Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

questions on lines 8-11 of the Aeneid

852 views
Skip to first unread message

Ken Quirici

unread,
Mar 17, 2006, 10:19:14 PM3/17/06
to
8 Musa, mihi causas memora, quo numine laeso,
9 quidve dolens, regina deum tot volvere casus
10 insignem pietate virum, tot adire labores
11 impulerit. Tantaene animis caelestibus irae?

My rough translation (partly cribbed from the Perseus English tr, or
at least inspired by it):

Muse, tell me the tale, what divine spirit offended,
or what [divine spirit] feeling aggrieved, drove
the queen of the gods to bring about in their turn
misfortunes on a man renowned for piety,
to endure such hardships. [Are there] so many rages
in the souls of the gods?

is insignem...virum a kind of second accusative with volvere (along
with casus)? or a subject of adire, to endure? What is
the construction of adire - infinitive, but why? Is it, [she]
drove hardships [on] a man, [so that?] he endured hardships?
But what is that construction, if that's it?

And is regina nominative because volvere is kind of a historical
infinitive, even tho it seems that it should be accusative
as the subject of the infinitive phrase with infinitive volvere.

This I think is a crucial few lines for me anyway, because
the construction, much more so than in Catullus, seems
strange, foreign (no pun intended), unknown. Is there a lot
of poetic license, that is, ungrammatical constructions that
'sort of' are used in 'such a way'?

If I can't built a kind of vocabulary of syntax, I'm not going to
get very far in the Aeneid. I like it tho. It seems one can get
used to the strange constructions and almost intuit them.
But I'm sure not there yet. I need to examine these few lines
really thoroughly. Hence my questions!

Thanks for any help.

Ken

B. T. Raven

unread,
Mar 18, 2006, 1:58:30 AM3/18/06
to

"Ken Quirici" <ken.q...@excite.com> wrote in message
news:1142651954.8...@e56g2000cwe.googlegroups.com...

Probably most of the mystery can be dispelled by expanding ellipses:

Mūsa, mihi causās memorā, quō nūmine laesō,
cūrve dolźns, rźgīna deūm tot cāsūs volvere impulerit.
cūrve īnsignem pietāte virum, tot labōrźs adīre impulerit.
Tantaene īrae (sunt) animīs caelestibus?

Eduardus


Ed Cryer

unread,
Mar 18, 2006, 6:26:36 AM3/18/06
to

"B. T. Raven" <eci...@peoplepc.com> wrote in message
news:qcOSf.5822$sL2....@newsread2.news.atl.earthlink.net...

Where have all these "curve"s come from?

The whole thing is very standard classical Latin. As is most of the Aeneid.
"Regina" is nominative singular. The construction with "impulere" is acc. +
inf, as with "iubere" and "prohibere".

Ed

Douglas G. Kilday

unread,
Mar 18, 2006, 1:48:39 AM3/18/06
to

"Ken Quirici" <ken.q...@excite.com> wrote ...

> 8 Musa, mihi causas memora, quo numine laeso,
> 9 quidve dolens, regina deum tot volvere casus
> 10 insignem pietate virum, tot adire labores
> 11 impulerit. Tantaene animis caelestibus irae?

I don't understand why an editor would put commas after 'laeso' or 'dolens'.
The latter is particularly misleading, even if the corresponding English
translation requires it.

> My rough translation (partly cribbed from the Perseus English tr, or
> at least inspired by it):
>
> Muse, tell me the tale, what divine spirit offended,
> or what [divine spirit] feeling aggrieved, drove
> the queen of the gods to bring about in their turn
> misfortunes on a man renowned for piety,
> to endure such hardships. [Are there] so many rages
> in the souls of the gods?

Changing the cases is very misleading. Making Juno the object of a verb
distorts the passage quite far from what Vergil intended.

> is insignem...virum a kind of second accusative with volvere (along
> with casus)? or a subject of adire, to endure? What is
> the construction of adire - infinitive, but why? Is it, [she]
> drove hardships [on] a man, [so that?] he endured hardships?
> But what is that construction, if that's it?

The infinitives and accusatives depend on 'impulerit', which is perfect
subjunctive in indirect discourse, since it in turn depends on the
imperative 'memora'. In direct discourse we would have the following
question:

Quo numine laeso quidve dolens regina deum tot volvere casus insignem
pietate virum, tot adire labores impulit?

By what outraged divine will (i.e. by what instance of outrage to her divine
will), or grieving what, did the queen of the gods compel a man
distinguished for piety to undergo so many hazards, to confront so many
hardships?

> And is regina nominative because volvere is kind of a historical
> infinitive, even tho it seems that it should be accusative
> as the subject of the infinitive phrase with infinitive volvere.

No, 'regina' is nominative because it is the subject. I haven't seen the
Perseus translation, but I suspect you have been led astray by excessive
translatorial license. It's always a good idea to keep subjects and objects
straight in translations, not jumble them around.

> This I think is a crucial few lines for me anyway, because
> the construction, much more so than in Catullus, seems
> strange, foreign (no pun intended), unknown. Is there a lot
> of poetic license, that is, ungrammatical constructions that
> 'sort of' are used in 'such a way'?

The only thing unusual about this construction is the asyndeton, the lack of
a connective between 'tot volvere casus' and 'tot adire labores'. Other
than that, the only peculiarities are the archaic gen. pl. 'deum' and the
sense of 'volvere'. Further on in the Aeneid you will find seriously
"ungrammatical" constructions not used in prose at all, e.g. case-attraction
in 'urbem quam statuo vestra est' and passive used as middle in 'tectus
vultum'.

I find it odd that your translation fails to render 'tot' as 'so many', then
incorrectly renders 'tantae' as 'so many' rather than 'so great'. Perseus
again?

Ken Quirici

unread,
Mar 18, 2006, 11:07:07 AM3/18/06
to
Douglas G. Kilday wrote:
> "Ken Quirici" <ken.q...@excite.com> wrote ...
>
> > 8 Musa, mihi causas memora, quo numine laeso,
> > 9 quidve dolens, regina deum tot volvere casus
> > 10 insignem pietate virum, tot adire labores
> > 11 impulerit. Tantaene animis caelestibus irae?
>
> I don't understand why an editor would put commas after 'laeso' or 'dolens'.
> The latter is particularly misleading, even if the corresponding English
> translation requires it.
>
> > My rough translation (partly cribbed from the Perseus English tr, or
> > at least inspired by it):
> >
> > Muse, tell me the tale, what divine spirit offended,
> > or what [divine spirit] feeling aggrieved, drove
> > the queen of the gods to bring about in their turn
> > misfortunes on a man renowned for piety,
> > to endure such hardships. [Are there] so many rages
> > in the souls of the gods?
>
> Changing the cases is very misleading. Making Juno the object of a verb
> distorts the passage quite far from what Vergil intended.
>

Yes, futher noted below.

