Orrajt (Eng Title: Alright) is an “Award Winning Maltese Short Film – 2021” about “The uphill bilingual battle George needs to face as being both an English and Maltese speaker.” The Maltese is not subtitled, which is fun and shouldn’t cause a problem — the English dialogue makes it clear what’s going on. It’s only nine minutes, and it’s a sympathetic look at a problem a lot of people have to deal with in this increasingly multilingual world. (It was shot in the Beggars Inn Pub, which looks a little rowdy for the likes of me but is certainly photogenic.) Via Slavo/bulbul at FaceBook.
And if that doesn’t interest you, I present HamBam: The Hamedan-Bamberg Corpus of Contemporary Spoken Persian:
The HamBam corpus contains annotated recordings of contemporary spoken Persian, compiled as part of a cooperation between Bu-Ali Sina University in Hamedan, Iran (team coordinator: Mohammad Rasekh-Mahand), and the University of Bamberg in Germany (Geoffrey Haig). […]
The corpus was primarily designed for investigations into the impact of information structure on word order variation in spoken Persian, but is freely adaptable to other research questions, including research on prosody, referential density, or usage-based approaches to grammar.
Clicking YouTube’s [cc] button should turn on English-language captions for both the English bits and the Maltese bits!
Ha, I didn’t even think of that!
The film was wonderful! Many thanks!
I like to collect micro-libraries for languages I might want to become literate in. Maltese is on the long list, and last year I managed to find several recent novels in Maltese, some with versions also in English, plus a grammar and dictionary. I got some valuable help from one of the leading publishers in Malta (Merlin https://merlinpublishers.com/product/fittixni/ ), which produces beautiful books. I’m just mentioning this because if you want to learn Maltese, the material is out there, though it takes some work to find it. I have not started working on the language yet, but maybe one day…
See also
https://www.midseabooks.com/shop/language-literature/concise-maltese-english-maltese-dictionary/
https://bdlbooks.com/author-2/lawrence-gatt/
The short was really enjoyable and cute. Thanks.
The corpus was primarily designed for investigations into the impact of information structure on word order variation in spoken Persian
What the heck is “information structure” ? It seems to have been a Thing in linguistics for a while, but the WiPe article left me clueless. It starts thus:
# In linguistics, information structure, also called information packaging, describes the way in which information is formally packaged within a sentence.[1] This generally includes only those aspects of information that “respond to the temporary state of the addressee’s mind”, and excludes other aspects of linguistic information such as references to background (encyclopedic/common) knowledge, choice of style, politeness, and so forth.[2] For example, the difference between an active clause (e.g., the police want him) and a corresponding passive (e.g., he is wanted by police) is a syntactic difference, but one motivated by information structuring considerations. Other structures motivated by information structure include preposing (e.g., that one I don’t like) and inversion (e.g., “the end”, said the man).[3]
The basic notions of information structure are focus, givenness, and topic,[2] as well as their complementary notions of background, newness, and comment respectively.[4] Focus “indicates the presence of alternatives that are relevant for the interpretation of linguistic expressions”, givenness indicates that “the denotation of an expression is present” in the immediate context of the utterance, and topic is “the entity that a speaker identifies, about which then information, the comment, is given”.[2] Additional notions in information structure may include contrast and exhaustivity, but there is no general agreement in the linguistic literature about extensions of the basic three notions.[4] There are many different approaches, such as generative or functional architectures, to information structure.[5] #
Looks like a bunch of people got tired of the traditional jargon based on Latin syntax, such as subject/object/verb, so they fashioned a new jargon to go on top. The word “indicates” does a lot of heavy lifting there. Part of the information structure of information structure seems to be avoidance of the word “means”.
1. Focus “indicates the presence of alternatives that are relevant for the interpretation of linguistic expressions”
The focus is the choice among alternatives to talk about. It’s the subject of the talk until the subject/focus changes. What’s the point of a new word here ? Elegant variation ?
2. givenness indicates that “the denotation of an expression is present” in the immediate context of the utterance
Does this mean that if what is being talked about is in the same room as the talkers, then it’s in the same room ?
3. topic is “the entity that a speaker identifies, about which then information, the comment, is given”
So topic = what is being talked about, and the comment is the talk ?
The WiPe passage I quoted has a topic, but no focus or givenness. We call that Geschwätz.
