Lesson 13
pacimoi seltadni
Modification Pairs
tanru
If you are current in your vocabulary study, you have worked with a
little more than *** gismu so far, and thus know how to express ***
basic concepts; we have studied about 50 cmavo as well.
Using conversion operators (selma'o SE), you can access the non-x1
sumti, effectively multiplying that number by 2 to 4 times.
Using the abstraction operators (selma'o NU), you can access the
event/state, property, and amount abstracts of each gismu and each
converted gismu, multiplying that total by an additional 3 times.
You thus can feel some amount of control over the equivalent of about
*** 'vocabulary words'. In this lesson, we will add another ***
gismu, and increase the total to ***. Using just 2-term tanru, you
can square this number. Did you think you would have a Lojban
vocabulary of *** million after just *** lessons? get ready to have
fun and creatively explore the power of Lojban vocabulary.
13.1
What is a tanru?
loi tanru du ma
A tanru is a Lojban metaphor. tanru are something like English
adjective-noun and adverb-verb combinations. They go beyond these
concepts by combining and expanding upon them. In this, they are
similar to Chinese metaphor more than to English. Since all Lojban
brivla can act as 'adjectives' or 'adverbs', any of them can be used
to modify another brivla acting as a 'noun' or 'verb'. There are no
restrictions at all. You can take a pair of words, a modifier and a
modificand, and attach still another brivla as a modifier to the pair,
or you can use the pair to express a modification to the single word.
The process can continue indefinitely. There is no reason why you
can't have a 100-brivla string of modifiers and modificands, grouped
successively in pairs. The concept of tanru expresses exactly that:
the combination of Lojban concepts in pairs such that the first
concept modifies the second.
A tanru is made up of brivla representing concepts that are related to
the concept you are attempting to communicate. The relationship is
not necessarily unambiguous. A blue-nest in some way nests someone or
something. It could be a nest for blue eggs, or blue people, or it
could be a house painted blue either partially or completely. It
takes a more elaborate tanru to distinguish these less ambiguously (if
such is important), or you can use non-tanru methods to expand your
communication unambiguously. Most of the time you won't: it is
long-winded, and unnecessary to most understanding.
(Before proceeding with the explanation of tanru, a special note is
needed concerning "zdani". In English, we make distinctions between
"lair", used for animals, "nest", usually used for birds, and "house",
used only for humans and domesticated animals (such as dogs and
birds). However, we often metaphorically interchange the terms, and
Lojban thus uses "zdani" to refer to that essence that all of these
hold in common. You can be more specific using tanru, where this is
important. Note also that the distinction between "house" and "home"
that is often referred to by English poets is conveyed by the
distinction between "zdani" and "se xabju".)
The order of modification in tanru is from left to right:
zdani vitke is a nest(house)-guest, while:
vitke zdani is a guest-nest(house).
This can be kept clear by imagining the resulting tanru as a selbri.
The place structure of any tanru is the place structure of its final
brivla. Thus the following place structures:
x1 is a zdani vitke (house-guest) of host x2 at location x3
x1 is a vitke zdani (guest-house) for x2
All tanru are ambiguous in meaning. zdani vitke could refer to the
unlikely event of a 'house which visits'. While visiting, it might be
considered a vitke zdani, saying nothing of who might live in it. But
note that the place structure convention limits interpretation of the
latter tanru. The x2 place of the second tanru always refers to the
house-dwellers. The concept of a house that visits a location is
better expressed by the first tanru, which has the place structure of
"vitke". Checking the place structure of the final unit of a tanru
may often help you determine which is the correct order.
The most common mistake in making tanru is to put them in the wrong
order, often because idiomatic English expresses the words in the
reverse order. So remember the desired place structure. Reversing
the following tanru could be embarrassing:
le mamta patfu
The mother's father. (The maternal grandfather, a male.)
le patfu mamta
The father's mother. (The paternal grandmother, a female.)
Some further examples of tanru will show the wide variety in meanings
that can be extracted from various tanru, if you are imaginative.
tanru are more specific than gismu they restrict the domain of meaning
being referred to, and provide additional information that isn't
carried in a single brivla; however, because the relationship between
the two terms is semantically ambiguous, tanru are thus metaphorically
vague.
klama jubme
one-who-goes table. Could be:
The table is the one-who-goes; e.g. a wheeled table.
