Manual
input database NasDat
This manual
is based on W. Leo
Wetzels & Rob Goedemans (2009), NASDAT, an Electronic Database
for
the Storage of the Parameters of Nasal Harmony. Amerinda (32),
322-344.
Last update: 8 July
2009. For the most recent
version of this manual, see http://www2.let.vu.nl/oz/nasdat/manuals/manualInputDatabase.htm
1. Introduction
2. Using IPA codes
3. Using the input form
4. Tab page Language info
5. Tab page Syllables
6. Tab page Contrastive nasality
7. Tab pages Assimilation and Harmony
8. Tab page Segments
9. Tab page Examples
10. Tab page Stress
11. Tab page Remarks
12. Tab page Source(s)
1. Introduction
The Microsoft Access database NasDat_input.mdb
can be used by researchers to supply data for the Nasal
Harmony Database (NasDat). After entering your data, you can
send the database to Leo Wetzels at wlm.wetzels@let.vu.nl.
The new data will then be uploaded to the web version of the
database, after which they are accessible and useable for anyone
interested in the properties of NH.
To use the database, Microsoft Access version 2002 or higher is
required. If you have any technical
questions regarding the database, you can contact Eric Akkerman at e.akkerman@let.vu.nl.
If you have questions regarding the contents of the database, please
contact Leo Wetzels (see above).
Important:
the database is
compressed as a zip-file. Please save the database to your own computer
(using the right mouse button) and uncompress it there (usually also
via the right mouse button). After completion, compress the database
again before mailing it to Leo Wetzels.
2. Using IPA codes
In this database, the font Arial
Unicode MS is used to facilitate
working with IPA-codes. If this font is not yet installed on your pc,
you can download it from
the following web site:
http://support.microsoft.com/kb/287247
You are advised to enter IPA codes in the input database by means of
one of the special
programs that are available for this purpose, e.g. The Typological Database System
IPA Console:
http://languagelink.let.uu.nl/tds/ipa/index.html
N.B. The Unicode character set does not provide for a number of
superscript diacritic symbols used in IPA. In this database, these
symbols are represented by the string 'sup', followed by the diacritic
symbol.
Index
3. Using the
input form
When you open the NasDat input database by double clicking on
the file name, you will see the main input form on your screen. This
consists of a header, showing the language that is described in the
record that has focus, a tab element consisting of 9 tab pages and a
footer with a number of control buttons. From left to right, the
buttons on the custom made NasDat control bar are (hovering the mouse
cursor over the buttons shows there function in a small popup window):
- First
Record: go to the first record in the database.
- Previous
Record: go to the previous record.
- Next Record:
go to the next record.
- Last Record:
go to the last record.
- Add Record:
create a new record (i.e. add
new language description).
- Find Record:
search a record using Access' built in search function. N.B. By
default, Access searches for a complete field match of your search
string, Click the Match
option, and select Any
part of field, to search for a string as part of the
content of a field.
- Undo Record:
undo changes in the current
record. Undo is useful if something was accidentally deleted. It
nullifies all
the
changes you have made to this record since you entered it or since it
was last saved. You cannot use after you have gone to another record.
- Delete
Record: delete the entire record. This option can also be
used to delete empty records that were created accidentally.
- Help: opens
this manual. For this button to function correctly, the file manualInputDatabase.htm
should be installed in the same directory (folder) as the Access
database.
- Info:
shows administrative
information about the database in a popup window.
- Segment
features: gives a popup form showing the features
of all segments.
N.B. The navigation bar at the
bottom of the screen (that is innate to Access) is not needed to
navigate through the database. You can use this bar to check the total
number of records in the database (with is empty when you start using
it), and the number of the record that has the focus.
All information regarding one language must be entered in the
various fields of the tab element, and is consequently stored in one
database record. In the following paragraphs, the fields on each
tab page will be explained briefly. N.B. A short description
of the field that has focus is given on the status bar at the
botton om the Access window.
You can navigate through the fields on a tab page with the
keyus
<Tab> and <Shift>+<Tab>. You
can go to another tab
page with <Ctrl>+<tab>.
Please note:
- Access automatically saves all your input (and later
changes to the content of a record). You should not use the
menu-option Save
or Save as... for this
purpose.