> > is insignem...virum a kind of second accusative with volvere (along
> > with casus)? or a subject of adire, to endure? What is
> > the construction of adire - infinitive, but why? Is it, [she]
> > drove hardships [on] a man, [so that?] he endured hardships?
> > But what is that construction, if that's it?
>
> The infinitives and accusatives depend on 'impulerit', which is perfect
> subjunctive in indirect discourse, since it in turn depends on the
> imperative 'memora'. In direct discourse we would have the following
> question:
>
> Quo numine laeso quidve dolens regina deum tot volvere casus insignem
> pietate virum, tot adire labores impulit?
>
> By what outraged divine will (i.e. by what instance of outrage to her divine
> will), or grieving what, did the queen of the gods compel a man
> distinguished for piety to undergo so many hazards, to confront so many
> hardships?
>

This is one of two major critical misreadings on my part. I took
volvere to be active in the sense of 'inflict' hardships, which made
it difficult to see the parallelism with 'adire labores'. I was
using Glare, and I couldn't find any meaning there that would
imply 'endure'. After reading your response, I checked Lewis & Short
online and they actually cited this line of the Aeneid
as a usage of volvere to mean 'endure'.

> > And is regina nominative because volvere is kind of a historical
> > infinitive, even tho it seems that it should be accusative
> > as the subject of the infinitive phrase with infinitive volvere.
>
> No, 'regina' is nominative because it is the subject. I haven't seen the
> Perseus translation, but I suspect you have been led astray by excessive
> translatorial license. It's always a good idea to keep subjects and objects
> straight in translations, not jumble them around.
>

This is the second major misunderstanding. For some reason I
took impulerit's subject to be something like 'divine anger' which
drove 'the queen of the gods to ...'. This misunderstanding might
actually have been caused by the first - if volvere is active,
then regina must be the subject, which means something else
must be the subject of impulerit.

You're right. I think subject/verb might at least initially be the
first targets. It would be nice to eventually read it in order keeping
everything that's 'pending' in my head until it resolves. Like
I read English. But not yet.

> > This I think is a crucial few lines for me anyway, because
> > the construction, much more so than in Catullus, seems
> > strange, foreign (no pun intended), unknown. Is there a lot
> > of poetic license, that is, ungrammatical constructions that
> > 'sort of' are used in 'such a way'?
>
> The only thing unusual about this construction is the asyndeton, the lack of
> a connective between 'tot volvere casus' and 'tot adire labores'. Other
> than that, the only peculiarities are the archaic gen. pl. 'deum' and the
> sense of 'volvere'. Further on in the Aeneid you will find seriously
> "ungrammatical" constructions not used in prose at all, e.g. case-attraction
> in 'urbem quam statuo vestra est' and passive used as middle in 'tectus
> vultum'.
>
> I find it odd that your translation fails to render 'tot' as 'so many', then
> incorrectly renders 'tantae' as 'so many' rather than 'so great'. Perseus
> again?

No those were Ken Quirci. For some strange reason I on
occasion jumble tantus and tot.

Thanks all for your help. This clarifies things immensely.

Ken

Robert Stonehouse

unread,
Mar 18, 2006, 12:46:17 PM3/18/06
to
On 17 Mar 2006 19:19:14 -0800, "Ken Quirici"
<ken.q...@excite.com> wrote:

>8 Musa, mihi causas memora, quo numine laeso,

Muse, bring to my mind the reasons, from what injury to her
divinity,


>9 quidve dolens, regina deum tot volvere casus

or angered at what, did the queen of the gods (to be
inviolved in such disasters)


>10 insignem pietate virum, tot adire labores

a man notable for his piety (to encounter so many toils)


>11 impulerit. Tantaene animis caelestibus irae?

drive on? Can there be such fits of rage in celestial minds?

If you extract the phrases in brackets and put them at the
end of the sentence, you should get something like
intelligible English. It won't do for a translation, of
course, but I am trying to show how it works. The two
phrases are parallel and could be connected by 'and'.

'Quo numine laeso' is more literally 'from what injured
divinity', but it is Juno's divinity that is the point. The
poet is not asking what divinity caused the trouble - it was
Juno. She has been wounded (laesa) and feels pain (dolens) -
at what?


>
>My rough translation (partly cribbed from the Perseus English tr, or
>at least inspired by it):
>
>Muse, tell me the tale, what divine spirit offended,
>or what [divine spirit] feeling aggrieved, drove
>the queen of the gods to bring about in their turn
>misfortunes on a man renowned for piety,
>to endure such hardships. [Are there] so many rages
>in the souls of the gods?
>
>is insignem...virum a kind of second accusative with volvere (along
>with casus)? or a subject of adire, to endure? What is
>the construction of adire - infinitive, but why? Is it, [she]
>drove hardships [on] a man, [so that?] he endured hardships?
>But what is that construction, if that's it?

'Volvere' and 'adire' are parallel; you could put 'and' in
to join them. The construction is accusative with
infinitive, 'virum volvere et adire impulerit', 'she forced
the man to revolve and encounter'.


>
>And is regina nominative because volvere is kind of a historical
>infinitive, even tho it seems that it should be accusative
>as the subject of the infinitive phrase with infinitive volvere.

Too complicated. It's the subject of the sentence, disguised
by artistic word-order.


>
>This I think is a crucial few lines for me anyway, because
>the construction, much more so than in Catullus, seems
>strange, foreign (no pun intended), unknown. Is there a lot
>of poetic license, that is, ungrammatical constructions that
>'sort of' are used in 'such a way'?
>
>If I can't built a kind of vocabulary of syntax, I'm not going to
>get very far in the Aeneid. I like it tho. It seems one can get
>used to the strange constructions and almost intuit them.
>But I'm sure not there yet. I need to examine these few lines
>really thoroughly. Hence my questions!

The beginning of the poem is important. Beginnings are often
difficult because the writer sets out to decorate them and
make them impressive - he certainly succeeds here. This is
not a fair general sample of the syntax of the Aeneid.

I'd suggest learning these few lines by heart and thinking
about them in odd moments - on the bus, say - and in time
they will sort themselves out enough to seem more or less
natural, though not enough to lose the grandeur.

--
Robert Stonehouse
To mail me, replace invalid with uk. Inconvenience regretted

B. T. Raven

unread,
Mar 18, 2006, 6:35:07 PM3/18/06
to

"Ed Cryer" <e...@somewhere.in.the.uk> wrote in message
news:dvgqpk$ms6$1...@emma.aioe.org...

"curve" is a kludge to distribute "impulerit" over both infinitives. But
how do we know that "quid" doesn't mean "why" here also? It's difficult to
prevent polysemy even in Kunstprosa, let alone in verse. If dolens goes
with regina then Quid is the object of the verbal. Verbum dolendi is a
real maverick that seems to complete its meaning with the genetive,
dative, accusative, ablative. Apparently at one time at least you could
say Caput doleo for Caput mihi dolet. I wanted to avoid thinking about
that.