I finally watched it.
Have they imported the entire English social stratification, complete with sociolects, or what. *shudder*
Subject/object/verb don’t always tell you what to pay attention to in a sentence, but a human reading a sequence of sentences has no problem. It looks like “intormation structure” is a way of talking about the extra layer of meaning / interpretation that exists.
That said, I can never get focus and topic straight and the stuff in the quote didn’t help. I think I’ve read that Japanese has separate particles for the two, so maybe I should learn that next.
@Stu:
Information packaging is a real thing, and not some modish fad of bygone linguists. CGEL has a whole chapter on it (86 pages of grammatical goodness …) Even my Kusaal grammar manages twelve.
It’s a hard subject for a lot of reasons. Different authors mean subtly different things by “focus”, for a start, and it doesn’t work the same way across languages; some nihilists even claim that there is not really any such thing as “focus” in the abstract. Moreover, the “alternative” kind of focus is not the only sort. The prototypical kind of focus is that given to a clause element in response to a content question:
“Who did that?”
“John did that.”
“What did John do?”
“John did that.”
Focus may involve contrast, but it doesn’t necessarily do so. It is also quite distinct from “emphasis” (a term so vague that it usually signifies that the grammar-writer has given up trying to explain the phenomena.)
Information packaging is also hard to study because it deals with how sentences with basically the “same” meaning vary grammatically depending on what is at the forefront of the speaker’s mind, what they expect you to already know etc etc. Trying to discover focus phenomena from texts can be a nightmare, and in fieldwork you continually run into Heisenbugs – if you ask someone to repeat a sentence, they say it differently, exactly because you asked them to repeat it. It’s even less use than usual trying to get the speaker to “explain” why they put it that way.
In English (and German), focus is typically marked by intonation, which doesn’t help. Even quite sophisticated native speakers react with incredulity and amazement when you try to explain to them that there are actual rules for English intonation …
“Givenness”, among other things, comes into the use of the articles in languages that have such things. Other languages do it differently (we had a discussion here about how Finnish uses different cases for direct objects partly reflecting givenness.)
“Topic” and “subject” often coincide in English and SAE languages in general, but they are not the same thing fundamentally. Japanese, for your convenience, neatly marks them with different particles (ga subject, wa topic.)
a human reading a sequence of sentences has no problem.
Not so ! That was the point of my comment. In that to-me-unintelligible passage, the subject/object/verb bits told me absolutely nothing about the meaning. Trying to use them in the standard way, to tease out the meaning, I found that the words were juxtaposed in a way worthy of a beginning student of ESL. Every formulation is ever so slightly wrong-footed.
For similar reasons I find the works of Talcott Parsons to be almost unintelligible. He writes English like an ESL German professor. It was no surprise when I read that he “studied” a few years in Germany. He was poisoned in his prime by the dog biscuit of Weber’s rhetoric.
Instead of “topic” vs. “focus”, I first came across “topic” vs. “comment”. With that classification it’s at least much easier to remember which is which – but I don’t know if it’s supposed to be exactly the same as “topic” vs. “focus”…
Chinese has topic-verb-comment word order.
Comment-topic-verb word order Master Yoda uses. Not object-subject-verb it is, because clearly discernible it remains even when no object there is.
Chinese has topic-verb-comment word order.
It’s more complicated than that: the comment can itself have a subject and predicate.
In his Grammar of Spoken Chinese, YR Chao has, for example (p95)
这个人耳朵软
Jeyg ren eel.dou roan (in Chao’s wonderful Gwoyeu Romatzyh)
“(As for) this man, the ear is soft.” (i.e. This man is gullible.)
David Marjanović: I’m almost certain I first came upon topic-focus rather than topic-comment, and I think it was that it was in the context of an Austronesian language, probably from the Philipines but maybe from geographically adjacent areas to the west. It’s been a long time. “focus” had a very technical meaning in the description. The grammar I remember reading referred to it having five slots, with topic and focus being obligatory.
…a topic and a comment, I would say.