- In Access, using the scroll wheel of the mouse always
results in scrolling through the database, not in scrolling through the
contents of the field that has focus. This cannot be changed in a
simple way; you are
therefore advised not to use the scroll wheel.
Index
4. Tab page
Language info
On this tab page, basic information about the language can be
entered.
Language:
Name of the language. Only one name is permissable in this field
– please enter the most common name that is used for the
language and enter possible alternative name in the following field.
Alternative names:
Alternative names of the language (if relevant), separated
by a semi-colon.
Dialect of:
Sometimes differences between speech communities are small but
significant, in which case one speaks of dialects of a single language.
If the language being described is considered a dialect of a larger
group, list the name of the dialect above under
‘language’ and list the group (parent language)
name here. If there is no name for the specific dialect, give the name
of the village, town, or region where the dialect is spoken.
N.B. It is important to spell the name of the parent language exactly
the same for all dialects. The drop down list for this field shows alle
dialect names that you have entered before. Please use this list to
enter identical dialect names.
People:
Name of the people that speaks the language. It is also possible to
enter multiple names if the people are referred to by several names in
the literature.
Genetic Affiliation:
Note here whether the language is a linguistic isolate, or provide the
information regarding its linguistic affiliation (group (in case of a
dialect), family, phylum).
Region:
More detailed geographical information, at least including the name of
the region in which the language is spoken, end with the country name.
Spelling of country names should conform to English, i.e. Brazil, not
Brasil; Italy, not Italia, etc.
Latitude and Longitude:
Mark latitude and longitude of the (centre of) the region where the
language is spoken.
Number of speakers:
Provide here the (estimated) number of speakers, and, if available, the
date of the source’s count.
Quote:
Insert a direct quote from the main primary source describing the Nasal
Harmony pattern.
Index
5. Tab page
Syllables
On this tab page, information about the syllable structure of the
language can be entered.
Syllable types:
This is basically a summation of the syllable structures possible in
the language given in the format CV, CVV, CCV, etc. Common onset and
coda restrictions can normally be included in this type of notation
(i.e. coda is always glottal stop CV ), but this will complicate
searches (a search for languages with codas would include CVC+CV+CVN,
etc.). Separate fields are therefore added for such restrictions, so
that it is possible to stick to a very broad description in this field.
Collapsed syllable structure notation of the type CV(C) is to be
avoided, since this also hampers searches. All syllable types should be
separated by commas.
Ambisyllabicity:
This field is designed to designate whether the language has segments
that belong to two syllables simultaneously. It may not be easy to
determine this at face value. Structural considerations play a role in
the decision of the status of, for instance, the /s/ in the Dutch noun
kassa or the /s/ in the English noun concert.
Restrictions on single
onsets:
If the language allows CV, CVV, CVC, CVVC (possibly with coda
clusters), can the single onset then be any consonant (including
glides) in the segment inventory of the language, or are certain
(classes of) segments ruled out? A well-known restriction, for
instance, is that English words cannot start with a velar nasal //,
while it is prolific as a word-initial segment in numerous Australian
languages.
Restrictions on onset
clusters:
Describe which types of onset clusters are allowed. In a CC onset
sequence, dependencies will exist between the first and the second
consonant. If the first is a stop, many languages will require the
second consonant to be a liquid. Another dependency might require the
first to be /s/ if the second is a stop. It is impossible to give
exhaustive lists here. Shorthand descriptions of the dependencies are
preferred: [Cstop]{l,r}, s[Cstop], etc.
Restrictions on single
codas:
If the language allows VC, VVC, CVC, CVVC (possibly with onset
clusters), can the single coda then be any consonant (including glides)
in the segment inventory of the language, or are certain (classes of)
segments ruled out? A well-known restriction, for instance, is that
Dutch words with long vowels cannot end with a velar nasal // *oo
*aa. This example shows that restrictions on single codas (and onsets)
need not apply across the board. If restrictions apply to a subset of
the possible syllables, they must be listed here.
Restrictions on coda
clusters:
Which types of coda clusters are allowed? In a CC coda sequence,
dependencies will exist between the first and the second consonant. If
the first is a stop, for instance, a language may require the second
consonant to be /s/. Another dependency might require the first to be a
liquid if the second is a stop. It is impossible to give exhaustive
lists here. Shorthand descriptions of the dependencies are preferred:
[Cstop]s, {l,r}[Cstop], etc.