Eduardus

Ken Quirici

unread,
Mar 18, 2006, 7:12:58 PM3/18/06
to
Robert Stonehouse wrote:
> On 17 Mar 2006 19:19:14 -0800, "Ken Quirici"
> <ken.q...@excite.com> wrote:
>
> >8 Musa, mihi causas memora, quo numine laeso,
> Muse, bring to my mind the reasons, from what injury to her
> divinity,
> >9 quidve dolens, regina deum tot volvere casus
> or angered at what, did the queen of the gods (to be
> inviolved in such disasters)
> >10 insignem pietate virum, tot adire labores
> a man notable for his piety (to encounter so many toils)
> >11 impulerit. Tantaene animis caelestibus irae?
> drive on? Can there be such fits of rage in celestial minds?
>
> If you extract the phrases in brackets and put them at the
> end of the sentence, you should get something like
> intelligible English.

And I guess put the 'drive on' before 'a man'?

> It won't do for a translation, of
> course, but I am trying to show how it works. The two
> phrases are parallel and could be connected by 'and'.
>
> 'Quo numine laeso' is more literally 'from what injured
> divinity', but it is Juno's divinity that is the point. The
> poet is not asking what divinity caused the trouble - it was
> Juno. She has been wounded (laesa) and feels pain (dolens) -
> at what?
> >
> >My rough translation (partly cribbed from the Perseus English tr, or
> >at least inspired by it):
> >
> >Muse, tell me the tale, what divine spirit offended,
> >or what [divine spirit] feeling aggrieved, drove
> >the queen of the gods to bring about in their turn
> >misfortunes on a man renowned for piety,
> >to endure such hardships. [Are there] so many rages
> >in the souls of the gods?
> >
> >is insignem...virum a kind of second accusative with volvere (along
> >with casus)? or a subject of adire, to endure? What is
> >the construction of adire - infinitive, but why? Is it, [she]
> >drove hardships [on] a man, [so that?] he endured hardships?
> >But what is that construction, if that's it?
>
> 'Volvere' and 'adire' are parallel; you could put 'and' in
> to join them.

I got the parallelism even tho I thought one's subject was
regina and the other virum :-(

That's a terrific idea! I've already memorized it and am
going over it at every odd moment (well, almost every).

> --
> Robert Stonehouse
> To mail me, replace invalid with uk. Inconvenience regretted

But now a metrical question intervenes:

the scansion of line 8:

Musa, mihi causas memora, quo numine laeso

Working backwards, we have:

laeso - spondee
numine - dactyl
-ra, quo - spondee
-sas memo - dactyl

and I'm stuck with

Musa, mihi cau-

which has 5 beats, which would be

(1) long long long short short or
(2) long short short long long

neither of which seem to fit - (1) fails because the terminal
-a in Musa is short, and (2) fails because the terminal
-hi of mihi is short.

A 'literal' scansion would read

long short short short long

Also, I wonder at memora. In particular, the mem is accented
and there's an instinctive tendency (didn't the Romans have it
too?) to lengthen accented syllables. This is not really a
scansion question; or maybe it is.

Any comments on this stuff?

Thanks.

Ken

J. W. Love

unread,
Mar 18, 2006, 7:51:46 PM3/18/06
to
Ken Quirici wrote:

> and I'm stuck with
> Musa, mihi cau-
> which has 5 beats, which would be
> (1) long long long short short or
> (2) long short short long long
> neither of which seem to fit - (1) fails because the terminal
> -a in Musa is short, and (2) fails because the terminal

> -hi of mihi is short. . . .


> Any comments on this stuff?

"The Latin _mihi_ derives from PIE _*h1meghio,_ with an innovative
Latin ending _-ei_ (i.e. _*h1meghei_). Classical Latin shows _mihi,_
with iambic shortening from an earlier _mihii,_ and an alternate
contracted form _mii_ from enclitic _*h1mei._"---Philip Baldi, _The
Foundations of Latin_ (Berlin and New York: Mouton de Gruyter, 2002),
p. 338. [For the internet, I've had to change the macron-marked vowels
into geminated vowels, and to lower the superscript <h> and raise the
subscript <1>].

So in this verse, the terminal <-hi> of <mihi> is long---and that
solves your problem!

Ben C

unread,
Mar 18, 2006, 8:12:31 PM3/18/06
to
On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:

On c. 50 Vergil wrote:
> Musa, mihi causas memora, quo numine laeso,

> Working backwards, we have:


>
> laeso - spondee
> numine - dactyl
> -ra, quo - spondee
> -sas memo - dactyl

I get the same.

> and I'm stuck with
>
> Musa, mihi cau-
>
> which has 5 beats, which would be
>
> (1) long long long short short or
> (2) long short short long long

It has to be (2).

> Also, I wonder at memora. In particular, the mem is accented and
> there's an instinctive tendency (didn't the Romans have it too?) to
> lengthen accented syllables. This is not really a scansion question;
> or maybe it is.

By accented do you mean stressed?

There is a tendency to stress long syllables in that the penultimate
syllable of a word is usually stressed if it scans long, otherwise it's
usually the third syllable from the end. This works much the same as in
English. But I don't know which came first, the stress or the long
vowel. I would have thought the long vowel, but that's a complete guess.

People do talk, though, about the relationship between stress and ictus.

In my quite simplistic understanding of it, this refers basically to how
often the natural stresses of the words coincide with the starts of
feet. It's much more subtle than that though.

I do think a lot of the natural music of the lines arises out of this
kind of thing, but, in my opinion, it isn't possible or necessary to
analyze it too literally.

Ken Quirici

unread,
Mar 18, 2006, 8:34:21 PM3/18/06
to

It works for me!

Thanks much.

Ken

Ken Quirici

unread,
Mar 18, 2006, 8:56:29 PM3/18/06
to
Ben C wrote:
> On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:
> On c. 50 Vergil wrote:
> > Musa, mihi causas memora, quo numine laeso,
>
> > Working backwards, we have:
> >
> > laeso - spondee
> > numine - dactyl
> > -ra, quo - spondee
> > -sas memo - dactyl
>
> I get the same.
>
> > and I'm stuck with
> >
> > Musa, mihi cau-
> >
> > which has 5 beats, which would be
> >
> > (1) long long long short short or
> > (2) long short short long long
>
> It has to be (2).
>
> > Also, I wonder at memora. In particular, the mem is accented and
> > there's an instinctive tendency (didn't the Romans have it too?) to
> > lengthen accented syllables. This is not really a scansion question;
> > or maybe it is.
>
> By accented do you mean stressed?
>

Yes I do. Latin has both length and accent, but the prosody is
entirely based on length.

> There is a tendency to stress long syllables in that the penultimate
> syllable of a word is usually stressed if it scans long, otherwise it's
> usually the third syllable from the end. This works much the same as in
> English. But I don't know which came first, the stress or the long
> vowel. I would have thought the long vowel, but that's a complete guess.
>
> People do talk, though, about the relationship between stress and ictus.
>

OED, if I interpret it correctly, says ictus is either length or
stress-based.
OTOH, English/American prosody uses the term ictus to refer simply
to the accent mark.

> In my quite simplistic understanding of it, this refers basically to how
> often the natural stresses of the words coincide with the starts of
> feet. It's much more subtle than that though.
>
> I do think a lot of the natural music of the lines arises out of this
> kind of thing, but, in my opinion, it isn't possible or necessary to
> analyze it too literally.

