I was thinking about the diminution of skills the other day. Like the half-life of an isotope, our ability to do a thing well will fade over time as we do it less frequently. At the time, I was thinking about foreign language skills. But it applies just as well to legal research and legal information.

I’ve picked up a smattering of languages over the years. Enough to be able to get by in French or German as a tourist, or to embarrass my kids in a foreign country. But proficiency drops quickly with disuse. One of our kids is starting a new language, and so I had language retention on the mind.

It applies to any skill set, though. Complexity matters: bike-riding comes to mind, although the ability to bike at any time of life may oversimplify what biking can be to a racer. One of the challenges for law library directors is that you can end up in a role where you’re no longer doing regular research. It varies with job context but it makes sense. As you transition from one skill set – researcher – to another – manager, leader – the performance expectations may change.

A simple visual to explain my thinking on how, when you pick up a new knowledge domain, you may experience loss in an existing skillset. There is no reason the old skillset would entirely disappear, though it may be rusty.

But I wonder if the performance perceptions change with it. My own experience suggests that people expect me to still be a crack legal researcher because, well, law librarian. And while it’s true that I can still find information in a pinch, I know that it’s not expertise. I recall the vocabulary, and can put together the sentences, but it’s not fluent any longer.

I was not at all surprised that someone had already coined the idea of the half life of knowledge. It also seems to be called the depreciation of knowledge. As the Wikipedia page notes, knowledge is not like krypton. The half-life of an isotope doesn’t mean the end of the isotope. And the half-life of knowledge need not mean exponential decay. But it’s a helpful analogy.

Legal Research

It made me think about how we treat legal research. My perception is that it is a bit like playing Pooh Sticks (not playing THE Pooh Sticks). We set the new researcher into the water and give them a push. Once that 1L or self-represented litigant or whatever floats away, they are carried on without our help. Whether they end up in the weeds or in a fast flowing current may have little to do with us.

One perspective I’ve posted on in relation to this was about how we can continue to impact the legal research skill set. Law libraries and others offer continued training and continuing legal education to support ongoing literacy. Our reach limits our ability to lengthen the half-life, though.

We know that legal research takes up just a portion of most lawyers’ daily work. New associates and articling students may spend more time on it. It also seems common that in many legal research contexts, the skillset is exercised by people who cost the client less than the lawyer handling the client matter.

The skills developed as a 1L, and which are integral to becoming a lawyer, have already started to diminish by the time a new lawyer starts practicing. We and others can attempt to slow the decline but it’s the doing – like speaking a language in context – that counts.

Ongoing refreshers – CLE/CPD, boot camps – can offset loss but cannot offset diminution of skillset or knowledge domain. It will maintain a baseline, not an expertise, in the skillset

If the legal researcher ends up in a learning organization, then I would expect that to staunch the decline as well. A place where it is safe to ask questions, to risk showing a lack of knowledge and to make mistakes so as to learn, to take time to stay current on emerging tools. I think that environment can’t be seen as a given for lawyers.

I was interested in this outcome from an OECD report out of Poland on learning at work, and the contrast between training modules (like CLE) and “on-the-job” expertise:

A learner-friendly environment in which learning at work would be stimulated and recognised would need to include training offers that combine formal (theoretical and vocational) learning and informal “on-the-job” learning and lead to formalised certification at the end of training modules. Certifying experience alone would, however, not be sufficient. The decreasing rates of return to experience on the labour market show that the rates of depreciation of knowledge are far too high to be compensated by increased productivity through learning by repetition.

Thematic Review on Adult Learning – Poland – Country Note, p. 26, March 2005

As the legal research skill set suffers attrition, I suppose one goal could be to lengthen it until it is overtaken by other professional skill sets. At some point, a law firm lawyer may be able to offload that skillset (perhaps that’s what’s already happening) to other staff as they focus on being a rainmaker or practice area lead. Or a solo lawyer may become more confident in her knowledge domain so that research is less important to success.

To the extent that lawyers are rarely corrected for professional misconduct due to a failure to do legal research, I expect the need to maintain a legal research skillset is not prioritized over other areas. You can perform at entry level and be considered competent.

I doubt someone has already measured the lawyer and the half-life of their legal research knowledge. I expect it would vary: not all legal research skills would have a half-life of, say, ten years like engineering or 3.3 years in some areas of psychology. Legal research requires a combination of understanding and tools. A lawyer whose knowledge was created in a wholly print world might have seen that skillset plummet as electronic resources came into being. When a major legal publisher pivots their interface, I expect this has a more dramatic impact on the skillset.