The table is owned by the go-er; i.e. the go-er's table.
The table is used for go-ers; a sports-doctor's examining table?
It is a table for use in a go-er; e.g. a ship's table.
It is a table when it is a go-er (otherwise it's a chair?)
sutra daspo
fast one-who-destroys. Could be:
The destroyer is fast.
It is a destroyer of fast-things.
It destroys things by use of a fast-thing.
The destroyer is owned by a fast-thing.
It destroys when it is fast.
karce kukte
car tasty-thing. Could be:
The tasty-thing is a car.
It is tasty to a car.
It is tasty with the flavor of a car.
The tasty-thing is for use in a car.
The tasty-thing is shaped like a car.
To learn the essence of tanru interpretation, you must learn to
imaginatively think of possible meanings (using some simple
conventions to limit the possibilities), and then determine why the
speaker used this particular tanru as opposed to some other to help
weed out the interpretations that are not intended by the speaker.
tanru interpretation is thus an exercise in "creative plausibility
analysis". You should choose the most plausible interpretation given
the known context, and be prepared to be wrong if the context is
uncertain.
For the essence of tanru making, the reverse process is used. Think
of a few possibilities, then try to analyze how a listener might
misinterpret each possibility. Thus, making tanru gets to the true
essence of human communication: putting yourself in the mind of the
other person, and figuring out what that mind is thinking. It's a
neat concept. It is obvious, though, that no Lojbanist will be good
at tanru-making for a while. Until we have many people thinking
Lojbanically - seeing the world through tanru - we won't have the
variety of thought needed for fluency in this form of mind-reading.
When we have people from a variety of cultures involved in
communication, the making and interpreting of tanru will be even more
challenging. Both German and Chinese use a form of metaphor-making
similar to tanru to express complex notions. In many cases, a native
Chinese or German speaker will each choose tanru for many concepts
that differ from one another, and which also differ from those an
English-speaker would choose. For Lojban to bridge multiple cultures,
it must in some way build its own culture, which cannot be too alien
to those of other languages, but rather a true blend or hybrid
culture. Alternatively, speakers from different cultures will have to
be more specific in their tanru, adding in extra terms that clarify
how one concept modifies the other.
We slipped something into the definition of tanru by using the word
'concept'. In addition to using the brivla as 'units' in a tanru, you
can use simple tanru as units for more complex tanru. Moreover, you
can use conversions and abstractions and negations of brivla as units.
In tanru, you should use the x1 place in building your meanings. If
you want to use another place, use the brivla converted with selma'o
SE:
ve mrilu tanxe
mail-origin box (a mailbox where you deposit mail to be delivered)
te mrilu tanxe
mail-destination box (a mailbox where mail is delivered to)
One problem/mistake that may come up in tanru-making is the use of the
x1 place of a gismu when you really want an event, property, or,
quantity abstraction.
le bajra terpa
The runner fear-er (One who fears runners.)
le terpa nu bajra [kei]
The fearing/fearful act-of-running. (What you do when you meet an
axe-murderer.)
Note that you can use a tanru as the selbri inside of a "nu ... kei"
abstraction clause; you can even specify places for the bridi,
resulting in a very complex tanru:
le nu mi bajra kei terpa
The act-of-me-running fear-er. (One who fears my running)
tanru of the last sort are permitted in descriptions, as shown in the
examples, or in selbri. It is difficult to convey the sense of some
of the latter expressions to an English speaker (we just don't phrase
things in that way), though examples can be found:
le nanmu [ku] cu nu le ninmu [ku] cu klama le zdani [ku] kei kurji le
cifnu (le nanmu [ku]) cu ({nu
kei} kurji) le cifnu The man is an
event-of-the-women-coming-to-the-house type-of caretaker of the baby.
The (male) babysitter has women over to the house (he's sitting at) a
lot.
A tordu bajra is a short-runner, not a short-run. To get the 'verb'
as in the latter, use nu bajra [kei] (the event of x1 running) in the
tanru, giving tordu nu bajra. You can also get an 'adjective' which
is otherwise hard to convey by using the quality abstractor cmavo 'ka'
(This is a little harder to come up with good English-translatable
examples for, so we won't. Let it be an exercise...).
