Studies In Second Language Acquisition
Studies in Second LAAdditional services for Studiesin SecondLanguage Acquisition:Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click hereOn determining developmental stages in naturalsecond language acquisitionJürgen M. Meisel, Harald Clahsen and Manfred PienemannStudies in Second Language Acquisition / Volume 3 / Issue 02 / March 1981, pp 109 135DOI: 10.1017/S0272263100004137, Published online: 07 November 2008Link to this article: http://journals.cambridge.org/abstract S0272263100004137How to cite this article:Jürgen M. Meisel, Harald Clahsen and Manfred Pienemann (1981). Ondetermining developmental stages in natural second language acquisition.Studies in Second Language Acquisition,3, pp 109 135 doi:10.1017/S0272263100004137Request Permissions : Click hereDownloaded from http://journals.cambridge.org/SLA, IP address: 132.206.27.25 on 01 Sep 2012
On DeterminingDevelopmental Stages inNatural Second LanguageAcquisitionJiirgen M. MeiselUniversitdt HamburgHarald Clahsen and Manfred PienemannUniveisitdt Wuppertal1. IntroductionResearch on Second Language (L2) Acquisition, over the past ten years, has undergone substantial changes by shifting its focus of interest away from an analysis oflinguistic structures alone, concentrating more on the learner himself or, rather, onthe process of learning. It had become obvious that one of the major shortcomingsin contrastive studies as well as in the usual kind of error analysis is that they lackthorough investigation of factors which determine the kind of approach a learnermay take to acquire a second language. This again implies that it is more fruitful tostudy the process of learning itself instead of merely analysing its outputs. It is bynow widely accepted that the learner takes an active part in the learning processand does not merely get trapped in structural gaps which linguists may find whencomparing the source language (the learner's LI) and the target language (L2).A well-documented example of this kind of problem is the case of structuralinterference from LI into L2. Take for example an error one can find in the speech ofSpanish or Italian learners of German: they sometimes use warum "why" whenstandard norms require weil "because." This is quite generally interpreted as aninterference from L1 since Spanish and Italian do not discriminate morphologicallybetweenpor que (Sp.), perche (It.) "why" andporque (Sp.), perche (It.) "because."However, Felix (1977:245/6) reports the following dialogue with a 7-year old learnerof German:F: wollen wir zu Julie gehen?let's go to JulieG: neinnoF: warum nicht?why not?G: warum die ist nicht hierwhy she is not hereThere are similar uses of warum instead of weil by the same boy. What is interesting about this error is that the learner's first language is English; had it been Spanishor Italian it would have been impossible to refute the interference hypothesis. Inother words, since in this case we cannot account for the use of warum, meaning"why" as well as "because," by referring to morphological identity in L1, we shouldlook for an operation the learner might have performed when constructing andusing the intermediate system which is available to him on his way to L2.The choice of this example does not necessarily imply that we deny the possibilityof interference in L2 acquisition. But it can illustrate the fact that many phenom109
110Studies in Second Language Acquisition, 3, No. 2, 1981ena studied in L2 acquisition—like interference—cannot be discussed in terms ofdifferences between linguistic structures (nor by merely classifying an error).Rather, the learning of a language must be viewed as a very complex process ofwhich the development of a grammatical system is only one part; properties(syntactic, phonological, etc.) of L1 and L2 certainly do have some influence on thisprocess and may account for some aspects of the learner's interlanguage. But again:other factors, especially psychological ones, are likely to be of much greater importance for our understanding of the process of L2 acquisition, including linguisticand non-linguistic strategies involved. This point gains further support by theobservation that there exist certain similarities between LI acquisition and L2acquisition. To return briefly to our wanim example, one should take into accountthat the same and similar phenomena appear in LI acquisition. Bailey (1972:3)gives the following conversation with a 5-year old boy, David:I came home and found the ice cream was all gone.(David:) That's why Markey ate it all up.Whatever reason one may find for the use of [that's) why replacing because (we donot want to speculate on this), it will at least be a good candidate for a betterexplanation of the above mentioned error by Spanish and Italian learners of GerSuch an approach to the investigation of L2 acquisition—one which describes it as asequence of approximative systems partly determined by rather general operationsand strategies, some of which may even be universal—almost inevitably leads tothe assumption that there exist developmental sequences which can be defined bythe appearance in a strict order of certain linguistic features. This again implies thata learner acquires part of the rules in an ordered sequence—as long as his learningprocess is not stopped by fossilization or by some other force. This definition allowsfor the possibility that a rule is not acquired at all; if, however, it is learned, thishypothesis