identifying units of secondary articulation: exten-
sive redundancy and relaxed production/reception rules make global realiza-
tion and reception necessary for such signs.
(c) In languages which distinguish nouns from verbs, the predication realized
by signs is mostly carried out by the verbs. Note however that noun/verb lan-
guages can also realize predications without a finite verb (e.g., so-called
(d) Many signs do not constitute predications; they carry out spatial-tempo-
ral or circumstantial deictic functions or naming functions (such as signs,
titles, name plates, listings).
(e) Certain signs – morphs and lexemes – have a strictly determined mean-
ing and only that meaning is, in theory, possible; the most typical case are
words designating numbers. Note however that even number words can, in
rare cases, be vague: for example, paucals (in particular in Arabic or Hopi,
but also in other languages – for example, “fare due passi”/ “fare quattro
passi”, “mangiare due spaghetti” in Italian), or multals (which, to be exact,
are degree-quantifiers more than actual numbers).
(f.) Most morphs and lexemes (and thus phrases) admit synonymy, whether
predictable and determinable, or unpredictable and determinable only by
16(g) Most lexemes can be used to designate either general species (universals)
or particular meanings.
(h) By custom or by an explicit convention, a part of the morphs and lexemes
can be used like number words, that is, to indicate given meanings exclu-
sively. This special use is tied to particular trades, professions, technical
fields or scientific domains.
(i) Otherwise morphs and lexemes – as well as words and phrases – are, gen-
erally speaking, indeterminate. This characteristic, together with the highly
determinate meanings that morphs and lexemes assume as part of a conven-
tional language, has the following consequences:
(i.1.) meanings are potentially extensible;
(i.2.) the noetic field of a language is unlimited;
(i.3.) meanings can be highly restricted in the case of special uses of lan-
(i.4.) morphs, lexemes, words and phrases have, in any case, the permanent
possibility of semantic variance.
(j) Lexemes and phrases (and every single part) can be used to designate
themselves; this is their so-called autonymic function and can occur either
linguistically (in suppositio formalis, to borrow a term from Scholastic logic)
or metalinguistically (in suppositio materialis).
(k.) Identical signifiers can have, as their signified, meanings that are diverse
because of the diversity of the paradigmatic series of morphs (this is the case
of textual homonyms) or because of the diversity of the synonymic relations
(this is the case of lexical homonyms).
4. Meaning understanding
Given the characteristics of signals (section 2) and signs
(section 3), the attribution of any given sense to a (syntagmat-
ic) utterance in any given language, can by no means be auto-
matic. Understanding meanings is never automatic. A meaning
is a probabilistic approximation that depends not only on the
quality of the perception of the utterance, but also on the qual-
ity of the passage from signal to sign and thus the recognition
of a given verbal string in a given language; in addition, mean-
ing depends on the receiver’s knowledge of the conditions of
enunciation of the utterance: the cultural context, the habits
and physical/psychological state of the sender, the setting, the
probable communicative intent, etc.
Of course, utterances expressed in a special (conventional)
language come close to permitting automatic understanding, in
17that the form of the utterance weighs heavily (in the case of
technical and scientific discourse) and even exclusively (in the
case of mathematical calculations) in determining meanings.
What lies outside the form becomes less important. But even in
this case, understanding is never fully automatic. For one
thing, it depends on whether the receiver truly understands and
shares the norms regulating the conventional language used.
Moreover, a full understanding requires, on the part of the
receiver, a certain degree (or even a high degree) of empathy
with the world in which the enunciation arose and thus the abil-
ity to decentre himself or herself into it, as Antonino Pagliaro
noted. Certainly it requires a high degree of “tolerance of
ambiguity” or “tolerance upon the field” (Eric Lenneberg) in
dealing with the problems that arise in attempting to grasp
another person’s utterances.
What appears to be phonetically or graphically the same
utterance can be understood differently (and even perceived
differently) according to the greater or lesser degree of empa-
thy, tolerance and, in the final analysis, “convergence” of the
various parties to the communicative act. Convergence also
implies a similar mastery of the resources of the language(s)
Indeed, the progressively increasing divergences in the uses
of a given language – which, over time, can lead to the rise of
a new and different idiom – may initially be rooted simply in
divergences of perception and understanding, due in part to
differences in mastery of the language’s resources. We may
call this the first root of change.
Languages do offer, however, a number of characteristics in
their structure and in their use, which serve to facilitate under-
standing and limit divergences of comprehension. These char-
acteristics, while sharing a common principle, vary from one
language to another as well as over time and space. They are:
(a) grammaticality, which permits anchoring an utterance – as it progres-
sively assumes the form of a syntagm in the mind of the receiver – to the cir-
cumstances of time, space and interlocution;
18(b) morphemic/lexical/grammatical creativity, which permits a language
community to add or eliminate components, thus increasing or decreasing
semantic coverage of given areas and thereby achieving a more desirable
specification of meanings (this characteristic can also be called “non non-
creativity” in opposition to the “non-creativity” of formalized and logico-
(c) lexico-semantic and syntactic redundancy, which occurs in the structure
of the language (e.g., alternative forms, synonymy) and in the linearity of
Understanding is also facilitated by the insertion of utter-
ances in a sequence of other utterances, i.e., in a dialog or in a
text The insertion can be signaled explicitly through the gram-
matical apparatus as well as by discourse signals. Dialogs and
texts are part and parcel of languages; they call upon the
receiver to make a supplementary effort to decentre, empathize
and show tolerance.