Now that you have learned many gismu as part of your vocabulary
studies, you have a lot of potential uses for them in tanru. Some of
the gismu are primarily intended for tanru use, such as "mabla" and
"zabna" from the last lesson. These turn a brivla or tanru into an
insult or derogative, or into a compliment or ameliorative,
respectively. This is important since most insults and compliments
are highly cultural. Expressions which might be compliments in one
culture or context might be insults in another, and vice versa. In
Lojban, by convention, an expression is not presumed to be intended as
an insult unless it contains "mabla" in some form, nor is it presumed
to be a compliment without "zabna". If you omit "mabla", you might be
perceived as culturally insensitive in your choice of expression, but
you should not be perceived as insulting. Since Lojban must be
relatively insensitive to culture in order to be culturally neutral,
this convention is vital to understanding.
Look over the various gismu that have been presented in the last few
lessons. Try to imagine how they could be used in tanru. After one
more topic, we will have an exercise that will stretch your mind
Lojbanically and give you ideas on how tanru are made and used.
13.2
3-or-more place tanru
su'ocimei terbridi tanru
tanru are composed of pairs of modifiers and modificands. We have
discussed using brivla, converted brivla and abstracted bridi as
components of either position. You can also fill either position with
tanru. Let us first discuss the use of tanru in the modifier
position.
Assuming you are in a class studying Lojban, to describe that class,
you might use:
le nu le lojbo tadni cu penmi [kei] [ku]
the event of (the lojbanic-studiers meet(ing))
You can drop the "le" to make this a selbri:
nu le lojbo tadni cu penmi [kei]
but the structure is complex and a bit convoluted for comfortable
speech. So why not move the "nu" to mark only "penmi", and use the x1
sumti that has been exposed as a modifying tanru:
lojbo tadni nu penmi [kei]
lojbanic-studier-meeting
The result is shorter, and almost certainly will not be misinterpreted.
Given one assumption, that is. You have to know which way to group
the tanru. If you group the tanru as:
(lojbo tadni) nu penmi [kei]
Lojbanic-studier type-of meeting
you will get the intended meaning. If, on the other hand, you group
the tanru (incorrectly, as marked with the asterisk):
*lojbo (tadni nu penmi [kei])
Lojbanic type-of studier-meeting
which could easily be referring to a class in philosophy which happens
to be taught in Lojban. (The first tanru might possibly be taken with
this meaning, but only by jumping to the conclusion that the speaker
is referring to students who are Lojbanic in that they attend classes
taught in Lojban. Without the modificand of "nu penmi [kei]", you
would be unlikely to choose this less plausible alternative for "lojbo
tadni". But in interpreting tanru, you should be presuming the most
plausible interpretation.)
Why is the second interpretation incorrect? How can you pick the
grouping which is uniquely intended? The answer lies in Lojban's
unambiguous grammar. Only one possible grouping structure can be
permitted for the tanru that was given. You need to be able to
uniquely determine which words have the modifier role, and which words
have the modificand role in a tanru.
In 3-or-more-place-tanru that is otherwise unmarked, you always assume
'left grouping'. This is easiest to describe by showing the result
with nested parenthesis. In a longer tanru, the grouping is:
[{(brivla1 brivla2) brivla3} brivla4] brivla5
What the parentheses mean is that you first take the tanru "brivla1
brivla2", determine its meaning, and use it to modify brivla3. This
three-place tanru is then used as the modifier, while brivla4 is the
modificand. This can be repeated for brivla5, using the 4-place tanru
as the modifier unit. If the tanru were longer, you could extend the
process indefinitely. In any group of three-or-more units without any
grouping markers as described below, you presume that the first unit
modifies the second unit in a modifier-modificand tanru pair. This
tanru in turn serves as a modifier unit for the third term, forming a
new tanru. This tanru then is a modifier for the fourth term, and so
forth.
But this isn't always the interpretation that you want. What if you
want the non-left-grouped interpretation that was marked with the
asterisk above. We now explore how to use a tanru as the modificand
unit of another tanru.
13.3
On ke and ke'e
me zo ke .e zo ke'e
The secret lies in the grouping cmavo "ke" and "ke'e". The use of
"ke" and "ke'e" overrides left grouping. You first determine the
interpretation of the contents of the inside of the 'parentheses'
using left-grouping rules, then use the result as a unit in further
left-grouping of tanru.