predicts when it will appear as compared to other rules ordered in thissequence. It is furthermore assumed by many authors that this order is normallythe same for each individual learning a second language, more or less independently of the learner's first language.In what follows we want to discuss the notion of 'developmental stage': how it isdefined and what kind of data constitutes evidence for the appearance of a newstage in the acquisition of L2. We will argue that one has to distinguish betweendifferent groups of learners who may follow different paths on their way to theirvariety of the target language. This does not exclude the possibility that there arestages in the linguistic development of L2 learners which may be defined in termsof rules emerging in a predictable order. However, one will also have to allow for thepossibility that within each such stage there is considerable variation. This variation, we suggest, characterizes the different learner groups linguistically; whethera learner belongs to one such group or another depends on socio-psychologicalfactors.Our arguments are based on data taken from informal interviews with adultSpanish, Italian, and Portuguese women and men working in Germany who did not
Stages in Second Language Acquisition IMiesel, Clahsen, Pienemann111receive formal instruction in German.' These conversations with 45 foreign workers were recorded, transcribed and analyzed by the members of our research group,Zweitsprachenerwerb italienischer und spanischer Arbeiter (ZISA).2 In the following we will only deal with the acquisition of syntax although the research project isalso concerned with problems of the lexicon.2. Determining Developmental Stages2.0 Leaving aside those studies which only speculate on what may constitutedevelopmental stages in second language acquisition without much empiricalbasis to support these claims, one can—roughly speaking—distinguish two types ofapproaches. The first consists of cross-sectional studies with—generally—a fairlylarge number of learners analyzed; the second comprises the few existing longitudinal studies investigating the acquisition of a second language by a very limitednumber of learners. We believe, however, that both share one important drawback:very often variation is indiscriminately interpreted as indicating a newdevelopmental stage. This criticism will have to be explained in detail.2.1 Longitudinal StudiesAs various authors have pointed out, longitudinal studies—designed according tothe methods used in LI acquisition—seem to be quite adequate for the study of L2acquisition, although L2 acquisition may develop less uniformly to the extent thatinterpersonal similarities in LI acquisition can be explained as a result of biologicalmaturation and cognitive development, whereas socio-psychological factors exerta stronger influence on the acquisition of a second language. There exist, however,only very few examples of this kind of research. One example of a study which doesdescribe the linguistic performance of a small number of people longitudinally isthe one by Cazden, et al. (1975). To our knowledge, there exist no other reports onlongitudinal studies of natural L2 acquisition by adults. 3 Although this kind ofresearch has led to a number of widely accepted hypotheses about developmentalsequences, it does not solve the problem of defining developmental sequences.A longitudinal study gives a description of linguistic performance at several pointsof time, analyzing changes which occur between point t[ and point t1 1; if a linguistic feature appears systematically it is said to have been acquired by the learner.Developmental stages may then be defined in terms of acquired rules. If this is anadequate account of what makes up a longitudinal study, there are at least threemajor problems: time, structure, interpretation of changes. One would of courselike to have only the latter as a free variable and the first two as bound. Unfortunately this is much less straightforward than one might believe at first sight.We only want to mention briefly the last point: given a certain linguistic structure,when can we say that it has been acquired? A single occurrence would obviouslynot indicate that it has been acquired in the sense that the person who used it couldrepeat it every time a similar context occurs. This seems to indicate that we have torely on some kind of quantification of the data. Brown (1973:258), followingCazden (1968), established as a criterion for morpheme acquisition (LI) that it mustbe used correctly in 90% of the obligatory contexts of three successive recordings;
112Studies in Second Language Acquisition, 3, No. 2, 1981in each recording there should at least be five obligatory contexts. However, asCazden et al. (1975:27) remarked, limitations of data sometimes make it impossible to apply this criterion in L2 acquisition:Often, we did not find five obligatory contexts even though the subject wasperforming 100% correctly on a particular linguistic feature.Brown's criterion may be more useful in LI than in L2 research (Brown 1973); also,it seems to work better with morpheme acquisition than with word order problems.A more serious problem with Brown's criterion is that—if applied as statedabove—it ignores the insights provided by interlanguage hypothesis, since onlystandard-like usage is counted. Schumann (1975:35-53), for example, analyzesAlberto's (his learner's) speech in this way and comes to the