Well, my understanding is that syllable length is not used as a
prosodic entity in English prosody. However it's real, in that I
believe stress, accent, causes length- as tho a person emphasizes
the stress by 'dwelling' on it, lengthening the syllable.

However, I like the 'natural music of the lines' idea - but how
can you determine what it is if you don't analyze it in detail. I'm
not sure what you mean by 'literally' - I take it as meaning,
according to certain rules. In particular, the rules that determine
stress in Latin, and the rules that determine length.

This whole thing about stress vs. length is something I'm still
trying to grasp. When I read a Catullus hendecasyllabic poem
it follows the rules for length-determined meter EXACTLY. It
can't be a purely intellectual exercise that doesn't have any
correspondence in the way the poem is spoken. And then there's
stress which you're also supposed to observe. I throw
my hands up, but still try to do both.

Thanks.

Ken

Ben C

unread,
Mar 20, 2006, 3:09:53 AM3/20/06
to
On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:
> Ben C wrote:
>> On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:

> Well, my understanding is that syllable length is not used as a
> prosodic entity in English prosody. However it's real, in that I
> believe stress, accent, causes length- as tho a person emphasizes the
> stress by 'dwelling' on it, lengthening the syllable.

You're right it does all work quite differently in English scansion. And
as you say, it's likely that stress does have a slight lengthening
effect also in Latin. This doesn't affect the way the line is scanned,
although it will have an affect on the way the line sounds.

> [...] When I read a Catullus hendecasyllabic poem it follows the rules
> for length-determined meter EXACTLY.

So do Vergil's poems follow their meters exactly.

> [...] It can't be a purely intellectual exercise that doesn't have any


> correspondence in the way the poem is spoken.

No of course not, it does have an effect. There are at least five
variables involved that affect each syllable:

1. Length
2. Stress
3. Ictus
4. Intonation
5. Meaning

By "intonation" I mean not just changes in pitch, but also changes in
timing.

Length and stress are properties of the words themselves. Ictus (a
slight accenting of the syllable at the start of each foot) is a
property of how they are fitted into the line, and intonation is a
property of whole sentences or clauses at a time. Intonation is also
affected by meaning.

The poet has to be in control of all these variables, and make them all
work together.

> And then there's stress which you're also supposed to observe. I throw
> my hands up, but still try to do both.

It's much more complicated than just stress and length, in fact I
suspect that a complete technical explanation is not possible. But on
the other hand people have instincts for this kind of thing. After all,
we all speak our own languages naturally without thinking about it.
That's why I think if you just read it the best you can the way it
looks, the more you read the more it will just take shape. Don't
self-consciously emphasize either meter or stress.

Ken Quirici

unread,
Mar 20, 2006, 10:05:31 AM3/20/06
to

The only difficulty I have with the above (apart from putting it into
practice - and it seems eminently useful in that area) is the
question of naturalness. Do you believe the length in Latin syllables
is a natural attribute of the language, or was it to some degree
artificially enhanced to follow Greek models? This is partly
an unfair question, because if you take an English iambic poetry,
write it out as a paragraph of prose, and read it, I'll bet the
iambic nature becomes much less apparent. So maybe the
length in Latin poetry was similarly artificially enhanced as
the reader (out loud) realized he was reading a poem in this or
that meter.

Ars longa, vita brevis!

Thanks.

Ken

Ben C

unread,
Mar 20, 2006, 10:40:28 AM3/20/06
to
On 2006-03-20, Ken Quirici <kqui...@yahoo.com> wrote:
>> No of course not, it does have an effect. There are at least five
>> variables involved that affect each syllable:
>>
>> 1. Length
>> 2. Stress
>> 3. Ictus
>> 4. Intonation
>> 5. Meaning
>>
>> By "intonation" I mean not just changes in pitch, but also changes in
>> timing.
>>
>> Length and stress are properties of the words themselves...

> The only difficulty I have with the above (apart from putting it into
> practice - and it seems eminently useful in that area) is the question
> of naturalness. Do you believe the length in Latin syllables is a
> natural attribute of the language, or was it to some degree
> artificially enhanced to follow Greek models?

I had always assumed it was a natural attribute-- some vowels just are
long, others short, and double consonants are actually pronounced as
double consonants.

Latin poetry uses for the most part Greek meters of course (including
the hexameter), and I understand that some of them are particularly
difficult to realize in Latin.

I think it's more a case of squeezing real Latin the way it really
sounded into the Greek meter than changing the way Latin sounded to make
it more like Greek, if you see what I mean.

But I would be as interested as you to know more about this, so perhaps
someone else will follow up who's more of an expert.

> an unfair question, because if you take an English iambic poetry,
> write it out as a paragraph of prose, and read it, I'll bet the iambic
> nature becomes much less apparent. So maybe the length in Latin poetry
> was similarly artificially enhanced as the reader (out loud) realized
> he was reading a poem in this or that meter.

I guess knowing it's a hexameter is always going to make some
difference. Although I think an experienced reader would spot that and
the hexameter rhythym would just emerge even if it was written out as a
paragraph of prose.

B. T. Raven

unread,
Mar 20, 2006, 11:54:47 AM3/20/06
to

"Ben C" <spam...@spam.eggs> wrote in message
news:slrne1tj2m....@bowser.marioworld...

Robert Sonkowsky is more of an expert. Google on his name and you may be
able to find some sound clips from his readings on the web.

>
> > an unfair question, because if you take an English iambic poetry,
> > write it out as a paragraph of prose, and read it, I'll bet the iambic
> > nature becomes much less apparent. So maybe the length in Latin poetry
> > was similarly artificially enhanced as the reader (out loud) realized
> > he was reading a poem in this or that meter.
>
> I guess knowing it's a hexameter is always going to make some
> difference. Although I think an experienced reader would spot that and
> the hexameter rhythym would just emerge even if it was written out as a
> paragraph of prose.

This is exactly right. As an experiment I collapsed the entire Aeneid into
a single null terminated character string and printed it out (only about
30 pages on both sides of 8.5 x 11). After marking all the macrons, I read
it out loud as prose. The rhythm is, however, much too deeply impressed by
the order of the words for it to pass as good prose.

Eduardus

Douglas G. Kilday

unread,
Mar 20, 2006, 2:09:06 PM3/20/06
to

"J. W. Love" <lov...@aol.com> wrote ...

That's an awfully odd way for Baldi to express it. Umbrian <mehe> 'mihi'
shows that the innovation in question was Proto-Italic *meghei, already in
use before there was any "Latin" distinct from the rest of Italic. And the
laryngeal is posited solely on the evidence of the Greek pronominal stem
<em-> < *h1m-; there is no basis for assuming its retention in Proto-Italic,
much less (Proto-)Latin, so the form *h1meghei is anachronistic.

John W. Kennedy

unread,
Mar 25, 2006, 8:48:41 PM3/25/06
to
Ken Quirici wrote:
> Well, my understanding is that syllable length is not used as a
> prosodic entity in English prosody.