Thus:
brivla1 ke brivla2 brivla3 ke'e
is grouped as:
brivla1 (ke brivla2 brivla3 ke'e)
and the tanru formed by "brivla2 brivla3" becomes the modificand,
while brivla1 becomes the modifier.
To get the asterisked interpretation from the last section, use a
ke/ke'e pair to force the grouping you want:
lojbo {ke (tadni nu penmi [kei]) ke'e}
Lojbanic type-of studier-meeting
You can see that "ke" is well-translated as "type-of".
We will see later in this lesson that in most cases the right
parenthesis "ke'e" can be elided. It usually occurs at the end of the
tanru, and the word that comes after the tanru usually makes it clear
that the scope of the "ke" has ended along with the tanru. We will
thus mark these elidable "ke'e"s with brackets, as we have been for
the NU-clause-closing "kei"s.
Let us look at the possible groupings of four terms that can be
expressed with "ke" and "ke'e", noting that you can use these cmavo
multiple times to achieve nested parentheses:
{(brivla1 brivla2) brivla3} brivla4
brivla1 (ke {brivla2 brivla3} brivla4 ke'e)
(brivla1 brivla2) (ke brivla3 brivla4 ke'e)
{(brivla1 brivla2) brivla3} (ke brivla4 ke'e) is the same as the
unmarked form (it is legal, though wasteful, to surround a single
brivla with a ke/ke'e pair)
brivla1 (ke brivla2 {ke brivla3 brivla4 ke'e} ke'e)
{brivla1 (ke brivla2 brivla3 ke'e)} brivla4
We will let you work out the possibilities for 5-unit tanru in the
exercise that follows, as well as show the interaction between
NU-clauses and ke/ke'e pairs.
Before doing that, however, let us give a couple of examples, which
incidentally will show why "cu" is sometimes necessary. We will show
the sentences with elidable terminators removed to make comparison
easier.
13.3.1
mi gleki prenu xebni
mi (gleki prenu) xebni
I happy-person hate. (I hate happy people.)
mi gleki ke prenu xebni
mi gleki ke prenu xebni [ke'e]
mi gleki (ke prenu xebni [ke'e])
I happily person-hate. (I am happy in my hatred of people.)
le gleki prenu xebni
le (gleki prenu) xebni
The happy-person hater. (The one who hates happy people.)
le gleki prenu cu xebni
le gleki prenu [ku] cu xebni
(le gleki prenu [ku]) cu xebni
The happy person hates. (The one who is a happy-person hates.)
le gleki cu prenu xebni
le gleki [ku] cu prenu xebni
(le gleki [ku]) cu prenu xebni
The happy-one person-hates. (The one who is happy hates people.)
le nu le gleki cu prenu cu se xebni
le nu le gleki [ku] cu prenu [kei] [ku] cu se xebni
(le {nu cu prenu [kei]} [ku]) cu se xebni
The-act-of the happy-one's being-a-person is hated.
(What is hated is that the happy-one is a person.)
le nu gleki prenu cu se xebni
le nu gleki prenu [kei] [ku] cu se xebni
(le {nu [kei]} [ku]) cu se xebni
The-act-of happily being-a-person is hated.
(What is hated is that someone is a happy at being a person.)
le nu le gleki prenu cu se xebni
le nu le gleki prenu [ku] [kei] [ku] cu se xebni
(le {nu [kei]} [ku]) cu se xebni
The-act-of the happy-person's being hating.
13.3.2
le sutra cenba karce stizu
le ({sutra cenba} karce) stizu
The (quickly-varying)-car type-of seat.
(The seat is in a quickly-varying car - a convertible that can
be quickly converted perhaps.)
le sutra ke cenba karce stizu
le sutra ke cenba karce stizu [ke'e]
le sutra (ke {cenba karce} stizu [ke'e])
The quick type of (varying-car)-seat.
(It is a quick seat for a varying-car - still a convertible, but it is
the seat that is fast, whatever that means. Perhaps the seat is
self-propelled and moves fast while removed from the car.)
le sutra ke cenba karce ke'e stizu
le {sutra (ke cenba karce ke'e)} stizu
The quick type-of varying-car, seat.