Anyway, beautiful example of a verbless sentence.
beautiful example of a verbless sentence
Not in Chao’s analysis: the adjective is basically behaving as a verb (though I believe that in Mandarin there are some complications in the way of just regarding predicative adjectives as verbs tout court. Still, they’re pretty verbal.) In the relevant passage Chao actually calls topics “subjects”, though as he writes with his characteristic clarity it’s apparent that he means what we Young People call “topics.” I doubt whether he would have objected to your analysis of the sentence as having two topics, anyhow. The significant thing is that the first topic can’t be taken as the subject/topic of the adjective; he illustrates this with a similar example which literally goes “He (is) face-skin thick” (i.e. shameless), where the adjective glossed as “thick” cannot apply to a person.
The ultimate language for lovers of grammatical focus (and aren’t we all, deep down, lovers of grammatical focus?) is surely Lavukaleve, in which 35% of all sentences in Angela Terrill’s corpus have explicit focus marking. There are three focus markers, all independent words inflected for person, gender (masculine/feminine/neuter) and number (singular/dual/plural.)
I had to do a lot of reading up on focus to make sense of the Kusaal particle nɛ, which marks focus within verb phrases and also interacts with the aspect system in a way which ought to cause a lot of ambiguity but in practice doesn’t (much.) It’s a difficult area to get to grips with … both theoretically and practically.
From a cross-linguistic point of view, at any rate, although comments are likely to be focused by default (just as VPs are focused by default in English), it ain’t necessarily so; consequently, I think it’s a bad idea to talk about “topic-focus” structure rather than “topic-comment.”
“Topic” and “subject” often coincide in English and SAE languages in general, but they are not the same thing fundamentally. Japanese, for your convenience, neatly marks them with different particles (ga subject, wa topic.)
Three’s also omission of topic, eg:
[…] the example “I am Misa” actually sounds better without わたしは. I am Misa. (in a true native way) = みさです。 Misa desu Of course, when it’s not obvious who you are talking about, then put it.
The Ultimate Guide To: は vs が (The ONLY lesson you need!)
Speaking of Japanese, I wonder where have all the ŋ’s have gone.
Speaking of Japanese, I wonder where have all the ŋ’s have gone.
Come again?
Are you speaking of [ga] vs [ŋa]?
When I started learning Japanese in the late 1980s, all audio materials had ŋ’s for all non-initial g’s. These days, when I listen to, eg, NHK podcasts, I almost never hear them.
@juha, see WP on that. It’s a known variation between groups of speakers, but why your audio materials were group B and NHK are C (or A), I won’t venture to guess. There are some references there that suggest either sociolectal or age/geographic variation.
This article looks at the issue in hand.
DeepLed:
(‘nasal sound’ refers to the ŋ.)
Will the nasal sound disappear?
Some of the questions we receive about the language of broadcasting include orders and opinions about NHK from outside parties.
Issues range from announcer language to advice on attire, makeup, and more.
One person from the Tohoku region sent us a tape with the following comment: “One of the announcers can’t pronounce nasal vowels. Fire him immediately.” I want you to fire him immediately.
If you can recognize the “nasal sound” immediately, you must be familiar with spoken language. The Japanese word “gago” actually refers to two sounds. I like movies. is pronounced “watashi wa ega gaskidesu” and “ha” is read as “wa.” In the same way, “ga” is written “ga” and has two sounds: a voiced “ga” sound and a nasal “nga” sound.
Roughly speaking, “ga” at the beginning of a word is a voiced sound, but “ga” in the middle of a word or “ga” in a particle is a nasal sound.
The complainant was a resident of the Tohoku region, but in the Chugoku and Kyushu regions, it is not customary to pronounce “ga” with this nasal sound, and it is not a problem in those regions.
However, since a common language is used in broadcasting, it can be said that nasal sound pronunciation is indispensable.
Recently, however, nasal sounds have been disappearing in eastern Japan, especially among young people. Enka (traditional Japanese folk songs) use beautiful nasal sounds, but since the beginning of what is called “new music,” the use of voiced sounds has been spreading even in the world of song. In “Tsugaru Kaikyo Fuyugeshiki” (Tsugaru Kaikyo Fuyugeshiki), “ga, ge” is a nasal sound.
Yumi Arai’s “Sotsugyo Shashin” (Graduation Photo) is sung with the nasal sound “gyo” in “sotsugyo shashin.
One reason for the increasing number of young people who are unable to produce nasal sounds is that there is no way to write the distinction in general notation. (Technically, the nasal sound is written with a semivoiced dot as “ga” for voiced sounds and “ka˚” for nasal sounds).