Not in traditional first-level analysis, but it is used as a tool, even
by Pope:
When Ajax strives some rock's vast weight to throw,
The line too labors, and the words move slow;
Not so, when swift Camilla scours the plain,
Flies o'er the unbending corn, and skims along the main.

(Not to mention in Lord Peter Wimsey's famous:
And dying so, sleep that sweet sleep no more.)

And it used to full extent from time to time, as in Bridges':
Now in wintry delights, and long fireside meditation,
'Twixt studies and routine paying due court to the Muses,
My solace in solitude, when broken roads barricade me
Mudbound, unvisited for months with my merry children,
Grateful t'ward Providence, and heeding a slander against me
Less than a rheum, think of me today, Dear Lionel, and take
This letter as some account of Will Stone's versification.
or Tennyson's rather sour:
These lame hexameters the strong-wing’d music of Homer!
No—but a most burlesque barbarous experiment.
When was a harsher sound ever heard, ye Muses, in England?
When did a frog coarser croak upon our Helicon?
Hexameters no worse than daring Germany gave us,
Barbarous experiment, barbarous hexameters.

--
John W. Kennedy
"But now is a new thing which is very old--
that the rich make themselves richer and not poorer,
which is the true Gospel, for the poor's sake."
-- Charles Williams. "Judgement at Chelmsford"

Neeraj Mathur

unread,
Mar 28, 2006, 5:39:01 AM3/28/06
to

Ben C wrote:
> On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:
> > Also, I wonder at memora. In particular, the mem is accented and
> > there's an instinctive tendency (didn't the Romans have it too?) to
> > lengthen accented syllables. This is not really a scansion question;
> > or maybe it is.
>
> By accented do you mean stressed?
>
> There is a tendency to stress long syllables in that the penultimate
> syllable of a word is usually stressed if it scans long, otherwise it's
> usually the third syllable from the end. This works much the same as in
> English. But I don't know which came first, the stress or the long
> vowel. I would have thought the long vowel, but that's a complete guess.
>
> People do talk, though, about the relationship between stress and ictus.
>
> In my quite simplistic understanding of it, this refers basically to how
> often the natural stresses of the words coincide with the starts of
> feet. It's much more subtle than that though.

First things first: the long and short vowels of Latin are absolutely
an integral feature of the language from its earliest times to the
development of the Romance tongues. The stress-based accent with the
rule about the length of the penultimate syllable can only have arisen
if length were already a part of the language.

Besides which, there is ample evidence that Latin used to have a rather
different accentuation before the system of the Classical period: it
earlier had the very simple rule that all words were stressed on their
first syllable. This caused changes to the vowels within words, namely
that all short vowels in open syllables in middle syllables tended to
become -i- (except -u- around labial consonants, and -e- in front of r)
and all short vowels in closed internal syllables (ie. those that were
followed by two or more consonants) became -e-. You will be familiar
with these alterations (I'm using the colon after a vowel to indicate
that it was long):
1) capio:
2) incipio: (note the change of -a- to -i-; the vowel in front of the r
is e)
3) factus
4) perfectus (note that the -a- changes to -e- because it's followed by
two consonants).

These kinds of changes would only have occurred if the first syllable
of every word got the accent.

Now, it can also be seen that long vowels were not affected by these
changes:
5) ce:do:
6) disce:do:
7) a:ctus
8) coa:ctus (note that this is not *coectus like perfectus above).

So long vowels existed long before the Classical stress pattern based
on the length of the second-last syllable came into effect.

Now, about ictus. 'Ictus' can best be considered like the downbeat in
music; appliec to the hexameter line, it basically refers to the
emphasis given to these syllables which must be long (that is, the
first syllable of every foot - one of the differences between the Latin
and Greek hexameters is that Latin measures the hexameter as six feet,
where each foot is a dactyl or a spondee, while the Greek line of
Homer, Hesiod and the Hymns has an analysis in terms of cola). When the
syllable of the word that is accented does not fall on the 'downbeat'
of a foot, the effect is like that of musical syncopation. When the
accents line up with ictus perfectly, the effect is like that of a
musical march or something like that.

There's an infamous line of Ennius where the accents and the ictus line
up perfectly: 'sparsis longis hastis campus splendet et horret'.
Apparently, an awful lot of Ennius lined up like that. While if that's
the case for one line it sounds punchy and militaristic, you can well
imagine that if that were your rhythm for line after line you would
find the reading extremely boring indeed.

What you find in writers from Lucretius on is the enforced use of the
caesura to ensure that that doesn't happen. The effect, then, is that
accent and ictus do not line up in the first half of the line, but they
do in the second. So in your lines, 8-11:

Musa, mihi causas memora, quo numine laeso,

quidve dolens, regina deum tot volvere casus

insignem pietate virum, tot adire labores

impulerit. Tantaene animis caelestibus irae?

we can mark the beginning of each foot:

Musa, mi-|hi cau-|sas memo-|ra, quo |numine |laeso,
quidve do-|lens, re-|gina de-|um tot |volvere |casus
insig-|nem pie-|tate vir-|um, tot a-|dire la-|bores
impule-|rit. Tan-|taene ani-|mis cae-|lestibus |irae?

You can see that the first sylllable of each foot (the one coming right
after my marking with the | sign) there is an ictus emphasis. Now, if
we look at the last two feet of every line, we see that the ictus is
coming on the syllable of the word that bears the stress accent. Before
these last two feet, however, the only two cases where the accent and
ictus line up (other than the first syllables of each line) are in the
words 'regina' and 'tantaene' - this latter of course has had its
accent shifted because of the enclitic.

In general it is true that in the hexameter line from Lucretius on, the
ictus matches the accent for the fifth and sixth feet. In general, it
is therefore often metrically noticeable to have the ictus match the
accent in the fourth or the third and fourth as well. Lines which have
five or all six feet where the ictus and accent line up, or those where
they don't line up even in the fifth or sixth feet, are meant to be
noticed as metrical effects.

Neeraj Mathur

Ben C

unread,
Mar 28, 2006, 7:09:56 AM3/28/06
to
On 2006-03-28, Neeraj Mathur <neeraj....@gmail.com> wrote:
>
> Ben C wrote:
>> On 2006-03-19, Ken Quirici <ken.q...@excite.com> wrote:
>> > Also, I wonder at memora. In particular, the mem is accented and
>> > there's an instinctive tendency (didn't the Romans have it too?) to
>> > lengthen accented syllables. This is not really a scansion question;
>> > or maybe it is.
>>
>> By accented do you mean stressed?
>>
>> There is a tendency to stress long syllables in that the penultimate
>> syllable of a word is usually stressed if it scans long, otherwise it's
>> usually the third syllable from the end. This works much the same as in
>> English. But I don't know which came first, the stress or the long
>> vowel. I would have thought the long vowel, but that's a complete guess.
>>
>> People do talk, though, about the relationship between stress and ictus.
>>
>> In my quite simplistic understanding of it, this refers basically to how
>> often the natural stresses of the words coincide with the starts of
>> feet. It's much more subtle than that though.
>
> First things first: [...]

