Deep Analytics for Technology Innovation Management
1. Introduction
Business analytics for technology innovation projects
Requirements engineering Technology schema Technology management schema
Scope Strategy Security Structure System Skill, Style, support Staff
Critical success
factors analysis
Value chain analysis
Business Process mapping
Computing schema Networking schema
Data schema Application schema Application integration
Project analytics Financial investment model Organization structure
Figure 1.1: Business Analytics for technology innovation projects
Let us highlight the organization of our work. The work starts with the problem of technology innovation project management. It presents various models of business analytics such as Deep analytics ‘7-S’ model, SWOT Analysis and Technology life cycle analysis in terms of S-curve. Next, the business analytics framework is decomposed following a decision tree like structure (Figure 1.1.)
Technology requirements engineering schema
o Scope
o System
Technology schema
o System
Computing schema
Networking schema
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Data schema
Application schema
o Structure
o Security
Technology management schema
o Staff–resources
o Skill
o Style
o Support
o Project analytics mechanism
o Innovation ecosystem
Next we have applied the aforesaid deep analytics to reason a set of test cases associated with top seven technology innovations today.
o Solar power electronics (chapter 2)
o Electrical and hybrid vehicles (chapter 3)
o AI and smart grid (chapter 4)
o IIOT, SCADA & ICS (chapter 5)
o Deep learning for cancer prediction and prevention (chapter 6)
o Biomedical instrumentation challenges (chapter 7)
o Artificial rainfall and cloud physics (chapter 8)
o Real-time search for astronomical hazards (chapter 9)
A technology innovation project is associated with a network of organizations that satisfies the demand of ultimate customer by producing values in the form of products and services. Project management is a novel management paradigm; the basic objective is to fulfill ultimate customer demands by integrating a network of organizational units through systematic coordination of material, information and financial flows [11,15]. A project chain includes all the stages involved directly or indirectly with a project like suppliers, manufacturers, distributors, project service providers, consultants and customer. Each stage performs different processes and interacts with other stages of the chain; there is a flow of material, information and funds between different stages.
Integration of organizational units and coordination of flows of material, information and funds are the challenges of today’s project management. A lack of coordination occurs if information is distorted as it moves across the project chain or if different stages of the chain focus on optimizing their local objectives. Efficient coordination and integration depends on choice of partners or strategic alliance, inter-organizational collaboration and leadership. Effective use of information and communication technology, integration of project planning and enterprise resource planning system and process orientation ensure improved
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
coordination of different flows in project management. Collaborative planning, forecasting and replenishment is a strategic tool for comprehensive value chain management of a project. This is an initiative among all the stakeholders of the project in order to improve their relationship through jointly managed planning, process and shared information. The ultimate goal is to improve a firm’s position in the competitive market and the optimization of its own value chain in terms of optimal inventory, improved sales, higher precision of forecast, reduced cost and improved reaction time to customer demands.
Collaborative intelligence highlights the importance of sharing appropriate strategic data for greater transparency and accuracy of resource constrained, multi-objective and stochastic project planning. Information technology allows project chain partners to interconnect, but trust is also important. The interplay between trust and technology encourages the commitment of collaboration among the organizations. The partners of a project chain are often reluctant to share their private information. It is an interesting option to explore how privacy can be ensured in exchange of strategic information for collaborative intelligence. The trading agents must define points of collaboration for information sharing in terms of DCOR, CCOR and SCOR plans. DCOR plan analyzes project scope, requirements engineering, design, coding, configuration and customization. CCOR plan analyzes the strategies of erection, testing, commissioning, maintenance and performance optimization. SCOR plan analyzes demand, inventory, production and capacity, sourcing, distribution, warehousing, transportation, reverse logistics and evaluate supply chain performance in terms of lead time, cost, quality and service. The agents also analyze the intelligence and feasibility of various types of contracts such as service, sourcing, push-pull, revenue sharing and buy back contract. The project analysts assess and mitigate risks of uncertainties in delivery and capacity planning adaptively and resiliently; adjust reference project plan against exceptions and compute revised plan (P’).
The outcome of collaborative intelligence is a set of intelligent contracts. Collaborative planning is a common approach for a group of decision-making agents (DMAs) to reach mutually beneficial agreements. This is an important conflict management and group decision-making approach for making an intelligent contract by a set of agents. The agents exchange information in the form of offers, counter-offers and arguments and search for a fair consensus. Efficient coordination mechanisms are essential to achieve a set of rational structured plan. The coordination mechanisms get competitive intelligence from a set of strategic business intelligence moves. The trading agents start interaction with their initial plans; call a specific set of moves; share strategic information
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
but do not disclose private data; negotiate and finally settle the desired output plans. The basic objective is to improve the financial and operational performance of a project plan through systematic coordination of the flows of information, resources and funds. The present work is organized as follows.
2. Deep Analytics - 7-S Model
This section presents a model of deep analytics for technology innovation project management. It is basically an integrated framework which is a perfect combination or fit of seven factors or dimensions. Many technology innovation projects fail due to the inability of the project managers to recognize the importance of the fit and their tendency to concentrate only on a few of these factors and ignore the others. These factors must be integrated, coordinated and synchronized for effective project management [15,16,17].