Is it rationalism to assume that “if there is no difference in appearance, then the pronunciation is the same?”
This trend is also seen among new announcers, and the percentage of those who can nasalize when they first join the station has been decreasing year by year.
In addition to nasalization, newcomers are given thorough training in specialized pronunciation such as voicelessness, but the only method of instruction was one-on-one teaching. Recent developments in personal computers have made it possible to easily analyze speech sounds, which previously could only be done with special machines. The terminology group has researched analysis methods that can be used for training, and has made it possible to visually distinguish the difference between voiced and nasal sounds to some extent objectively.
Details will be reported in the November issue of “Broadcasting Research and Survey. We hope that this method will help new announcers to train in their assigned areas.
Wow, thanks to juha and Lars for the [ŋ]/[ɡ] information — like juha, when I learned Japanese it was (as far as I remember, but this is certainly what I internalized) always [ŋ] intervocalically, and it’s fascinating to learn of the variation.
I just remembered coming across it before in that WP article, so when juha mentioned it it was easy to find. I have yet to learn enough Japanese to have to know that stuff.
But, trying to synch up the DeepL version with the Japanese, it looks like it translates 「が」 as “ga” and 「ガ」 as “nga”. How does that work?
Later it says that “technically” “nga” should be written as 「カ゜」 (“with the semi-voiced dot”) which DeepL cannot grok (“ka˚”) and which my browser treats as two double-wide chars in the default font. (But not when displaying the NHK article. I think it is a single-wide in that font [that does not combine with カ], but the circle is at the left in the space so it looks combining).
ŋ
ng
か゚ カ゚
き゚ キ゚
く゚ ク゚
け゚ ケ゚
こ゚ コ゚
https://en.wikipedia.org/wiki/Kana#Diacritics
Aren’t there also supposed to be accents with a prenasalized [ŋg]?
My “analysis” (i.e. immediate impression) is that there’s one topic and one comment – but the comment itself consists of a topic and a comment. It’s nested like a compound noun.
It’s nested like a compound noun.
Quite so; the interesting thing it that if it were a compound, the second component of the top-level compound would be a bahuvrihi. It’s an exocentric proposition.
Reminds me of the famous Japanese restaurant order: 私はタコです “I am an octopus.”
私はタコです “I am an octopus.”
“Boku wa unagi da” no bunpō (The grammar of ‘I’m an eel’)
Japanese title 「ボクハ ウナギダ」の文法
Author Keiichirō Okutsu
Publisher Kuroshio
https://www.sljfaq.org/books/book-387.html
Unagi-sentences in Japanese and mutual knowledge
https://www.academia.edu/45444102/Unagi_sentences_in_Japanese_and_mutual_knowledge
[ŋg]
Well, the consonant denoted by ん・ン is realized by different allophones depending on what follows.
The pronunciation can also change depending on what sounds surround it. These are a few of the ways it can change:
[n] (before n, t, d, r, ts, z, ch and j )
[m] (before m, p and b )
[ŋ] (before k and g)
[ɴ] (at the end of utterances)
[ũ͍] (before vowels, palatal approximants (y), consonants h, f, s, sh and w)
[ĩ] (after the vowel i if another vowel, palatal approximant or consonant f, s, sh, h or w follows.)
https://en.wikipedia.org/wiki/N_(kana)
“I am an octopus.”
German has something similar, mostly used by waiters and similar service personnel responding to orders. A typical question from a waiter serving the orders is Wer war das Steak? “Who ordered the steak?” (lit. “Who was the steak?”), and having got a response, they may then deduce Dann sind / waren Sie der Salat. “In that case, you are / were the Salad.”
Hans, that’s used in English too when the server brings the food or drinks. “I’m the steak” would be understandable in that context.
I think in the German/English cases the mechanism is rather different. In “I’m the steak”, I think that “the steak” is just elliptical* for “the one who ordered the steak”, whereas in the Japanese there’s no ellipsis: the sentence is just fine as it is, as topic-comment (it’s also perfectly grammatical without the topic.) On the other hand, if you substituted the subject marker ga for the topic marker wa you probably would get some odd looks …
* Perhaps “elliptical” is not the right term. It’s a waiter’s term of art. But the Japanese sentence still doesn’t rely on this being borrowed from Waitronics; it’s a perfectly cromulent way of putting it in Japanese. As the paper juha linked to points out, it needs to be set up properly, and would make no sense used in that way out of the blue; but this interaction of grammatical form and larger discourse context is what this whole thing is about.