(It is a seat in a quick type of varying-car - the seat is in a fast
convertible)
le sutra cenba ke karce stizu
le sutra cenba ke karce stizu [ke'e]
le (sutra cenba) (ke karce stizu [ke'e])
The quickly-varying type-of car-seat.
(The car-seat is quickly varying - probably with a lever under
the seat, but this needn't be in a convertible.)
le sutra ke cenba ke karce stizu
le sutra ke cenba ke karce stizu [ke'e] [ke'e]
le sutra (ke cenba {ke karce stizu [ke'e]} [ke'e])
The quick type-of varying type-of (car-seat).
(It is a quick car-seat-that-varies - this is another
adjustable car-seat, but it really moves, not necessarily while in a car.)
13.4
Grouping in Longer tanru
ve fendi fe lo tanru
The following three-part and four-part tanru illustrate in more detail
the uses of ke and ke'e.
blanu botpi tanxe
ta1 is a <(exhibiting the color blue) kind-of bottle
containing bo2, and made of material bo3> kind-of
box/carton containing ta2, and made of material ta3
blanu ke botpi tanxe ke'e
ta1 is a (exhibiting the color blue) kind-of containing ta2, and made of material ta3
ke/ke'e grouping first groups the last two terms as a tanru ("botpi
tanxe") forcing the blanu to modify this unit. In short, "blanu botpi
tanxe" is a "blue-bottle type of box", while "blanu ke botpi tanxe
ke'e" is a "blue type of bottle-box".
When tanru are extended to four or more terms, left grouping
precedence makes ke/ke'e grouping even more valuable. If we wanted to
precede the above with "cmalu", the result becomes:
cmalu cm1 is small in property/dimension cm2
blanu botpi tanxe
ta1 is a <(exhibiting the color blue) kind-of bottle
containing bo2, and made of material bo3> kind-of
box/carton containing ta2, and made of material ta3
cmalu blanu botpi tanxe
ta1 is a <[(small in property cm2) kind-of (exhibiting
the color blue)] kind-of bottle containing bo2, and
made of material bo3> kind-of box/carton containing
ta2, and made of material ta3
schematically: <[cmalu blanu] botpi> tanxe
More simply, the 4-place tanru translates roughly as a
smallishly-blue-bottle box, necessitating one to figure out what
smallishly-blue might mean, especially when applied to bottles.
ke/ke'e grouping can produce several variations on this grouping, and
hence on the resulting meaning (although in all cases, the place
structure of the entire relation is that of "tanxe", the final term).
With one ke/ke'e pair, we can produce three variations of the original
grouping. Compare the marked and unmarked forms:
cmalu blanu botpi tanxe
schematically:
<[cmalu blanu] botpi> tanxe or smallishly-blue bottle type-of box
cmalu ke blanu botpi ke'e tanxe
ta1 is a <(small in property cm2) kind-of [(exhibiting the color
blue) kind-of bottle containing bo2, and made of material
bo3]> kind-of box/carton containing ta2, and made of material
ta3
schematically:
tanxe or small blue-bottle kind-of box
In this example, the ke/ke'e pair forces "blanu botpi" to group first,
then normal grouping from the left takes over, and "cmalu" modifies
the 2-part tanru. Then this 3-part tanru is the modifier for "tanxe".
cmalu blanu ke botpi tanxe ke'e
ta1 is a <(small in property cm2) kind-of (exhibiting the
color blue)> kind-of containing ta2, and made of
material ta3
schematically:
or smallishly-blue type-of bottle-box
In (1.8-4b), the ke/ke'e pair forces "botpi tanxe" to group first.
Then, normal left-grouping determines the rest. The priority pairing
does not prevent the two left terms ("cmalu blanu") from grouping as
they normally would, but it does force this tanru as the left term to
modify the right two terms as a single unit.
cmalu ke blanu botpi tanxe ke'e
ta1 is a (small in property cm2) kind-of [<(exhibiting the color
blue) kind-of bottle containing bo2, and made of material bo3>
kind-of box/carton] containing ta2, and made of material ta3
schematically:
cmalu [blanu botpi] tanxe or small blue-bottle kind-of box
In this example, the ke/ke'e pair forces the three terms to the right
to be evaluated first. Within that unit, "blanu botpi" groups first
by normal left-grouping, and this pair modifies "tanxe". Then "cmalu"
modifies the 3-part tanru.
