Who, investigating English morphology, if not word formation more generally, does not owe some, and often a large debt, to Hans Marchand's The Categories and Types of Present-day English Word-formation (first published in 1960, the second revised and enlarged edition usually referred to published in 1969)? As explained by Pennanen (Reference Pennanen1971: 9), it took in the study of ‘the vast subject of word-formation … the decisive step from the traditional diachronic-comparative to a synchronic descriptive treatment’, and, according to Štekauer (Reference Štekauer2000: 29), derived its importance from two key characteristics; ‘[f]irst, it delimited word-formation as a linguistic discipline of its own, and secondly, … drawing on the structuralist traditions …, it provided an account of many fundamental problems of word-formation in English in a systematic and comprehensive way’. Here, now, is its successor, and more. The three authors of this new The Oxford Reference Guide to English Morphology (henceforth: ORGEM), Laurie Bauer of Victoria University of Wellington in New Zealand, Rochelle Lieber of the University of New Hampshire, and Ingo Plag of Heinrich-Heine-Universität Düsseldorf in Germany, merrily smiling at the reader from the book's dustjacket, observe in their earliest sentences that ‘there is no single-volume work that focuses on the current state of English morphology’ and that therefore their aim will be ‘to fill that gap by offering a comprehensive descriptive overview of English morphology’. Acknowledging their predecessor, they intend to do more than simply revisit Marchand's territory. His ‘milestone monograph … [w]ritten over forty years ago, [is] inevitably dated: we would expect that there is much that is new to say about the state of English morphology in the second decade of the twenty-first century’. They aim at reaching these goals by invoking modern technological tools because ‘our ability to find an manipulate data has improved considerably with the development of vast computerized corpora’ (3). Ambitiously, they extend the descriptive overview with a discussion of the theoretical challenges that are posed by the morphological phenomena described (something Marchand was unable to do, of course). Compared to Marchand and other work, such as Adams (Reference Adams2001), inflection is newly included. Different varieties of English are addressed. Some areas are excluded, a ‘comprehensive overview’ is not an exhaustive one. The authors focus ‘on structure and use, to the exclusion of various other interesting topics (e.g. processing or acquisition)’ (4). Not included either are verb–particle constructions, which the authors consider syntactic rather than morphological, cases such as forget-me-not, which they consider ‘lexicalized phrases’ (so?), technical and scientific ‘bits’ such as the ending of the word glucose, and affixes when they ‘are clearly no longer available to native speakers’ (4), i.e, truly unproductive ones, although the term is avoid here.

ORGEM has six parts, helpfully summarized also in the first chapter. Part I (3–57) contains discussions of core terminology, methodology underlying the collecting and presentation of the data, and orthographic practices. Parts II–V (61–582) descriptively address, respectively, inflection, derivation, compounding and the interaction of morphological processes in English. Its nineteen chapters include formal semantic considerations, conversion as different from affixation, back-formation, clippings, blends and acronyms, the phonological processes recurring in English derivation, locatives and evaluatives, affix ordering (‘stacking’), the formal properties and semantic interpretation of compounds, and paradigms. Part VI (533–642) is the admirably ambitious ‘Themes’, and ‘is devoted to some of the many areas of theoretical concern that have been raised by the enormous amount of research on English morphology in recent decades’ showing where ‘the descriptive data we have set out in sections II–V bears on them’ from the vantage point of no particular theoretical stance (5). References and indexes (643–691) close the book.

ORGEM without a shadow of a doubt fulfils many of the expectations one may have about it, especially when accepting the authors' aims and ambitions. Once begun reading, it will be unputdownable for considerable time to anyone interested in ‘words’, how they yield to and defy morphological analysis, and how their incredible range of properties can be classified and commented upon by dyed-in-the-wool morphologists, employing ‘modern’ ways of data collecting (digital corpora, lexical databases, ‘opportunistic scaveng[ing]’ (43)). Words as illustrations of some idea, generalization or principle, as counterexamples to some proposal, as intriguing, surprising, awkward, creative or odd pieces of language. ORGEM is mostly admirably well written; in any case it being written by three different authors with as many different backgrounds is hardly ever an issue. Presentational style combines matter-of-factness, well-informedness and occasional good humour. An amazing amount of the literature is summarized and sifted, which the authors combined have an enviable command of. Based on pertinent data, theoretical proposals of the past few decades are dissected and often found wanting. The book ends with a brief assessment of where morphological investigation, as founded on the study of English, now stands, and provides a small catalogue of ideas on how to proceed. For both of these latter issues, see further below.