The ability to produce and understand utterances is visibly
intertwined with the ability of senders and receivers to partici-
pate, to varying degrees, in the organization of a “culture of
survival,” in the creation of technical innovations and “disci-
plines,” in the production of works of intellectual creativity
(music, pictorial and representational art, literature, science) –
the three forms of culture that Kant distinguished.
The varying degrees to which users posses the resources of a
language and its potential reflexive/metalinguistic uses, condi-
tions their access to and participation in the various forms of
culture of their community. This in turn conditions the degree
to which they possess the language and are able to pro-
duce/receive linguistic objects and carry out metalinguistic
reflection. It is in this sense that, in describing a language, we
describe a form of life (Wittgenstein).
The cultural stratification of the members of a given com-
munity, the diversity of their personal experiences and of their
experience of the “common” cultural forms, all go to deter-
mine a different degree of convergence in using and under-
standing the community’s linguistic repertory – which is, as a
consequence, determined precisely by these differences.
19Variations in the “culture of survival” of a community, in its
technological development, in its members’ most creative
intellectual experiences, therefore intertwine with variations in
the language toward which, in varying degrees, the members
converge. And this is the second root, ineradicable, of the vari-
ation of languages in time and space.
1 This paper, originally drafted by T.D.M., is the fruit of discussions with
Patrick Boylan, who then carried out the translation. Silvana Ferreri also con-
tributed extensively. κοινά τά των φίλων.
EXPERIE TIAL CO STRUC TIO S I
THE CIRCUM-MEDITERRA EA AREA
UNIVERSITY OF PAVIA
UNIVERSITY OF PAVIA
Classical languages have dedicated verbs for expressing feel-
ings like hunger or thirst such as Gk. πεινάω, διψάω, Lat.
ēsuriō, sitiō. Both Gk. verbs are denominative (πεῖνα ‘hunger’,
δίψα ‘thirst’). The same holds for sitiō (sitis ‘thirst’), whereas
ēsuriō does not derive from a noun (ēsuriēs is attested only in
Late Latin) but goes back to the verbal root ed- ‘to eat’.
While Modern Greek has maintained πεινώ and διψώ along
with the possibility of the periphrastic constructions έχω πείνα,
έχω δίψα, Romance languages have lost the original Latin
verbs and introduced periphrastic constructions: It. ho fame, ho
sete; Fr. j’ai faim, j’ai soif; Span. tengo hambre, tengo sed;
Port. estou com fome, estou com sede.
The ‘have’-type for expressing many different kinds of pos-
session seems to be an ancient option in Indo-European
(Stassen 2009: 564): cf., e.g., Gk. ἔχειν in constructions such
as ἔχειν χρήµατα ‘to possess goods, to be rich’, and also ἔχειν
πένθος, ἄλγεα ‘to have sorrows, pain’, ἔχειν ἣβην, γῆρaς‘ to
be young, old’ (lit.: ‘to have youth, old age’), Lat. habēre coro-
nam, anulum ‘to have the/a crown, the/a ring’ and also habēre
uulnus, febrim ‘to be wounded, to have a temperature’ (lit.: ‘to
1have a wound, fever’).
Expressions with habēre were fairly frequent already in
Classical Latin, especially for denoting mental states: habēre
21timorem, habēre gaudium etc. instead of timēre, gaudēre etc.
However, we find also habēre frigora, habēre aestum, instead
of frigēre, aestuāre with reference to physical states. As for the
so-called Abstract Possession construction, see 2.2.
Under this point of view the following quote from Seneca
Ep. ad Luc. 19.12 is very interesting:
(1) 'am quod ad illos pertinet apud quos falso divitiarum nomen invasit
occupata paupertas, sic divitias habent quomodo habere dicimus febrem,
cum illa nos habeat. E contrario dicere solemus ‘febris illum tenet’: eodem
modo dicendum est ‘divitiae illum tenent’ [our emphasis]
‘For as far as those persons are concerned, in whose minds bustling poverty
has wrongly stolen the title of rich – these individuals have riches just as we
say that “we have a fever”, when really the fever has us. Conversely, we are
accustomed to say: “A fever grips him”. And in the same way we should say:
“Riches grip him”’ [transl. by Gummere, The Loeb Classical Library]
Seneca’s words testify in a very clear way to the conceptual
2multi-faceted Experiencer (EXP) role. Its multifunctionality sur-
faces at the linguistic level with habēre and tenēre-constructions,
which feature an interesting alternation: on the one hand, the EXP
can be expressed as subject and thus “possesses” the experience,
3as in febrim habeo, ‘I have a fever’; on the other hand, even in
such cases the EXP reacts, rather than acting, feels, but not fully
“owns” the Stimulus (STIM). By contrast, a reverse construction
is also attested, where it is the STIM that, realized as subject, takes
hold of the EXP: examples are desiderium me tenet, aegritudo me
tenet... As is known, similar constructions, where the STIM is per-
sonalized in the nominative case, can be found also in Greek:
τότε καί µιν ἀνἠϖκεστον λάβε ἄλγoς (Il. 3.394)
than and he.ACC incurable took pain.NOM
‘and than an incurable pain took him’
τί δέ σε φρένας ἲkετo πένθος; (Il. 1.362)
what you.ACC heart.ACC reached sadness.NOM?
‘which sadness reached your heart?’