Index
6. Tab
page Contrastive nasality
On the tab page Contrastive nasality, questions are asked
about the contrastive properties of consonants and vowels of the sound
system within which nasal harmony functions. In particular, questions
focus on the nature of contrasts present in the set of non-continuant
consonants. Together with the parameter setting for the nasal/oral
opposition on vowels, these facts could reveal interdependencies
between the presence/absence of specific contrasts and the
presence/absence of specific types of contour segments in the language
and possibly suggest explanations for the occurrence of the latter.
Voiced-voiceless stop
contrast: Yes/No
Does the segment inventory of the language show a contrast between
voiced and voiceless stops?
Nasal contrast on consonants: Yes/No
Does the segment inventory of the language show consonants that are
only differentiated by their nasality?
Nasal-voiced stop contrast:
Yes/No
Does the segment inventory of the language show a contrast between
nasal and voiced stops?
Nasal-voiced stop
contrast restrictions:
A language may have an opposition between voiced and nasal stops for
some places of articulation only, either across the board, or in a
given environment. If this is the case, state these restrictions.
Nasal-voiceless stop
contrast: Yes/No
Does the segment inventory of the language show a contrast between
nasal and voiceless stops?
Nasal-voiceless stop
contrast restrictions:
A language may have an opposition between voiceless and nasal stops for
some places of articulation only, either across the board, or in a
given environment. If this is the case, state these restrictions.
Nasal-oral vowel contrast:
Yes/No
Does the segment inventory of the language show vowels that are only
differentiated by their nasality?
Nasal-oral vowel contrast
restrictions:
Often the system of nasal vowels is more restricted than the system of
oral vowels. Does this apply to the language in question? One need not
state the restrictions here. Below in the window headed Examples, there
is a special field for entering the nasal vowel system of the language).
Index
7. Tab pages
Assimilation and Harmony
The tab pages Assimilation and Harmony contains the
parameters involved in nasal
assimilation and long-distance nasal harmony. Languages may have
assimilation, harmony, or both. Some languages have multiple
assimilation or nasal harmony processes. In this case, open a new
language entry for each different process and number the entries L 1, L
2, L 3, etc., where L represents the name of the language. Under
'Quote', on the tab page Language info, stipulate for
all the language entries created
that L has more than one assimilation and/or harmony process, provide a
concise characterization of the different types and state which type is
being described in the present entry. Only the entry in L 1 contains
the information that is characteristic of the language as a whole.
The Assimilation
and Harmony
tab pages have a
number of fields with preselection lists. Click the black arrow and
select the relevant option. If some lists appear incomplete, please
contact the database developers and/or mention this in the
field Remarks on the tab page Remarks.
Although most of the
parameters that appear on these tab pages are self-explanatory,
please find a succinct explanation below.
Assimilation
Nasal assimilation:
Yes/No
Nasal assimilation is understood here as the local spreading of a nasal
feature, to maximally one segment at one or each side of the triggering
segment (but see above).
Directionality:
Does nasality spread from the trigger to the left (regressive), to the
right (progressive), or does it move in both directions (bidirectional)?
Trigger segments class:
Note here which segments trigger nasal harmony/assimilation. Provide
the set of trigger segments as a class, if possible.
Trigger segments list:
Give here the set of trigger segments as a list.
Conditions on trigger
segments:
If there are conditions under which the trigger segments function as
such, list them here.
Target segments class:
Identify which segments receive nasality as the result of the
harmony/assimilation rule. Provide the set of target segments as a
class, if possible.
Target segments list:
List the set of target segments.
Conditions on target
segments:
If there are conditions under which the target segments become nasal,
register them in this field.
Harmony
For information about the fields regarding trigger and target segments, see above under Assimilation.
Nasal harmony:
Yes/No
Nasal harmony refers to cases of spreading whereby a feature spreads
further than the neighboring segment, when the harmony is obtained from
a floating nasal feature, or when the domain of spreading is a prosodic
or morphological domain, such as the syllable. This excludes cases of
one-directional assimilation (one
segment to the left or to the right), as in languages where the nasal
feature of an onset stop spreads to the preceding vowel or
bi-directional assimilation (one segment to the left and to the right),
as in languages where the nasal feature of an onset consonant spreads
to the preceding and following vowel only.