Even so, inevitably, ORGEM shows a number of flaws. A surprise, especially in view of the authors' clarity about their aims, is the absence of a characterization of the target audience. Fair enough, ‘[t]his book is intended to be used as a reference work’ and readers can ‘dip in’ or ‘read the book through systematically’ (6), but who are they, in the authors' minds? My hardcore theory colleague next door? The Shakespeare expert next door to him? My daughter's university language teacher? My home neighbour theologian with a lay but keen interest in the intricacies and (to him) oddities of the English language? This is a slightly disconcerting and unexpected omission in a work that purposefully aims to get beyond description to ‘theory’, implying specific expository characteristics and responsibilities. ORGEM's second chapter, on terminology, no fewer than 33 pages long, makes this gap immediately felt. This chapter is not for the faint-hearted. Non-specialists, who would certainly find their intellectual lives improved by consulting this book, will be well advised to stay away from it. Lexeme, lexical item, words of several varieties, morpheme, morph, allomorph, base, root, bound root, obligatorily bound root – term after term is introduced and showcased, the list seems endless. Even when this chapter is considered an inescapable prerequisite to the further exposition, it seems ill-advised to (not just end but repeatedly) start paragraphs using qualifications such as ‘[l]ike so many fundamental notions in linguistics, “word” has resisted definition’ and ‘the elements that make up words … is an area where there is a great deal of controversy’, interspersed with ‘it is not always clear how to apply’, ‘[h]owever, what is controversial is the status of’, ‘[a] third vexed case’, ‘[a]nother difficult case’, ‘another problematic case’, ‘is notoriously fraught with difficulty’, ‘[h]owever useful these criteria may be, there are problems with them’ (non-exhaustive examples taken from just pages 7–17). And who is served, in the midst of all this, by the condescending remark ‘[a]n error often made by beginning students is to confuse …’? For specialists from other linguistic fields, this chapter in this form is likely to confirm their worst impressions about ‘the state of morphology’, at an early stage of reading. Bread-and-butter morphologists will leaf through it, wondering why its contents was not condensed into a glossary, hoping to find the ‘problems’ and ‘vexed cases’ lucidly treated as ‘challenges’ later in the book, in which they will then not be disappointed.

For the sake of brevity, substantial topics allowing comment will be subsumed here under the three headings of interfaces, typology, and productivity, branching out occasionally towards issues of theoretical depth, the role of argumentation based on ‘processing’, and of ‘storage vs. computation’. Discussion of these issues will hopefully contribute in a small way to further assessing the current ‘state of English morphology’.

We begin with interfaces. Lexical Phonology and Morphology (Siegel Reference Siegel1974, Kiparsky e.g. Reference Kiparsky1982) is a theory of how morphology interacts with phonology ‘in the lexicon’. It hypothesizes a layered lexicon, where ‘stratally’ specified suffixes combine with stems to form words. The sum total of the output represents the word-knowledge of a speaker. The theory implies that language-specific morphological operations consistently fall into as many classes as there are levels, and in its classical form makes two predictions: that in complex words ‘linear’ affix-order reflects that stratification, and that specific phonological properties can be consistently associated with each level, prosodic properties prominent among them. Work showing that the first prediction is wrong (impeding the study of the second) includes, with the mildest possible conclusions, Aronoff (Reference Aronoff1976) and, more fully damagingly, Fabb (Reference Fabb1988), the latter showing that affixes' independently necessary selectional restrictions go a long way towards descriptive success. This issue is discussed elaborately in ORGEM's Chapter 21 fully crediting Fabb, but unnecessarily digressing into a summary of one of the authors' own attempts at providing an account based on ‘processing’ (pace the book's introduction), which, though self-assessed to be ‘very explanatory’, does not seem to fully work, eventually confirming Fabb's approach. Chapter 27, ‘The nature of stratification’, in the book's Part VI, revives the issue, arguing that Lexical Phonology and Morphology's ‘theoretical claims have largely been based on an incomplete picture’ (383), refraining from crediting Fabb any longer but providing a full 26 pages of affixal combinations data, leading to the conclusions that ‘English derivational morphology actually displays very little in the way of stratification’ and that ‘affixes in English … are far more integrated than might at first be imagined’, but admitting that ‘the classic theory of Lexical Phonology and Morphology has been largely discredited, and therefore our observations might be deemed unnecessary nails in an already sealed coffin’ (610–614). Here restraint might have benefitted the authors given that most of the affixal combinations data are already published in Plag & Baayen (Reference Plag and Baayen2009).