> [...]

> Now, it can also be seen that long vowels were not affected by these
> changes:

> 5) ce:do:
> 6) disce:do:
> 7) a:ctus
> 8) coa:ctus (note that this is not *coectus like perfectus above).

I didn't realize the a of actus was long-- it scans long anyway because
of the double consonant, but so does the a in factus, although the vowel
itself in factus is short.

> [...] In general it is true that in the hexameter line from Lucretius


> on, the ictus matches the accent for the fifth and sixth feet. In
> general, it is therefore often metrically noticeable to have the ictus
> match the accent in the fourth or the third and fourth as well. Lines
> which have five or all six feet where the ictus and accent line up, or
> those where they don't line up even in the fifth or sixth feet, are
> meant to be noticed as metrical effects.

Yes, and there are examples of such metrical effects in the Aeneid.
These are lines that stand out, almost by breaking the rules, but I feel
sure that the way stress and ictus work together is important in the
"normal" lines too.

> What you find in writers from Lucretius on is the enforced use of the

> caesura to ensure that [stress and ictus coinciding in first part of
> line] doesn't happen. [...]

The caesura I always understood to be a word-break typically after 2.5
feet in Vergil (although I know my understanding of this is very
incomplete). If this is a "strong caesura" (break after the first long
syllable, as opposed to between the two shorts of a dactyl), it would
guarantee that the third foot starts with the last syllable of a word,
and cannot be stressed, since the stress is never on the last syllable.
I had never thought of this relationship between caesura, stress and
ictus before. We see this strong caesura in the third foot only in the
first of these four lines:

Musa, mi-|hi cau-|sas memo-|ra, quo |numine |laeso,
quidve do-|lens, re-|gina de-|um tot |volvere |casus
insig-|nem pie-|tate vir-|um, tot a-|dire la-|bores
impule-|rit. Tan-|taene ani-|mis cae-|lestibus |ira

What are the rules for where the caesurae occur in Latin hexameter?

Many thanks for the detailed explanation you've given.

Ed Cryer

unread,
Mar 28, 2006, 8:15:07 AM3/28/06
to

"Ben C" <spam...@spam.eggs> wrote in message
news:slrne2i9ns....@bowser.marioworld...

Interesting comment on caesura. It's noticeable that the 7 lines preceding
these (ie. the famous opening 7) all have strong caesuras in the third foot.
But 9-11 have weak caesura splits in 2nd and 4th feet, after which Vergil
returns to strong again.

The effect is to slow things down as we move from epic declamation to a more
sombre questioning of what brought it all about. And then we're off on the
fast ride again.

Ed

Neeraj Mathur

unread,
Mar 28, 2006, 9:44:28 AM3/28/06
to

Ben C wrote:
> On 2006-03-28, Neeraj Mathur <neeraj....@gmail.com> wrote:
> > 7) a:ctus
> > 8) coa:ctus (note that this is not *coectus like perfectus above).
>
> I didn't realize the a of actus was long-- it scans long anyway because
> of the double consonant, but so does the a in factus, although the vowel
> itself in factus is short.

It is long; it's a classic problem about why it's long, something
called Lachmann's Law (which, as a Law, is pretty useless and is a
matter of controversy for Indo-European linguistics).

There are basically three main ways to tell if a Latin vowel was long
or short, which are scansion, behaviour in vowel weakening, and
developments in Romance (because long and short vowels developed
differently). Sometimes, information from one or more of these sources
is not useful - like with actus, the scansion doesn't help - but
between the three we can be relatively sure. There are also other
sources - epigraphic evidence, for instance, sometimes marked long
vowels - and these can add to the argument on either side when
available.

...


> > What you find in writers from Lucretius on is the enforced use of the
> > caesura to ensure that [stress and ictus coinciding in first part of
> > line] doesn't happen. [...]
>
> The caesura I always understood to be a word-break typically after 2.5
> feet in Vergil (although I know my understanding of this is very
> incomplete). If this is a "strong caesura" (break after the first long
> syllable, as opposed to between the two shorts of a dactyl), it would
> guarantee that the third foot starts with the last syllable of a word,
> and cannot be stressed, since the stress is never on the last syllable.
> I had never thought of this relationship between caesura, stress and
> ictus before. We see this strong caesura in the third foot only in the
> first of these four lines:
>
> Musa, mi-|hi cau-|sas memo-|ra, quo |numine |laeso,
> quidve do-|lens, re-|gina de-|um tot |volvere |casus
> insig-|nem pie-|tate vir-|um, tot a-|dire la-|bores
> impule-|rit. Tan-|taene ani-|mis cae-|lestibus |ira
>
> What are the rules for where the caesurae occur in Latin hexameter?

Well I'd say that the caesura in that line (line 8 of the poem) is
actually after 'memora', because 'causas memora' makes up a syntactic
unit, as does 'quo numine laeso' - it's one of these fourth-foot
caesuras that could technically be scanned either way but work better,
going with syntax and sense, in the fourth foot. (That also supports
Ed's point about these four lines being transitional and so marked
metrically.)

Now, rules for the caesura - my university is very big on finding the
caesuras when you scan a line, an exercise which I personally find
rather pointless. I don't for a second believe that fluid writers like
Virgil or Ovid were thinking about managing their caesuras - rather,
they would have just felt the pulse of the metre and the rhythm of
their words. Maybe I think like that just because I'm a musician and
have an instinct for rhythm; I don't know.

At any rate, you'll find that most caesuras in Latin dactylic hexameter
from Lucretius on are a strong caesura in the third foot:
Arma virumque cano || Troiae qui primus ab oris

and less frequently in the fourth foot:
insignem pietate virum || tot volvere casus.

You'll also find, still less frequently, a weak caesura in the third
foot, which occurs between the two short syllables in that foot (and is
the most common type of caesura in Homer, although Greek hexameters
should not be analysed that way):
cum sociis natoque || penatibus et magnis dis.

You'll also find something called a 'diaeresis' - the difference
between a diaeresis and a caesura is that diaeresis occurs at the end
of a foot, while caesura occurs in the middle. One type is called the
'Bucolic diaeresis' because it's supposed to be most common in pastoral
poetry - this is a diaeresis at the end of the fourth foot:
namque erit ille mihi semper deus, // illius aram (Ecl. 1.7)

and another:
Non equidem invideo, miror magis: // undique totis (Ecl. 1.11)

Note that in this latter line, you could mark a strong caesura in the
third foot (after 'invideo') in addition to the caesura. With the
earlier line, if you really wanted to, you could mark a caesura after
'mihi', but I think it would be wrong to do so - there should be no
pause at all in that line until after deus, which is where the
diaeresis is.

Note also that I don't mark a diaeresis in any of the lines we were
looking at earlier, any of the first 11 lines of the Aeneid. Even
though there are lots of places where there is a word-break at the end
of the fourth foot, they seem virtually always to be connected with the
word in front without any obvious need for a break.

I guess that's the main thing about caesuras and diaereses: they are
not just word divisions at any of the given points, they are actual
pauses. They exist because of a break in sense, and this interacts with
stress and ictus to create the rhythm of the line.