Directionality:
Does nasality spread from the trigger to the left (regressive), to the
right (progressive), or does it move in both directions (bidirectional)?
Domain of spreading
(both phonological and morphological):
Provide here the largest morphological or prosodic unit within which
spreading takes place, for example: root, stem, word, syllable, foot,
phonological word, clitic group, etc.
Valence:
Specify unary or binary. Valence is unary if the single value [nasal]
is sufficient to account for the data. Valence is binary if
[±nasal] is required, as is proposed here for Kaingang, and
which, for different reasons, can be argued for Desano and Barasana
(for Desano, see the contribution of Wilson Silva to this volume).
Transparent segments class:
Provide here the class of segments that let nasality pass, but that do
not become nasal as part of the spreading process.
Transparent segments list:
Provide here the list of segments that let nasality pass, but that do
not become nasal as part of the spreading process
Blockers class:
Provide here the class of segments that block the spreading process, if
possible.
Blockers list:
Provide here the list of segments that block the spreading process
Prenasalization of
blockers: Yes/No
In some languages the blocking segments are prenasalized. If this
happens in the language in question, state whether or not this applies
to the class (or a subclass) of blocking segments.
Postnasalization of
blockers: Yes/No
In some languages the blocking segments are postnasalized. If this
happens in the language being entered, state whether or not this
applies to the class (or a subclass) of blocking segments.
Contour segments:
Does the language have contour segments that cannot be considered
variants of an underlying segment that is not a contour segment? This
means that the contour property is itself contrastive in the language.
Notice that it is possible to choose ‘No’ in this
entry and still have contour segments in the language, as in Kaingang,
where contour segments are derived.
Source of contour segments:
Languages may have contour segments that are derived from plain
segments. Here we list three possible sources.
- Spreading of nasality.
Contours can arise through the contextual influence of nasal segments.
In a typical case VNC arise as variants of underlying sequences VC
through progressive nasal spreading from the nasal vowel to the
consonant (/ãp/> [amp]).
- Spreading of orality.
Contours can arise through the contextual influence of oral segments.
In a typical case VCN arise as variants of underlying sequences VN
sequence through progressive spreading of orality from the nasal vowel
to the consonant (/am/ > [abm]).
- Spontaneous contours (phonetic enhancement of non-sonorant
voice). These contour segments typically occur as variants of
underlying
voiced oral stops, especially in the syllable onset (see for example
discussion by Ohala 1983, Iverson & Salmons 1996, Keyser
& Stevens 2006, Wetzels 2008).
It may sometimes be difficult to
decide whether derived contour
consonants are caused by the spreading of orality or by phonetic
enhancement of non-sonorant voice.
Nasal morphemes:
Yes/No
Sometimes the nasal feature represents some morphological category in
the language, in which case it must be listed in the lexicon as a
morpheme by itself, as, for example, in Terena, where /N/ represents
the 1person singular morpheme.
Commutation level:
Establish the level at which nasality is contrastive; specify
segment/syllable/morpheme/other.
Spreading source:
Indicate phoneme or nasal morpheme
Floating nasal:
Yes/No
In many languages exhibiting nasal harmony, words are either
‘entirely’ nasal or entirely oral. In such cases
nasality is usually represented as a floating feature that is linked by
the harmonic segments by association.
Stylistic conditions:
Assert whether the rule is optional, if it only
applies in fast speech, etc.
Adjacency conditions:
Specify here any conditions on the adjacency of nasal(ized) segments,
such as segment adjacency, syllable adjacency, or none.
Index
8. Tab page Segments
On this tab page, segment information about the language can be
entered. This tab page is the most complex section of this database in
all respects. It links the main sections of NasDat to a table that
contains all the segment information stored in the UCLA Phonological
Segment Inventory Database (UPSID, which can be found at http://www.linguistics.ucla.edu/faciliti/sales/software.htm#upsid).
Note that only the segments with their properties are included. UPSID
contains phoneme inventories for hundreds of languages based on
selections from the present list of segments. These inventories are not
included in NasDat. On the segment tab page, one can select which
segments from the UPSID range are present in the phoneme inventory of
the language being inputted. Because there are 920 different segments
represented in UPSID, data entry is far from trivial. We have chosen
for a list and sub-list presentation to guide encoders through the
selection of segments. The procedure is as follows:
- Select a main segment type from the list labeled Segment type (Plosives, Implosives, Ejective
stops, etc.)