A central tenet of Lexical Phonology and Morphology is the application of phonological rules along morphologically ‘layered’ structure. Phenomena such as stress assignment to a complex word when one of an affix's selectional restrictions is prosodic, and the phonological effects typical of (non-native) affixation in English, such as the avoidance of ‘stress clash’, are dealt with in brief subsections throughout ORGEM, but the deeper theoretical issues of the ‘cyclic’ application of the rules and the ‘unmarkedness’ of output structure (Brame Reference Brame1974, Goldsmith Reference Goldsmith1990) are never opened up. Discussion of theoretical claims examined and called into question on empirical grounds, and often quite convincingly so, tend toward the shallow, shying away from deeper ‘explanation’ which some theorists were or are trying to aim at. Another such striking example are the self-embedding anti-missile (anti-anti-anti-missile-missile-missile, etc.) cases mentioned in Chapter 21 but their presentation is dissociated from their potentially exciting theoretical implications in the area of ‘generative capacity beyond that of finite-state grammars’ (Di Sciullo Reference Di Sciullo, Kiefer, Ladányi and Siptár2012). Discussion of the morphology–syntax interface is limited, it comprises the issue when to call a string a compound or a phrase, given these constructions' semantic and prosodic overlap, and the well-argued and empirically supported possibility of using phrases as the input to affixation. Hard to understand is the omission of verb–particle constructions, beyond the unsubstantiated claim, in the book's introduction, that they are syntactic rather than morphological, and no less hard to understand are the mere two isolated references to ‘clitics’, a term never properly explained.

We next consider typology. Chapter 28, ‘English morphology in a typological perspective’, is ORGEM's most disappointing, a missed opportunity to grasp and convey the excitement that doing morphology can generate. It is argued that large-scale classification using schemes in which a language can be captured as being ‘isolating, agglutinating, fusional, or polysynthetic … do not tell us much that is interesting or useful’ (616). If, the authors say, following Comrie (Reference Comrie1981), among the ‘agglutinative’ languages ‘English would be ranked relatively low on the index of fusion, with few examples where more than one meaning is packed into the same morpheme, and somewhere midway on the index of synthesis, with words – at least multisyllabic ones – frequently having more than one morpheme’ (618), and Turkish ranked even lower for fusion and higher for synthesis, then first, calling English agglutinative at all would leave out ‘significant [non-concatenative] processes [such as] conversion, back-formation, and clipping’, which (as shown) are surprisingly productive in English, and second, ‘we are left with the feeling that this kind of classification does not give any real insight into English in comparison to other languages’ because ‘nothing … in terms of correlated structural properties or historical properties … follows from these classificatory schemes’ (618). But surely beyond scales, empirical illustration of what other languages, or language families, Turkish, Chinese, Semitic, Amerindian, Pacific or Bantu, can do and do in their morphologies, and English cannot, in degrees or in principle, tells us things about English? Why not devote a few pages to the fun of showing that? How does the claim of Chapter 21 that the ‘number of derivational affixes that can be found in a single [English] word seems in principle unlimited, but processing [of] words with five or more derivational affixes becomes increasingly hard’ (507) bear exactly on the fabulously disheartening sequences happily manipulated by speakers of Turkish? Compared to this language's Cekoslovakyalılaştıramadıklarımızdanmışçasına (Wikipedia, June 2014, Agglutination), is its English gloss ‘as if you were one of those whom we could not make resemble the Czechoslovakian people’ just more of the same? Is the Boumaa Fijian doubly reduplicated form ca-ca-lidi-lidi (Dixon Reference Dixon1988) essentially just a variant of English chitchat? Unimpressed, the authors promote a ‘finer-grained approach’, comparing English to Germanic and Romance languages in a handful of morphological characteristics. They argue ‘that typological perspectives on a more local level have the potential to be more illuminating’ (615) to reach the conclusion, bafflingly, that a ‘typological approach of the kind that has been taken here does not necessarily tell us a great deal about the operation of English morphology’ although it ‘raises questions about processes of diachronic development and the way they interact with typological tendencies’ (627). Which questions of any analytical depth are raised (these are ‘theoretical themes’, remember) beyond relatively superficial observations on differences between languages and cases of borrowing remains unexplained. Given the languages compared, morphological gender is one of the most striking absentee phenomena. Intriguing typological areas related to issues of language acquisition, morphological markedness and the like are not just unexplored but left unalluded to.