> Many thanks for the detailed explanation you've given.

No problem - glad you've found me useful!

Neeraj Mathur

Ben C

unread,
Mar 29, 2006, 11:58:43 AM3/29/06
to
On 2006-03-28, Neeraj Mathur <neeraj....@gmail.com> wrote:

> [..]


> Now, rules for the caesura - my university is very big on finding the
> caesuras when you scan a line, an exercise which I personally find
> rather pointless. I don't for a second believe that fluid writers like
> Virgil or Ovid were thinking about managing their caesuras - rather,
> they would have just felt the pulse of the metre and the rhythm of
> their words. Maybe I think like that just because I'm a musician and
> have an instinct for rhythm; I don't know.

I think that's the right way to think about it, anyone who doesn't see
that poetry is music hasn't read Vergil or Ovid.

This is roughly what I was trying to tell the OP in the first place.

But this isn't "free verse", and it is interesting to know the broad
outlines of what the rules are (which you have explained).

Ed Cryer

unread,
Mar 29, 2006, 12:55:36 PM3/29/06
to

"Ben C" <spam...@spam.eggs> wrote in message
news:slrne2lf1b....@bowser.marioworld...

I go with this music parallel. After all a hexameter consists of six feet.
And a foot consists of arsis and thesis, raise and place. Foot up, foot
down.
Just try it with your right foot, beating time. Keep your heel on the floor
and raise the foot, then down. Equal time for up/down = a spondee.
But then you can tap the floor twice on the down beat; = a dactyl.

A practised wordsmith in Latin would have internalised all this so well that
in actual composition he would be conscious not of the smaller elements, but
of the overall flow across several lines; maybe even further. Just as say,
Johann Strauss the waltz king wouldn't have to keep repeating to himself
123, 123 in order to stay within 3/4 time.

Ed

Simon Pugh

unread,
Mar 29, 2006, 1:09:09 PM3/29/06
to
As a Vergil beginner, I have very much enjoyed this thread and found it
useful. One of the things that someone advised a while ago was learning
passages.

I have tried learning the opening lines of books 1 and IV and I must say
it makes a tremendous difference to be able so recite the lines out loud
from memory. The metre and caesuras seem to come naturally and I think
it helps to understand the Latin as Latin rather than as a puzzle to be
solved. I have always had a problem with remembering what as gone before
in a long sentences, but this isn't an issue if it has been memorised.
--
Simon Pugh
Remove X for mail

Ken Quirici

unread,
Mar 29, 2006, 1:12:20 PM3/29/06
to
Ed Cryer wrote:
> "Ben C" <spam...@spam.eggs> wrote in message
> news:slrne2lf1b....@bowser.marioworld...
> > On 2006-03-28, Neeraj Mathur <neeraj....@gmail.com> wrote:
> >
> >> [..]
> >> Now, rules for the caesura - my university is very big on finding the
> >> caesuras when you scan a line, an exercise which I personally find
> >> rather pointless. I don't for a second believe that fluid writers like
> >> Virgil or Ovid were thinking about managing their caesuras - rather,
> >> they would have just felt the pulse of the metre and the rhythm of
> >> their words. Maybe I think like that just because I'm a musician and
> >> have an instinct for rhythm; I don't know.
> >
> > I think that's the right way to think about it, anyone who doesn't see
> > that poetry is music hasn't read Vergil or Ovid.
> >
> > This is roughly what I was trying to tell the OP in the first place.
> >
> > But this isn't "free verse", and it is interesting to know the broad
> > outlines of what the rules are (which you have explained).
>
> I go with this music parallel. After all a hexameter consists of six feet.
> And a foot consists of arsis and thesis, raise and place. Foot up, foot
> down.
> Just try it with your right foot, beating time. Keep your heel on the floor
> and raise the foot, then down. Equal time for up/down = a spondee.
> But then you can tap the floor twice on the down beat; = a dactyl.
>

Actually I seem to do the reverse as my natural 'keeping time' foot
taps. A dactyl to me is down up up (of the foot - where down means
hit the floor). An anapest would of course be up up down.

If Neeraj meant downbeat like this, that is, the accented beat,
then we're in agreement.

> A practised wordsmith in Latin would have internalised all this so well that
> in actual composition he would be conscious not of the smaller elements, but
> of the overall flow across several lines; maybe even further. Just as say,
> Johann Strauss the waltz king wouldn't have to keep repeating to himself
> 123, 123 in order to stay within 3/4 time.
>

My only issue here, which I apologize for taking up time on, is
that the natural rhythm of a waltz - dactyl - and with no ictuses
(icti?) and no length considerations (except slow waltz fast
waltz) - is much easier to flow naturally into. But of course Latin
poetry is simply more complicated, which does not mean it's
impervious to a natural musical rhythm. There are extremely
complicated (I am told) rhythms in much tribal music in Africa
(among other places).

Also Latin poetry lacks the rhythm section :)

Thanks.

Ken

> Ed

Ken Quirici

unread,
Mar 29, 2006, 1:29:13 PM3/29/06
to

What is the relative strength of the ictus and the accent? My
impression is that the ictus is slightly less prominent. That it's
sort of like the way English iambic verse tends to become
slightly artificially iambic once the real iambs 'set the rhythm' -
kind of sing-songy.

When reading classical Latin prose aloud, is it true that
there's no ictus?

I thought the 'a' in pietate was long. In your scheme it makes it
also line up the stress and ictus. And make 3 instances of such.
This would also emphasize metrically the notion that the wrath
of the goddess fell on a famously 'virtuous' man.

> In general it is true that in the hexameter line from Lucretius on, the
> ictus matches the accent for the fifth and sixth feet. In general, it
> is therefore often metrically noticeable to have the ictus match the
> accent in the fourth or the third and fourth as well. Lines which have
> five or all six feet where the ictus and accent line up, or those where
> they don't line up even in the fifth or sixth feet, are meant to be
> noticed as metrical effects.
>
> Neeraj Mathur

Fascinating stuff. Thanks.

Ken

John W. Kennedy

unread,
Mar 29, 2006, 1:37:38 PM3/29/06
to
Ken Quirici wrote:
> Actually I seem to do the reverse as my natural 'keeping time' foot
> taps. A dactyl to me is down up up (of the foot - where down means
> hit the floor). An anapest would of course be up up down.

This is alt.language.latin, not alt.language.english. A Latin (or Greek)
dactyl is not BOOM-chink-chink but daahh-da-da. A few English examples
of dactyls such as Latin verse employs might be:

kingmaker
blonde Suzie
raw footage
gross liar

Ken Quirici

unread,
Mar 29, 2006, 1:55:41 PM3/29/06
to

John W. Kennedy wrote:
> Ken Quirici wrote:
> > Actually I seem to do the reverse as my natural 'keeping time' foot
> > taps. A dactyl to me is down up up (of the foot - where down means
> > hit the floor). An anapest would of course be up up down.
>
> This is alt.language.latin, not alt.language.english. A Latin (or Greek)
> dactyl is not BOOM-chink-chink but daahh-da-da. A few English examples
> of dactyls such as Latin verse employs might be:
>
> kingmaker
> blonde Suzie
> raw footage
> gross liar
>

Especially good examples because the second syllable (except
to my ears 'foot') is a 'long' vowel. Actually also I tend to
lengthen the first syllable of Suzie. But I get your point.
It's also Ed's 'strawberry'.