Common or garden metonymy.
Indeed. That’s the word … alors, the Japanese sentence involves no metonymy …
Maybe the Japanese parallels English “He is chair of the meeting” vs. *”He is a chair of the meeting”, or ?“…the chair…”?
No, I don’t think so. It’s an unnecessary supposition. Although the unagi-ordering has given its name to the phenomenon (I remembered the wrong seafood), it’s all a perfectly normal consequence of how wa and topics work in Japanese, and not limited to ordering in restaurants. The sentence Boku wa unagi da does not have an expressed subject at all, neither boku “I”, nor anything else, so there is no need to do any mental gymnastics with the eel.
SAE-speakers have some difficulty actually believing that a topic need not necessarily be a subject, because SAE languages (and many others) characteristically conflate the two categories. The paper juha linked to, before describing the surely correct analysis, goes through quite a gallery of Awful Warnings of what happens when Chomskyans tried to force the topic-comment structures of Japanese into a Form They Were Comfortable With. (I think they’ve got more sophisticated nowadays; Japanese is one of their go-to languages for “proving” that they aren’t really obsessed with trying to pretend that specifically English syntactic quirks belong to their imaginary Universal Grammar. Nobody is fooled …)
You could paraphrase Boku wa unagi da into “As for me, it is the eel”, which, while it is not really English, gives a better idea of the structure of the Japanese. The “as for me” is misleading, however, because it falsely suggests a marked character to what is quite a neutral construction. juha will Actually Know, but I would guess that Japanese in the wild has a lot more expressed topics than expressed subjects.
I left Japan in 1993 and I seem to remember that [ga] was already becoming common then.
By the way, the two songs mentioned in the article are from around the mid-70s.
That doesn’t work with German masculines, because you are the nominative but you ordered the accusative.
There is a mistake in the translation, actually. Tsugaru-kaikyō Fuyu-geshiki* is an enka (traditional ballad-style song), Sotsugyō Shashin** is so-called “New Music”. In Tsugaru-kaikyō Fuyu-geshiki the nasal sound is used. In Sotsugyō Shashin the unnasalised sound is used.
* Winter scenery in the Tsugaru Strait (between Honshu and Hokkaido, which was once serviced by ferry, now replaced by a train tunnel. The song references the ferry).
** Graduation photo.
That doesn’t work with German masculines
Yes, Y was right to label this metonymy, and I was wrong to suppose it was ellipsis.
Ok, then the issue is with the rendering of the Japanese sentence as I am an octopus; if I understand correctly, a better approximation would be “I/Me? Octopus.”, using intonation to express the roles of the Japanese clitics.
@juha, I understand that か゚ and カ゚ are the conventional hiragana and katakana expressions of /ŋa/. But as far as I can identify the first place where the NHK article contrasts [ga] and [ŋa], they use hiragana が for [ga] and and katakana ガ for [ŋa] — at least that’s what DeepL translates those tokens to, and it makes sense in context. But I didn’t think the “same syllable” in hiragana and katakana could carry different meanings like that?
EDIT: I figured it out — I was out of sync, and they actually spell “nga” as katakana ﾝガ which makes much more sense. Though why they can use hiragana for “ga” and have to use katakana for that I don’t know — wouldn’t んが work? I am beginning to suspect that DeepL didn’t “know” that “nga” meant [ŋa] — and maybe it didn’t, maybe it means [ŋga] as I think somebody else mentioned.
(And then later in the piece DeepL gets confused by the “technically correct” か゚ and renders it “ka ̊ ” — but that’s not a question about Japanese except insofar as it points to か゚ being too rare in the training input, and presumably in Japanese text in general, to be handled properly).
I can never get focus and topic straight and the stuff in the quote didn’t help
Short, simple, and imperfect version: the focus is what is foregrounded (and they alliterate), whereas the topic is what is backgrounded (and they don’t). Protip: if anyone comes at you claiming that Austronesian syntax has topics or focuses, run away at once: they are either a Chomskyite, terminally confused, or in many cases both.
though I believe that in Mandarin there are some complications in the way of just regarding predicative adjectives as verbs tout court
Much ado about adjectives, mostly from 2009.