Legal Information

The change in law library information is probably the most talked about change in our context. “Everything is on the internet” is a rallying cry heard far and wide by law library funders. It remains a perception without a basis in fact.

A screenshot of a Google search showing suggested matches for “everything is on the internet”

And the people who talk about the half-life of knowledge or facts also talk about libraries:

Librarians were some of the first people to do this. In the 1970s people started looking around and noticing that scientific knowledge was growing very rapidly, but papers had not been digitised yet, and libraries were finite in size and had finite resources. And so librarians had to grapple with the question what to carry on their shelves. They had to calculate which fields get overturned really rapidly, in other words, which papers and books people were unlikely to care about in the future

Q&A with Samuel Arbesman: the Half-Life of Facts, The Economist 11/28/2012

This leads directly to law library collection decisions. Your context will inform your discussion, but the basic point is: how much do we need to collect, and then keep, to enable our researchers to be successful. Courthouse law libraries may only carry a decade of collected materials. A larger library may collect longer. A law school university may play an archival role, reaching back to the year dot.

Library collections frequently represent a funnel, with a wider top (current) and a narrow tip (historic). As a law library, we have a bulge in the year Canada was created.

The Canadian Legal Information Institute commissioned some research into citation longevity, which is useful. Ironically, the link in the blog post to the full report is broken, but you can get it on the Wayback Machine or reprinted by the author in the Journal of Open Access to Law. One finding the report reiterated was this from a 2010 Illinois law journal article:

[C]ase citation networks are scale-free, which is to say that a very small number of cases receive the most citations, and most other cases are cited infrequently or not at allCitation Analysis of Canadian Case Law, 2013 JOAL 1, 8. (citing to “Citations in the U.S. Supreme Court: an Empirical Study of Their Use and Significance”, 2010 Ill. Law Rev. 489, 523.

This concept has stuck with me over the years. Law libraries pride themselves on comprehensiveness – and argue over the quality of resources based on that – but we also know that not every object has equal value to legal researchers. That’s easy to do with print – we buy one text, we skip another – but we don’t have that granularity in digital objects. We buy the collection the publisher offers, not the one we might want.

When we make purchasing decisions about databases, we often contrast publisher A against publisher B. We try to line their content offerings up and, as I said to another law librarian yesterday, determine which one leaves the fewest gaps. There are always gaps.

Perhaps a better approach would be to come at it horizontally rather than vertically. It would require the publishers to rethink how they make their content available. But if I know that most case law usage is recent, why shouldn’t I be able to license just the last 30 years of decisions? I’m going to have gaps in my collection no matter what, so why not base the gaps on object age?

The same thing with texts. At some point, are we literally paying just for the updates because the competent practitioner already knows the text? They – and we – just need to know what’s changed since last access?

I don’t foresee a time when publishers would slice their data this way. An exception to this is Fastcase, which supports the Public Library of Law. The PLOL only goes back to 1999 but it emphasizes breadth over depth. Google Scholar‘s case law database is similar but less transparent.

We’ve accepted that with law journals. When I started in Canada, law journal access in the two primary commercial publishers was from 1970. HeinOnline changed that but I expect there is a similar pareto principle operating within their collection. It’s my recollection that other knowledge domains (like medicine) have similar cut offs on digital journals (I seem to remember they stop(ped) in the 1950s).

In a sense, law libraries are struggling against the natural decline in value of the outputs of the judicial and legislative systems. It’s not that much of a struggle, because technology and the money we pay publishers allows us to breathe longevity into the objects. And I don’t know that it should conjure a sense of blame: funder and researcher perceptions may create pressure to retain content long past its usefulness.

All of this gets me to a final thought: are we struggling to maintain something that doesn’t need to be maintained? If there is a natural decline in skill sets, and no apparent negative impact from the decline, could we rethink how we use our resources?

It means taking greater risk in collection development, accepting larger gaps to see what the impact is (on older content). It could mean that our inter-reliance on law libraries to maintain older collections falls apart.

It could mean rethinking our approaches to legal research training outside the 1L stream. If there is a need to lengthen the half-life, an inevitable decline, how do we help with that? Less to keep it aloft but more to keep it alive while the lawyer gets seasoning.

I haven’t come to the end of my thinking on this. As we face budget cuts and staff reductions – or at least a context where law libraries are not, as a whole, growing – I think we’ll need to spend more time rethinking what we do.