With all these examples I wonder whether the long
cherished hope of some English-language versifiers
for a true length-based metric might not be more
possible than at least I had thought. Actually you
yourself in an earlier post showed some examples,
if I remember correctly.

Also allow me to point out that BOOM-chink-chink is in
(again to my ears) Latin long - latin short - latin short :-)
i.e., another example of a 'Latin' dactyl.

In any event what I was talking about was the accent/ictus
in Latin verse. Unfortunately I used a word used in Latin
scansion that is not ABOUT accent/ictus (i.e. the word
dactyl) - thanks for pointing that out.

Thanks.

Ken

Ed Cryer

unread,
Mar 29, 2006, 4:21:49 PM3/29/06
to

"Ken Quirici" <kqui...@yahoo.com> wrote in message
news:1143655940.9...@z34g2000cwc.googlegroups.com...

>>
>
> Actually I seem to do the reverse as my natural 'keeping time' foot
> taps. A dactyl to me is down up up (of the foot - where down means
> hit the floor). An anapest would of course be up up down.
>

I've been doing some experiments on this with The Beautiful Blue Danube
playing.
I started off with up-down-down. Then I realized that I'd kind of programmed
myself into that by talking it, so I set about just "letting it go" as they
say. I waltzed around the room (nobody else present) and then sat down again
and let it happen.
And I found that I too went for the down-up-up; ie down on the "arsis". The
more I did it, the more it became natural to emphasise the first beat with a
down.
After not too long I ended up with what seemed to be an even more natural
growth. Down beats only. Both feet doing down beats, the right three to a
bar, the left just one. Ie. it was emphasising the first beat.

Ed

Ben C

unread,
Mar 29, 2006, 4:32:54 PM3/29/06
to
On 2006-03-29, Ken Quirici <kqui...@yahoo.com> wrote:

> [...] the natural rhythm of a waltz - dactyl - and with no ictuses
> (icti?)

It would be ictus (with a long u). I think ictuses is fine :)

Ben C

unread,
Mar 29, 2006, 4:34:44 PM3/29/06
to
On 2006-03-29, Ed Cryer <e...@somewhere.in.the.uk> wrote:
>
> "Ken Quirici" <kqui...@yahoo.com> wrote in message
> news:1143655940.9...@z34g2000cwc.googlegroups.com...
>>>
>>
>> Actually I seem to do the reverse as my natural 'keeping time' foot
>> taps. A dactyl to me is down up up (of the foot - where down means
>> hit the floor). An anapest would of course be up up down.

> I've been doing some experiments on this with The Beautiful Blue

> Danube playing. [...]

Now repeat the exercise with Beethoven Symphony No 7, second movement.

John W. Kennedy

unread,
Mar 29, 2006, 9:51:05 PM3/29/06
to
Ken Quirici wrote:
> John W. Kennedy wrote:
>> Ken Quirici wrote:
>>> Actually I seem to do the reverse as my natural 'keeping time' foot
>>> taps. A dactyl to me is down up up (of the foot - where down means
>>> hit the floor). An anapest would of course be up up down.
>> This is alt.language.latin, not alt.language.english. A Latin (or Greek)
>> dactyl is not BOOM-chink-chink but daahh-da-da. A few English examples
>> of dactyls such as Latin verse employs might be:
>>
>> kingmaker
>> blonde Suzie
>> raw footage
>> gross liar
>>
>
> Especially good examples because the second syllable (except
> to my ears 'foot') is a 'long' vowel. Actually also I tend to
> lengthen the first syllable of Suzie. But I get your point.
> It's also Ed's 'strawberry'.
>
> With all these examples I wonder whether the long
> cherished hope of some English-language versifiers
> for a true length-based metric might not be more
> possible than at least I had thought.

It is /possible/. Bridges did it. Tennyson did it, though only in
mockery. I imagine any English-speaking Latin student has done it in the
privacy of his own home. (I've done it myself.) But it's simply not
native to English, or to any other Germanic language. (Neither, for that
matter, are the purely syllabic meters of French or Japanese.)

Other languages can do better, e.g., Boito's

Notte cupa, truce, senza fine funèbre!
Orrida notte d'Illio! implacato rimorso!
Nugoli d'arsa polvere al vento surgono e fanno
più cieca la tenèbra. Di cozzantisi scudi,
di carri scroscianti, di catapulte sonanti
l'etere è scossa! si muta il suol in volutàbro
di sangue. I numi terribili ruggono, l'ire
inferocendo della pugna; l'ispide torri
ergonsi tragiche, negre, fra la caligine densa.
L'incendio già lambe le case. Veggonsi l'ombre
degli Achèi proiette (bui profili giganti)
vagolar le pareti al lume torvo de' roghi.
Ahimè! tremano basi e vertici! Crollano mura!
Si diroccano torri e tuona e sfolgora l'orbe!
[pausa]
Alto silenzio regna poscia dove fu Troia.

Of course, Boito had a rare genius, albeit an unreliable one.

> Actually you
> yourself in an earlier post showed some examples,
> if I remember correctly.

Yes, examples both of quantity used as a strong, but secondary resource
in standard English verse and of direct English use of quantitative
hexameters and elegiacs. (On the whole, I have always found elegiacs to
be the most successful quantitative form in English.)

> Also allow me to point out that BOOM-chink-chink is in
> (again to my ears) Latin long - latin short - latin short :-)
> i.e., another example of a 'Latin' dactyl.

But you went on to describe it as a "waltz", which it most certainly is not.

Ed Cryer

unread,
Mar 30, 2006, 5:41:28 AM3/30/06
to

"Ben C" <spam...@spam.eggs> wrote in message
news:slrne2lv6r....@bowser.marioworld...

You'll be pleased to hear that I have done just that.
Wagner called Beethoven's 7th the very apotheosis of dance, and the second
movement is in simple 2/4 time.
I found that I was swaying my body rather than stepping. The first movement
definitely calls for stepping about, but the second made me sway about. And
when I sat down and tried to adjust a foot tap to it, I found that I could
still use both feet but the left did one tap to the right's two. I also
found that with the left I could tap interchangeably with heel or toe.

We'll have to try some highly syncopated jazz rhythms. I've got a feeling
I'm getting enough practice in here to be able to perform to some Miles
Davis. {:-

Ed

P.S. Don't ask about The Rite of Spring or the scherzo from Dvorak's 9th.
I've done both of those with admirable success.

Ken Quirici

unread,
Mar 30, 2006, 12:15:12 PM3/30/06
to

With tap shoes on a hardwood floor :)

Actually if there are any 5-beat metra you could maybe try 'Take Five'
(Brubeck?).

Thanks.

Ken

0 new messages