conclusion that noauxiliaries (including do-support, copula, modals) can be said to be acquired, wherewe think one should conclude that Alberto has acquired a rather elaborate auxiliarysystem; but he doesn't use the correct form, e.g. supplying is where second personsingular are would have been expected. Even then, Alberto does use the copula is,correctly fulfilling the requirements of Brown's criterion. In this case, Schumann aswell as Cazden et al. (1975) argue that this is the result of positive transfer fromSpanish.4 Similarly, Alberto is said not to have acquired past tense endings (30-33).Although this is undoubtedly the case, an interesting question might be whether hehas developed means to express past tense or not. From the data quoted bySchumann, it seems as if Alberto only deletes the endings when there is anadverbial like yesterday, after . . . three years, etc. This would be in accordancewith some of our own findings and shouldn't come as a surprise to a creolist, sincesimilar constructions may be found in pidgins and Creole languages.All this seems to indicate that Brown's criterion of acquisition is much less usefulin L2 studies than it may be for LI research.5 We believe that the most satisfyingresults are obtained by quantifying all features under consideration in the samefashion, indicating the number of actual occurrences relative to the total number ofpossible occurrences. Cazden et al. (1975) followed this procedure when studyingnegation and the interrogative. They showed that it allows for a description oflinguistic development as well as of variation in the development, for standard-likestructures as well as for features which deviate from the standard norm.In this section we have argued that although we think that only longitudinalstudies can reveal convincing evidence for developmental stages in language acquisition, such evidence does not automatically follow from an analysis of severalconsecutive samples of speech of the same learner. Regular or irregular intervalsbetween a number of recordings do not guarantee either that a new stage in thelinguistic development has been attained, or that one has not missed such a stage.The process of second language acquisition has to be described primarily in terms oflinguistic features.Even the most sophisticated linguistic examination of the data, however, can onlyreveal that changes did in fact occur. The main task then still remains to be done,namely to interpret these results. This brings us back to our remark at the begin-
Stages in Second Language Acquisition I Miesel, Clahsen, Pienemann113ning of this second section: so far it has always been taken for granted that allchanges except certain idiosyncratic features in the developing system may beinterpreted as an indication of a new step on a learner's path from zero knowledge ofthe target language towards the standard variety. Furthermore, it has been tacitlyassumed that one may safely expect to find the same kind of development in alllearners, at least with adults who share the same first language. This is a rathersimplistic view of language acquisition, but it appears to serve quite generally as aguideline for research.2.2 Cross-Sectional StudiesApart from the points mentioned in 2.1., which are also valid for cross-sectionalstudies, we will show that these shortcomings are aggravated by certain characteristics of cross-sectional studies themselves. The procedure generally followedin such works is to analyze the linguistic performance of a number of L2 learners ata certain point of time. The results are then interpreted in one of two (similar oreven identical) ways: a) either one tries to find out which areas of the grammar ofthe target language present most difficulties for the learner by counting the numberof errors, or b) one attempts to give the learner's distance from the standard norm interms of grammatical complexity. The latter is normally computed on the basis ofthe numbers of rules acquired; this includes the possibility of special interlanguagerules which do not exist in the source and in the target language. In both cases thelearners studied are then plotted on a scale which reaches from close to zero(highest score of errors, or greatest distance from the standard variety) to almoststandard. As a result, different learners, or groups of learners who show comparablelinguistic behavior, are then said to represent different developmental stages. Thus,cross-sectional studies which are intended to determine developmental stages insecond language acquisition are necessarily based on the assumption that theprocess of acquisition is strictly linear and uniform.The first question to be raised is whether one may in fact draw any conclusionsabout the order of acquisition from the number of mistakes. This is the workinghypothesis for several cross-sectional studies although it is not always statedexplicitly. The central question to be discussed here is the following one: does ahigh number of errors in a certain area of grammar really indicate that this structure(or rather, the set of rules which generate it) has been acquired late, and vice versa?First of all, the inherent logic does not appeal to us as imperative. There is absolutely no reason to believe that an L2 learner, especially in a natural setting, shouldalways start with the "easy" parts of the grammar and leave the "most difficult"ones for later. Rather, he uses whatever is necessary to express his communicativeneeds, possibly choosing the least difficult of several alternatives. (This remains anopen question as long as linguists and psycholinguists have great difficulties inexplaining what is simple (Meisel 1980).Therefore, a structure which shows a high risk for errors may be acquired fairlyearly and continued to be used deviantly until very late. On the other hand, somerules, once they are acquired, may be used correctly right from the beginning. Inother words, what is discovered by research like Fathman's, analyzing the linguistic performance of "approximately 500 non-native English-speaking children