Could this be a use of katakana for emphasis?
Ok, then the issue is with the rendering of the Japanese sentence as I am an octopus
It can mean that. It’s just that it doesn’t have to.
the focus is what is foregrounded
Often; but foregrounding and focus are distinct in principle. The account of all this in CGEL is about as clear as possible, given the slippery and confusing nature of the actual facts themselves. In English, focus is marked by intonation, and foregrounding by word-order things like clefting. They may or may not coincide.
Kusaal has two distinct focus particles, but like English achieves foregrounding by playing with the word order, though Kusaal needs more formal machinery to do that than English does (it doesn’t even do wh-movement: wh-words often are fronted, but the aforesaid special preposing measures are needed if you decide to put them first – it can’t just happen.) Again, foregrounding and focus can occur together, but they don’t have to.
A peculiarity of Kusaal is that wh-words must be focused if they are subjects, but that otherwise focus is not possible in content questions.
Ok, then the issue is with the rendering of the Japanese sentence as I am an octopus
Sōseki’s famous novel 吾輩は猫である wagahai wa neko de aru “I am a Cat” illustrates that topic-comment is also the usual way of saying that you are, in fact, an animal …
One thing that makes focus and foregrounding hard [NB look, foregrounding!] is that you have to consider not just the sentence itself, but other possible sentences it is distinguishing itself from.
At least, that’s how I think about it right now.
Thanks for the pointer to CGEL. It really does do a good job exposing a complex subject.
Though why they can use hiragana for “ga” and have to use katakana for that I don’t know — wouldn’t んが work?
I think it works just as a transcription device.
Ok, then the issue is with the rendering of the Japanese sentence as I am an octopus
It can mean that. It’s just that it doesn’t have to.
I understand. My point is that saying “Japanese say I am an octopus when ordering octopus” is misleading; it’s rather that they use a construction that also can mean that (and maybe some things more), but doesn’t contain an actual copula.
Ah, but it does contain an actual copula: that’s what da/desu/de aru is.
What it doesn’t contain is an explicit subject.
it does contain an actual copula
Which is not strictly essential: see the unagi paper, pp.730–1.
Okay, I’ll stop commenting on Japanese eel and octopus ordering before I dig myself into a Japanese hole – they seem to be very difficult to escape from.
Great movie …
I have only heard of a topic – comment word order in Chinese. A topic or a comment can be made up from a verb, or a phrase with a verb in it, of course. Grammar books typically list examples with all sorts of word classes. I personally like the short and easy sentences where both are nouns. The type of sentence David Eddyshaw mentions, with a comment made of a topic – comment phrase, is one of the things that confuse me about Chinese grammar. I’m happy about the reminder to read up on that. As it turns out, I left a bookmark in one of my grammar books in exactly the spot where it’s explained… I guess I have had the ambition to learn about this for a long time. Oh, well.
By the way, the mention of predicates make me nostalgic, it seems like this term has fallen out of favour.
the mention of predicates make me nostalgic, it seems like this term has fallen out of favour.
This is the sort of thing that has driven Haspelmath to try to put some order in linguistic terminology. I sometime disagree with the way he does it, but certainly not with the intent. Unfortunately he has not done much with syntactical categories.
In school, “predicate” was what we called the verb in syntax, as in Subjekt, Prädikat, Objekt. Apparently that was a very parochial phenomenon.
Not at all. Or at least not always. So did we when we started analysing sentences around 1980. Until we didn’t and it became subjekt, verbal, objekt.
Russian school grammar terminology is weird. The simple SVO sentence parts are named подлежащее, сказуемое, and дополнение. If first 2 words ever had any use outside grammar, it’s long gone. Подлежащее is some sort of olden-days bureaucratism meaning “something to which an action should be applied” (and compositional meaning “lying under”), сказуемое in the same manner some (faux-)18c. word meaning “what is being said” and дополнение is an ordinary word meaning more or less “addition” or “complement”.
The English language instruction I had in school always divided sentences into only two basic pieces: subject and predicate. The direct object was part of the latter. My first exposure to a tripartite separation was in the “Pirate school” Far Side cartoon, which shows a subject-predicate-object division of the sentence: “I will keel-haul him.” (That was also, indirectly, how I came to learn about keel-hauling as a form of maritime punishment.)