Finally, we turn to productivity. Rule one of morphological analysis is not to ask an unsuspecting practitioner for her definition of ‘productivity’. While in the main body of the book, because ‘even in simple cases, it can be difficult to determine the productivity of morphological processes’, the authors ‘have restricted [them]selves to rather impressionistic statements of degree of productivity’ (581), they try to make sense of their own findings in Chapter 26, driving home two points. First, productivity must be assessed in terms of ‘domains’, a concept which for affixation approaches that of the selectional restriction. Regarding the notion of ‘competition’ between morphological processes within a single domain, where the application of one process may ‘block’ the application of another (as in the irregular past tense blocking the regular form, swam – *swimmed), they find that ‘[o]ur data shows so much competition in domains of productivity that we cannot find strong support for any notion of blocking’ (581). This conclusion in this strong form is likely to be judged controversial. Secondly, they recommend promising work by Baayen (Reference Baayen1993) who (rather illuminatingly) proposes to adduce productivity from the number of hapaxes in a corpus, taking domains into account. Two disappointments remain, however. The discussion surrounding ‘dual route’ models (Pinker & Prince Reference Pinker, Prince, Sutton, Johnson, Shields, Lima, Corrigan and Iverson1991, Pinker Reference Pinker1999, Nooteboom, Weerman & Wijnen Reference Nooteboom, Weerman and Wijnen2002, and many others), which is simply unavoidable in this context and a potential source of interesting speculation if only on the subject of processing, is half-hearted at best. To not seriously discuss the concept of ‘the default’ seems an omission, and to suggest that the computation (‘rules’) vs. storage (‘lists’) dichotomy is equivalent to productive vs. non-productive is simply wrong. Regarding the productivity of compounding, also left undiscussed, interesting material for both a typological comparison and a methodological exposition, might have been taken from a ‘dual route’ study such as Ricca (Reference Ricca, Scalise and Vogel2010). Although it is the authors' aim (Introduction) to include different varieties of English, their impression is that ‘differences between varieties seem to be rather minor’ (4). Even when taken just to refer to major world variants, this raises eyebrows, moreover it seems to have led to the exclusion from OGREM's database examples taken from pop(ular) culture or less middle-of-the-road language in which ‘creativity’ if not ‘productivity’ (in ‘domains’) is taken to limits. So blues-rock band Free's stealer (usually supposed to be blocked by thief) is not here, nor are Orwell's plusgood, the Sherman Bros.’ supercalifragilisticexpialidocious, and Philip Larkin's immensements.

The authors' proposed list of boundary conditions on morphological investigation in their final few pages fully reveals the bias of the above remarks, grounded as these proposals are in the entire volume if not even more general modes of research, rather than purposely selected chapters. Hence the list bears summarizing (640–641). Although morphology is ‘highly computational’ and ‘[r]ules of morphology must minimally allow for the generation of an infinite number of new words …, [m]orphology is not syntax or phonology’; these three must be kept separate, also formally (although no clarification is added as to nature of such ‘formal’ differences). ‘Storage and computation are not mutually exclusive’, even low-frequency forms can be stored. ‘Synonymy is rampant in word-formation’, which subsumes the conclusion ‘that blocking is at best a tendency and at worst a myth’. Models of inflection must be able to deal with probabilistic variation in the expression of morphosyntactic categories. Allowing that productivity is gradient, morphological rules must be violable when the need arises. Beyond being ‘rule-governed’, the lexicon can be subject to paradigmatic pressure, with phonological effects as in internal stress preservation pairs such as còndènsátion vs. compensation, and morphological results in coinages such as doubleton. This is a point of exceeding interest, leading to the intuition, finally (642), that ‘we might think about diverting our attention from the question of how forms get generated and focus instead on how well potential outputs conform to canonical patters, rules, sets of morphologically related words, or constraints’.

There are obvious points that can be taken issue with in this list (again, for instance, where is language acquisition?), just as there are for other passages of this book, as indicated. But this is due to the authors' aiming very high, and they must be unreservedly applauded for their ambition to undertake the enterprise, and their preparedness to present the result in its current form. Highly recommended.