114Studies in Second Language Acquisition, 3, No. 2, 1981learning English in public schools in the United States" (ages 6-14) (Fathman1979:31), is not the order of acquisition but rather the relative difficulty of somegrammatical structures of English for foreign learners. It should therefore not comeas a surprise that these findings are in agreement with results reported by otherresearchers for different populations, but working along the same lines asFathman.6It is most likely that it will turn out that the two hierarchies for difficulty and forlearning sequences, are not identical. The few existing longitudinal studies on L2acquisition show that the frequency of a certain error may go up and down. Molony(1977:279), for example, reports "a dramatic increase in verb errors from three and ahalf months (March) to five months, then a steady decline so that by October thereare almost no errors." The explanation for such development may be linguistic (e.g.structural contexts), as well as non-linguistic; and by careful analysis one should beable to find out about this. Cross-sectional studies, however, are not likely to detectthese facts, let alone to find solutions for the problem. Rather, samples taken atpoints of time which correspond to the end of the fifth month and sometime duringthe sixth moiith, respectively, will probably lead to erroneous interpretations,especially if compared to errors in other areas of grammar which show a differentdevelopment than the verbal constructions mentioned. This indicates, as is wellknown from first language acquisition research, that certain structural propertiesof the learner's performance can only be explained if one also knows preceding andfollowing developments. Some features even disappear after a certain period oftime and occasionally reappear later on.These remarks should suffice to illustrate that the process of L2 acquisition cannotsimply be viewed as a system of an increasing number of rules. While some newrules are acquired, others are dropped after a certain period of time (specific interlanguage rules, but possibly also standard rules), and some are changed (e.g. addingmore specific information to their structural description, thus restricting the rangeof possible applications which had been the result of overgeneralizations, as compared with the target variety.) As a consequence, grammatical complexity of onesample of a learner is a useful but not sufficient criterion to determine thedevelopmental stage he/she has reached. Therefore, it does not seem to be admissible to plot learners on a scale, implying that the one with a "less complex"grammatical system necessarily represents an earlier stage through which the oneon a higher grade of the scale should have gone. This leads to another point. Somestudies apparently take it for granted that all parts of the grammar of a learnerdevelop uniformly, i.e. if he/she progresses in one area of his grammatical competence, there must be similar developments in others. If this were not the basicassumption, one could not claim that a mean value for several groups of rulesindicates an overall level of linguistic development. Not only does common logicnot force us to believe in such a uniformity of development, longitudinal studiesclearly show that the contrary is much more likely to happen (Felix 1976, Molony1977). This remark should certainly not come as a surprise, since for many yearslanguage acquisition research has dealt with fossilization, and as far as we knowfossilization may affect different parts of the grammar in different learners. Thismeans that on the basis of cross-sectional studies one cannot decide, on
Stages in Second Language Acquisition I Miesel, Clahsen, Pienemann115principle, whether a certain structure shows 'normal' development, or whetherthis pattern has been fossilized just in the case of the learner under consideration.There is a last point which we only want to mention briefly since it would take uptoo much space to go into the necessary details. Although it is generally acceptedthat a speaker's performance does not in every respect reflect his competence (usingthese highly disputed terms for the sake of the discussion), there is a tendency towrite a rule of grammar for each phenomenon which occurs systematically andregularly. This is especially problematic when dealing with changing systems, as inlanguage acquisition. The linguist cannot even appeal to a "competent speaker"who, by definition, does not exist in this case.In spite of the manifold theoretical and practical difficulties, we think that it isworth the effort to try to sort out what a person's underlying knowledge of thelanguage must look like and how he uses this knowledge. This difference will beimportant as soon as the results of language acquisition studies are put intopractice, e.g. in designing language classes. To give