- For some segment types, sub types are distinguished. If
this is the
case, a sub-list opens below the main segment type list, from which you
can select the relevant sub type. Upon clicking Plosives, for
instance, a sub-list opens below it, containg sub types like bilabial, labiodental, dental,
etc.
- After selecting a type and, if relevant, a sub type, a list labeled Segment
will open,
containing all segments that are of the
articulation type that was selected. In this third list click the
desired
segment, and
with the buttons below it choose the relevant action to take regarding
this
segment.
[Add Segment] puts it in the phoneme inventory of
the language, which is shown
in the window to the right of the buttons and lists.
[Delete
Segment] removes
it from that inventory. It will be updated in real time.
[Show Features] shows the
relevant UPSID info pertaining to the selected segment in a
popup window.
- Repeat this process to enter all segments that occur in the
language.
Segment inventory:
This lists shows all segments that have been entered for the language.
Clicking the button [By
Category] will yield a popup window with an overview of
these segments in three lists of
phonemes: one for consonants, one for vowels and one
for glottals and approximants. Use the “back”
button to go back to the main form. N.B. This list cannot be edited
directly. You can only delete a segment from this list by selecting it
first, following the procedure described above, and then clicking the
button [Delete Segment].
Anomalous Segments:
In this field the rare segments of the language are listed. Note that
these segments are also present in the inventory that can be found in
the list on the right hand side of the segments page. This field simply
conveys which of the segments in that list are anomalous according to
the UPSID definition (rare segments, loans, etc.). For 317 languages
the status of segments can be checked in UPSID. For the rest, the
decision is up to the encoder.
Index
9. Tab page Examples
On this tab page, you can insert examples that illustrate the
contrastive segments of the language. Remember that these examples can
be entered easily in phonetic transcription using the IPA Console mentioned in paragraph
2 above.
Index
10. Tab page Stress
On this tab page, the stress system of the language can be
described.
Stress rule(s):
Describe the stress rule(s) here. As such descriptions may
often be quite prolix, and thus not conducive to easy searching, stress
patterns should be coded. To facilitate easy encoding, we use the
scheme used in the StressTyp database (where it has
had 15 years to mature) Stress codes for 510 languages can be found
there. For stress systems that are not yet represented in StressTyp,
new codes need to be created. To do so consult the StressTyp manual on
line http: stresstyp.leidenuniv.nl/ or Rob Goedemans (who will be glad
to do this provided the stress information of the language). His email
address is r.w.n.goedemans@hum.leidenuniv.nl.
Stress Source:
Provide the descriptive source for the data in the previous field. Only
use primary sources. Second hand descriptions in theoretical works on
stress may be inaccurate or biased.
Index
11. Tab page Remarks
On this tab page, any remarks pertinent to the complete entry
for
this language can be entered. This field typically contains information
that the encoder could not put into one of the regular fields, but is
relevant nonetheless, or information regarding alternative analyses of
the data. If there is doubt about certain data, that can be expressed
here also. Lastly, if the language has nasal morphemes, the
morphological function of the nasal morpheme may be stipulated in this
area. The field ends with the name of the person who entered the
remark, if it is not the author of the record.
Index
12. Tab page
Source(s)
On this tab page, you can provide information about the sources of the
data that entered in the database.. First, name, affiliation and email
address of the reasearcher who is responsible for entering
the data for
the language this record should be entered here. Next, you can indicate
if the data
are based on acoustic
evidence (yes/no) and/or tested in a
laboratory (yes/no).
Data collected:
In this field, you can indicate whether the data that you have entered
for the language are collected by yourself ('the
author'), from secondary sources, or both. Here, the
notion ‘secondary source’ means that the data have
been encoded from published sources that were not personally collected
by the person who enters the data.
Primary source(s):
List here the literature that describes the primary linguistic data of
the language being entered. These are usually descriptive grammars,
dissertations, or papers, but may be theoretical papers that are based
upon data gathered by the author(s). N.B. Remember that youy cannot use
italics in the fields of this database.
Secondary source(s):
List here the literature that uses the data of this language gathered
by others for theoretical, comparative, typological, or other purposes.
N.B. Remember that youy cannot use italics in the fields of
this database.
Index