NOVES SL. SUMMER 2003. THE SYSTEM OF LINGUISTIC INDICATORS (2)

The term "System of Linguistic Indicators" is adopted, as adequate for the purposes of this project. It will be seen that the list given above is not in itself a system in the full sense. In its present version it consists of 7 sections, 35 subsections and 85 indicators. See Appendix 2: List of sections and subsections. (PDF 12 KB).

The SIL is a collection of variables expressed in the form of indicators relative to the situation of the language in a given context and with specific language policy objectives; these variables need to be sensitive to exterior change, they have to be quantifiable, and lend themselves to analysis in an integral way.

Turning now to the sources of information for the SIL, these have to contain series of data, not isolated or one-off data, either derived from official figures or from studies over a period of time. The distinction needs to be made between official or non-official surveys (observational data and counts) and answers to direct questioning. The SIL is a statistic which is included in the Pla estadístic anual (Annual Statistical Plan) of the Generalitat of Catalonia. It is proposed that part of the data collected via Ofercat or Indexplà should also be used for the System of Linguistic Indicators.

In terms of types of indicator, it is useful to distinguish between: language knowledge, language use, and language provision. There are also "structural" indicators (every five years and containing few variables) to be distinguished from "conjunctural" indicators (annual, and more sensitive). In general, it is observed that many of the indicators of the current version are indicators of presence or provision of Catalan and need to be filled out with language use indicators. Graph 4 features a "provision" indicator, indicating supply or availability of Catalan.

Graph 4. Percentage of commercial radio stations that broadcast more than 50% in Catalan 1996-2002

In terms of structure, the SIL can be visualised as a matrix in three dimensions: type of indicator (knowledge, use and provision), ongoing nature of the indicator (structural and conjunctural) and its sector (current or otherwise). It is not so important to have data for all cells of the matrix, so much as to see its limits and shortcomings.

While the SIL is mainly used to gauge the Catalan language, it can equally be used to obtain data on both of the official languages of Catalonia -that is, Catalan and Spanish. It will prove useful to compare data on the two official languages, and indeed compare with other languages where necessary.

Where possible, the indicators have to be comparable internationally (to be used for international comparisons). The proposal is to establish the necessary contacts to be able to compare with other languages and other multilingual countries. Often the comparison between different indicators in different ambits or subambits and the comparability with different years is more important that the data seen in isolation.

In the first version, it is recommended that the SIL have a basic battery of indicators which can be progressively added to. It will be acceptable, in the initial phase, to have a good if not so extensive selection of indicators. Special emphasis is put on the fact that the SIL may be "read" at two levels: a complex level with all the indicators and a simpler level with few indicators. Table 1 shows the present number of indicators with data and with no sectorial classifications.

It would be worthwhile having available aggregate or synthetic indicators for each of the sectors of the current list. It would also be useful to give priority to indicators in each sector or to introduce objective weightings by means of objective information relating to the economy and productive sectors. Also under consideration is the creation of an overall synthetic indicator and the difficulties involved in weighting and aggregating. To give a familiar example of a synthetic indicator, there is the Retail Price Index (measuring inflation) which aggregates information on various products. It should be noted that there are two ways of looking at a synthetic indicator: a) as an aggregate indicator resulting from a mathematical operation and b) as a representative indicator, which results from a strategic selection and incorporates other indicators. It can be seen that not all sectors, nor all indicators will be aggregatable in a single general or sectorial indicator. For example: "external dissemination" indicators will not be able to be aggregated with the remainder of the indicators.

The SIL has to be able to integrate new phenomena as they emerge such as (modern) immigration or the information technologies which are important enough to have an influence on the whole of Catalan society.

Positively valued is attached to the search for consensus in the SLI, across experts on social issues and the professionals of language planning.

Apart from the indicators and weightings in SIL about which there is a generalised consensus, the commission of experts made the following recommendations:

The indicators have to put emphasis the interests of the general public, not just satisfy the experts’ interests and priorities. That is, ordinary folk have to be able to see that the linguistic indicators are useful. To that end we suggest that there should be a qualitative analysis with a focus group who would evaluate the proposed indicators put forward by the experts. Elements have to be found that connect the indicators put forward by the experts and those which are perceived as necessary by ordinary citizens.

A distinction should be made between a SIL which evaluates language policy and a SIL which provides information on the sociolinguistic situation and measures knowledge, provision and language use. Both systems are possible, but they have different objectives and methods.

Another positive aspect is the fact of an institution like the Directorate General for Language Policy proposing a public system of linguistic indicators, and thus enhancing its credibility and prestige. It would be very desirable for the political authorities to take stock periodically of the evaluation of the policies in their particular ambits.

It is proposed to unify protocols of questionnaires on language use and language identity. For example. "Read the answers / don't read them". It is also proposed to use the same linguistic protocol in questionnaires (whether statistical or not) promoted by the Generalitat and other administrations. This would give us access to sociolinguistic data at a low cost. This unification would affect both the sociolinguistic studies sponsored by the Generalitat and other research that might contain one or more questions on language. Below, we give exemplification in the shape of 4 studies that ask questions on language identity in different ways, with the result that the data obtained are not strictly comparable. See tables 2 and 3.

Year	Source	Question asked
1993	CIS	What is your main language?
1997, 1999	DYM	Which language do you habitually use?
2000	Fabà et al.	Which do you consider your language?
2002	IN3-UOC	What is your language?

	1993	1997	1999	2000	2002
Catalan	50	49,1	49,6	51,4	40,1
Spanish	49	50,2	49,9	37,3	43,2
Both	1			8,1	14,2
Others		0,7	0,5	3	2,2

There is seen to be a need for data on interpersonal usage. And under the heading of language use we can distinguish various types of use: "listening, speaking, reading and writing", in the same way as we can in language knowledge (understand, know how to speak, know how to read, know how to write).

Regarding the usefulness of the indicators it will be remembered that within the processes of social intervention there are three types of elements that form a pyramid consisting of social perceptions, arguments and evidence. In this pyramid, the indicators take their place as evidence. Good indicators can help to maintain a worthwhile debate and can in the long run modify social perceptions.

It is observed that in the case of institutions with a social function synthetic indicators in use can be iceberg-like in grouping many other indicators or aspects, as well as being easily interpreted or understood. The system of indicators will have a visible part (the tip of the iceberg) with public indicators on the "situation of the language" and a non-visible part with different indicators on evaluation and management of policies. Example: "When the wild salmon can live in the river again" is a synthetic indicator of sustainable environment implying or subsuming other ecological indicators.

It was judged useful to have data on the subjective perception on the situation of the language which speaker may have and which would complement objective data.