just one example: a tacitassumption shared by many researchers seems to be that a learner's performancereflects the best of his linguistic abilities. This is in conflict, however, with anotherwise equally well-known fact. Namely, someone who wants to communicateefficiently in the L2 cannot at every instance display the best of his/her internalizedgrammar since it would cost too much time and effort to handle certain parts of it,the use of which is not sufficiently habitualized. Instead, the learner will be lookingfor convenient shortcuts.Such shortcuts can be special production strategies. Similar to the strategies discussed by Bever (1970); Clahsen (1979) shows the usefulness of this notion in L2studies. More pertinent to the present discussion, however, is what is generallycalled "simplification" of linguistic structures. Meisel (1980) suggests that one hasto distinguish between at least two kinds of simplification. The first kind, elabolative simplification, serves the purpose of bringing the learner's interim grammarcloser to the target norm. The second kind, restrictive simplification, reduces thegrammar to make it easier for the learner to handle. Stauble (1977) argues that athird kind should be added, conformative simplification, "in which the learnersimplifies his grammar by conforming to the model system and eliminating anynon-standard forms."The second kind of strategy (restrictive simplification), in particular will make thelearner's speech look more deviant from the target variety of the L2 than what itcould actually be. Take, for example, the omission of certain syntactic categories orof morphological material. The learner 'knows' of course that there must normallybe a verb in each sentence and that it carries morphological information; in his/herLI this is quite the same, and one hardly finds a learner who consequently deletesall verbs, and not many always use the infinitive. But deletions of this kind make iteasier to use the grammar; they help to avoid the choice of the right position of theverb and the correct inflectional ending.If our interpretation of this kind of simplification is correct, the linguist whowishes to write a transitional grammar has to try to differentiate between what the
116Studies in Second Language Acquisition, 3, No. 2, 1981learner has not yet acquired and what he/she is not yet able to use correctly in everycontext. Otherwise one cannot achieve psychological plausibility for the grammarwhich seems necessary for an adequate explanation of a learning process, be itlinguistic or other.The problem of this section—to discover what is part of the internalized grammarof a L2 learner—has to be solved by longitudinal studies as well as by crosssectional studies. The former, however, have the advantage of being able to studyone person's speech at different points of time and possibly at different places and indifferent contexts. This may help to discover which strategies are applied by thelearner.We have tried above to show that cross-sectional studies, in addition to difficultiesmentioned in 2.1., have to cope with certain problems which make it ratherunlikely that they could be an adequate way to discover developmental stages in L2acquisition. Depending on methodological choices, some problems become more,others less important. Works which attempt to describe L2 acquisition by writingtransitional grammars have serious difficulties with the point mentioned last. Onthe other hand, they can avoid comparing the learner's speech with the norm of thetarget variety—the latter being hard to define in precise terms anyway. This again isa major shortcoming of studies which measure the learner's achievements bycounting the relative frequency of errors. One risks violating the interlanguagehypothesis completely; at least a difference should be made between a) what hasbeen acquired according to the standard norm, b) progress which does not reach thenorm, c) nothing acquired at all. Dulay and Burt (1973, 1974) try to take this intoaccount:no functor suppliedmisformed functor suppliedcorrect functor supplied 0 (she's dance) 0.5 (she's dances) 1.0(she's dancing)The same procedure is followed by Bailey, Madden, and Krashen (1974). Theseauthors are, furthermore, aware of the fact that difficulty ordering is not identical tothe sequence of acquisition.One obvious possibility of combining the advantage of a cross-sectional study withat least some of the strong points of longitudinal research is to see all or a few of theconsultants a second time after a certain period of time. There is the danger,however, of combining the shortcomings of both methods instead of the advantages, for the number of learners has to be reduced, and at the same time one doesnot get the same number of samples from each person.Hyltenstam (1977), on the other hand, seems to have used this technique quitesuccessfully. He tested 160 subjects on two occasions. At time 1 they had studiedSwedish for three weeks; time 2 was five weeks later, and 160 persons participatedon both occasions. But again, one has to pay a high price. Hyltenstam could notrecord natural conversations, but had to use a highly structured test to be able tohandle this amount of data: learners "were required to choose one of two marked