A single ke/ke'e pair can be placed around other pieces of the 4-part
tanru with no effect. For example, placing the words around the first
two terms forces them to be evaluated first, but this is what
left-grouping requires anyway. We note in passing that a ke/ke'e pair
is grammatically permitted but totally meaningless when surrounding
the entire 4-part tanru, or when surrounding an individual word.
You can put two ke/ke'e pairs into a tanru as well. With two "ke"s in
a tanru, the result is like nested parentheses in a mathematical
expression. In a 4-term tanru, only one nested grouping is truly
productive; the others simplify into a single pair of ke/ke'e, or into
the totally unmarked form. The one useful form is:
cmalu ke blanu ke botpi tanxe ke'e ke'e
ta1 is a (small in property cm2) kind-of < (exhibiting the color
blue) kind-of [bottle containing bo2, and made of material bo3
kind-of box/carton] > containing ta2, and made of material ta3
schematically:
cmalu or small kind-of blue bottle-box
In this construction, the two explicit groupings both take precedence
over left-grouping. The innermost parenthesis takes precedence over
the outermost.
Since we read (and, in effect, hear) Lojban from left-to-right, this
type of nesting may prove cumbersome for a listener to interpret.
Other techniques in tanru grouping, discussed later in this text, are
simpler than nested ke/ke'e constructs, reducing the chance of errors
in interpretation. 4- part tanru are uncommon in spoken Lojban
anyway, since a speaker has to build such tanru mentally before saying
them.
You can explore the groupings possible in 5-or-more part tanru on your
own. You will seldom have a need to use such complex tanru, but you
can get used to the effects of ke/ke'e grouping as opposed to unmarked
left-grouping by such 'word-play'.
13.5
Family Relationship Words
lanzu gismu
This section gives a variety of terms used in family relationships.
The expression of such relationships is a rich avenue for exploration
of the power of tanru.
The primary Lojban words for family relationships do not refer to
genetic relationship, but rather to culturally defined roles. In
English, your father is usually your male genetic parent, but might be
an adoptive parent or even someone else who assumes a father-like
role. It is also a general principle that there are three sets of
words for major family relationships: one sexually neutral, one
specifically female, and one specifically male. This structure is
paralleled in the words for human beings generally, without reference
to family. Where there are no separate gismu for specifically sexed
relatives, tanru can be formed using fetsi (female) or nakni (male).
The major family relationship words can be summarized in the following
table:
Category Sexually Neutral Female Male
or Female Role or Male Role
human being remna fetsi prenu nakni prenu
adult makcu ninmu nanmu
(adult, mature)
child (young human) verba nixli nanla
parent rirni mamta patfu
(raise/rear)
sibling tunba mensi bruna
offspring panzi tixnu bersa
grandparent/ancestor dzena fetsi dzena nakni dzena
aunt/uncle famti fetsi famti nakni famti
cousin tamne fetsi tamne nakni tamne
Aunts, uncles, and cousins are defined more loosely in Lojban than in
English; an aunt/uncle is a relative of one's parent's generation, and
a cousin is a relative of one's own generation who is not a sibling.
The general word for "family" in Lojban is "lanzu", which is not
limited to the nuclear family. It can be used to refer to an extended
family (lanzu ckini or darno lanzu), a social family-like structure
(jikca lanzu), or a tribe (cecmu lanzu, with "cecmu" meaning
"community"), using appropriate tanru.
There are also gismu which are explicitly tied to genetic
relationships, specifically "rorci", whose place structure is:
x1 engenders, procreates, or begets x2 with co-parent x3
and is specific to sexual reproduction. Similarly, "panzi" has the
place structure:
x1 is the offspring of x2
and can be used both for human beings and animals, and also for
plants, microbes, or other creatures that don't require two
individuals to reproduce.
The relationship of birth is represented by "jbena":
x1 is born to x2 at time x3 and place x4
In the modern world, it is very possible for all of these
relationships to be distinct. A child may have A as his genetic
mother (fetsi rorci or fetsi se panzi), B as his birth mother (se
jbena), and C as his adoptive/rearing mother (fetsi rirni). Probably
C will be mamta as well.