The proposal that newspapers which are members of the National Newspapers of Ireland (NNI) should appoint readers' representatives was made first by two lawyers, consultants to the NNI, who were interested in both press freedom and in individual human rights. They proposed, in a report to the NNI in 1988, that if each newspaper were to appoint its own readers' representative this might be one way of reducing the very costly libel cases that were being taken against papers at the time.

The cost of defending The Irish Times against libel cases was running at close to £1 million a year, so any reduction was worth attempting, especially since some readers saw a libel action as an easy way of making large sums of money. One greedy reader who thought he had been libelled actually got his lawyers to write to this newspaper demanding a cash payment to cover the damages he claimed, meanwhile telling us not to bother publishing an apology to restore the imagined damage to his reputation. Perhaps understandably, his lawyers never took the case to court so he got no money anyway.

In fact, there was no evidence to show that, after the post of readers' representative was established in The Irish Times in 1989, there was any significant decline in the number of libel cases being taken against the paper. Maybe many of the cases, like that of the man who wanted money but did not want a correction or an apology, were driven by a desire for profit and not by any real concern for an individual's reputation after all.

However, we did find that making a corrections and clarifications column available to readers (and also to journalists who might make inadvertent errors in their reports) helped to increase the credibility of the newspaper in the eyes of its readers. It became the policy that, where errors occurred, they would be corrected or clarified (under the responsibilities given to the new readers' representative). This would be done as quickly as possible and to the satisfaction of both the readers' representative, who would have to ascertain the facts of the case, and the person who had pointed out the error in the first place.

This sounds easy and logical, although the task of checking the facts could sometimes take longer than the original report had taken to prepare and write, especially if the facts were unclear or difficult to ascertain. It would reassure the readers that, if no correction appeared, the original report was factually correct, and so it transpired.

There was considerable and very gratifying feedback from readers to this effect. Far from feeling that the publication of corrections would indicate that the paper kept on making mistakes, the readers found it reassuring that errors were corrected as a matter of policy, under the aegis of an independent arbitrator.

But was the readers' representative really independent? He was, after all, a senior member of the editor's own staff and newspaper editors are generally regarded in their offices as God: omnipotent in their own hectic domain. Would it be possible for the readers' representative to stand up to the editor in a situation of potential conflict, and even to over-rule the editor if that became necessary in the matter of a correction?

As it turned out in this office, these questions never arose, except in theory. The editor was as fully committed to the policy as was the readers' representative and no conflict ever arose. Indeed, the facts spoke for themselves in the vast majority of cases. If the readers' representative had done his checking adequately, it was quite clear whether there had been, or had not been, an error, and the policy had been accepted by everyone that all errors would be corrected.

So, what were the kinds of errors that had to be corrected? Well, not spelling mistakes - unless the misspelling had altered the meaning of the sentence in which it occurred, or it had been in somebody's name and had upset that person. And (though far too many complaining readers never seemed to accept this) opinions published in the paper cannot and will not be "corrected". The best means of expressing contrary opinions is to write a letter to the editor for publication.

The well-intentioned pedants were often very difficult to deal with, complaining about an apparent misuse of words or phrases, but adamantly refusing to accept that the meanings of words and phrases change with time and usage. Often, even if they had a technical case to make for their interpretation of the words or phrases, they did not win a correction or clarification.

Most complainants, however, were calling to draw the attention of the readers' representative to what they saw as an error in the previous day's, or the previous week's, editions. They were dealt with simply enough: if they were right, a correction was published; if wrong, it was not.

Sometimes, the callers were strikingly obsessive about the subject of their desired corrections, and they often took time to deal with. Sometimes they were clearly and obsessively misinformed and they took longer. Occasionally, they were just lonely and looking for someone to talk to and, if one had the time, one listened. No further action was usually required.

Some callers adopted an almost proprietorial attitude: how could their newspaper do this to them and how was their newspaper going to remedy the situation? Many of these were regular callers and it was good to know that we had a significantly regular readership which was genuinely and actively concerned with what we were doing.

Sometimes, a long and busy day with nothing but negative comments coming in could become dispiriting, even depressing. But the occasional surprise of a reader calling in to praise the newspaper for some of its coverage of the news or its expressions of opinion (not usually a species included in the readers' representative's terms of reference) could greatly lighten the step and raise the spirits of that reader's representative - at least for a while during a busy day.