Stages in Second Language Acquisition I Miesel, Clahsen, Pienemann117spaces in which to place negation" (386). This does, of course, not leave any roomfor a creative use of an interim grammar.To sum up, we believe that the only promising way to find out aboutdevelopmental sequences in natural L2 acquisition—and possibly in languageacquisition research quite generally—is to study the process of language acquisition over a longer period of time. The results should then be backed up by crosssectional studies. This procedure will certainly not eliminate automatically allproblems discussed in this second section of our paper, but it will at least help tocome to grips with them.3. A multi-dimensional model of second language acquisition3.0. In sections 2.1. and 2.2. we have tried to explain our remark that cross-sectionalstudies as well as longitudinal investigations both share one important drawback:researchers tend to interpret all systematic variation as indicating a newdevelopmental stage. Part of our criticism was concerned with methodologicalproblems which will not be pursued any further in this paper. Our main point,however, is that second language acquisition is quite generally viewed as a linearprocess, as a straight line leading from zero to the target variety of L2. Systematicdifferences between the interlanguages of either different learners or of the samelearner at different points of time are indiscriminately said to constitute evidencefor his/her progress along this line toward L2. As far as we can see, for crosssectional studies it is difficult to avoid this uniformity hypothesis about L2 acquisition; this is why we think they cannot furnish conclusive evidence for orderedsequences of rules. Viewing language acquisition as a linear and uniform processdoes not allow one to attribute different degrees of importance to different structural properties of the learner's speech, either on language internal or on social orpsychological grounds. For the social scientist this is a rather implausible assumption in the case of second language acquisition and calls for further investigation.3.1. Works by E. Haugen (1956), Gardner and Lambert (1972), Gardner et al. (1974),Schumann (1975) and others, as well as our own studies (compare footnote 2),indicate that socio-psychological factors play a prominent role in second languageacquisition. Social distance from the target group, intensity of contact, attitudes,motivation, etc. have been shown to be more influential than, e.g. duration of stay,intelligence, etc. If it is widely accepted that such factors do influence secondlanguage acqui
acquisition, although L2 acquisition may develop less uniformly to the extent that interpersonal similarities in LI acquisition can be explained as a result of biological maturation and cognitive development, whereas socio-psychological factors exert a stronger influence on the acquisition of a second language. There exist, however,
The influence factors of second language acquisition . The main factors that affect second language acquisition. Individual factors of learners . The main process of second language acquisition includes perception, understanding, consolidation and application is very important for the study of the second language acquisition ,it .
Second Language Acquisition (SLA) refers to the study of how students learn a second language (L2) additionally to their first language (L1). Although it is referred as Second Language Acquisition, it is the process of learning any language after the first language whether it is the second, third
Language Acquisition Module 1: First language acquisition, second language acquisition, bilingualism: definitions, differences and similarities "When we study human language, we are approaching what some might call the human essence, the distinctive qualities of mind that are, so far as we know, unique to [humans]." (Chomsky, 1968, p. 100)
Introducing Second Language Acquisition Written for students encountering the topic for the first time, this is a clear and practical introduction to second language acquisition (SLA). Using non-technical language, it explains how a second language is acquired; what the learner of a second language needs to know; and why
3. the inevitability of child directed speech 62 matthew saxton 4. universal grammar approaches to language acquisition 87 maria teresa guasti 5. second language acquisition 109 susan gass part 2. windows on language acquisition 6. language and the many faces of emotion 143 judy s. reilly 9780230_500303_01_prexx.indd v 4/15/2009 8:43:26 PM. vi language acquisition 7. complements enable .
mastering a second language and begun to learn it. While research on second language acquisition has been under the way for a long time on a large scale. With the development of second language teaching, many researchers and educators try to find out the factors affecting second language Acquisition (SLA). SLA is a
Second language acquisition is needed for education, employment and other purposes, and it is typically an official or societal language (Ali Derakhshan, Elham Karimi, 2015). However, second language acquisition usually refers to any other language acquisition after that of a native language. As an independent discipline, research on second
Although adventure tourism is recognized as an important, growing tourism segment, primary research to quantify the size and scope of this market in the U.S. or internationally (Schneider 2006) has been lacking. For this reason, George Washington University, along with its partners, the Adventure Travel Trade Associ-ation (ATTA) and Xola Consulting, sought to better understand the adventure .