Novelty Objectives Constraints Priority Technology
Scope (S 1 ): Application, Primary & secondary scope, Portfolio rationalization
Security (S6): Threat analytics
Risks assessment
Risk mitigation
Skill & Style (S 5 ): Pace, Planning, Execution ERP, SCM, CRM, Support, Leadership style
Strategy (S 7 ): Shared vision Strategic alliance Performance metrics
System (S 2 ): Computing, Data, Networking, Application & Security schema
Complexity
Structure (S 3 ): Activities, Nodes Ordering constraint Time & resources
Staff-Resources (S 4 ): 5M – Machine, Material, Method, Money, Man (Technical, Mgmt., Legal)
Figure 1.2: Deep Analytics ‘7-S’ model
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Traditional approaches to project management focus on long-term planning and stability to mitigate various risks. But, complex technology innovation project management requires a mix of traditional and agile approach to cope with uncertainties [8,10,12,13]. The intension driven role develops collaboration. The event driven role integrates planning and review with learning. The other important roles of the project managers are to prevent major disruptions and maintaining forward momentum continuously. They must acknowledge the emergence of a problem and then try to minimize the frequency and negative impact of unexpected events in a dynamic environment. They must be people oriented, information oriented and action oriented [14].
Traditional project management approach follows four steps such as definition, planning, execution and termination. But, no projects are so linear. Once project execution starts, reality may demand exception management i.e. the adjustment and amendment in the planning or definition phases. Each industry has a different profile of risk. Deep analytics is applicable to both adaptive and linear project management approaches for technology innovation. Many projects fail due to conventional approach which may not adapt to a dynamic business environment. Deep analytics is essential to understand the nature of a technology innovation project and identify the gaps between as-is and to-be capabilities in a systematic and compelling way. The 7-S model uses seven factors to analyze a project. The first dimension is scope. At first, it is very crucial to identify the scope of a project rationally through feasibility study and cost-benefit analysis. It is essential to identify the primary and secondary scopes through portfolio rationalization and analysis of priority, novelty, objectives and constraints of a set of projects. Perception based emotional and readymade thoughts may affect the correctness of scope analysis. Scope creep is a serious concern in project management. The second dimension is system; it is essential to decide appropriate technology. For instance, an information technology project requires sound computing, networking, data, application and security schema. The third dimension is structure of a project which is explored through complexity analysis of a graphical project network in terms of nodes, activities, estimation of time and resources and ordering constraints (e.g. sequential, parallel). The fourth dimension is staff-resources which estimates resource allocation plan in terms of man i.e. human resources (e.g. technical, management and legal staff), machine, material, method and money (fund or capital). The fifth dimension is skillandstyle; an efficient technology innovation project management approach demands multiple skills of the stakeholders such as pace or time management, resource planning, supply chain and customer relationship management, group dynamics and leadership style. The sixth dimension is security which is focused on threat analytics, risk assessment and mitigation techniques. Finally, the seventh
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
dimension is strategy which defines shared vision, communication protocol, strategic alliance, collaboration and performance measurement in terms of KPIs. ‘7-S’ model highlights a set of intelligent strategic moves to tackle uncertainties and complexities in time and resource constrained project management. Novelty indicates how intensely new innovations are crucial aspects of a project. A technology innovation project should be assessed on the scale of sophistication of technology, which may be low, medium or high. Another critical factor is the complexity of project in terms of product, service and process. Pace indicates the urgency of a project – normal, fast, time critical or blitz. Different projects have varying degrees of newness or novelty. A derivative product development project may have low risk and few future concerns. A new version of an existing product needs detailed analysis and market research. Breakthrough product development projects face high risks. Each project is unique, but not in every respect and may have some common features. The uncertainty in a project is a measure of the mix of new and mature technology and existing knowledge base; it may arise from technological aspects, new service offering or new market segments. High technology projects are subject to time delays, cost overruns and risks of product failure. The complexity base measures three different types of complications within a project such as assembly (low), system (medium) and array (high). High complexity requires efficient coordination and integration among various phases and systems of a project. Pace indicates a sense of urgency and time sensitivity. The failure of time critical projects results from the violation of milestone deadlines and related opportunity loss; blitz projects are crisis projects with extremely urgent timing. There are various strategies for optimal pace management such as contingency plans, alternative solutions in parallel, resilient approach and business cases to manage emergency and to overcome uncertainties and unexpected surprises. A technology innovation project may be delivered on time and budget through the efforts, skill and professionalism of the project managers. But, it may not meet the needs of the customer due to uncertainty and misunderstanding. The basic objective of the deep analytics is to figure out the actual structure of a project as compared with the existing capabilities, the gap and the measure of project success in terms of efficiency, impact on the customer, impact on the team, business success and preparation for the future. It is rational to take both short and long term view of a project plan since success may change during the life-cycle of a project with the change of environmental parameters and information. Does anything change from a future point of view? Does a project have sufficient flexibility to adapt to new requirements of a dynamic business environment? Are the incentives aligned properly with customer satisfaction, system performance, deadline and budget requirements? The deep analytics is useful to find the gaps between as-is
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
and to-be requirements of a project, efficient resource planning, uncertainty and risk management. Correct use of deep analytics clearly highlights low-mediumhigh benefit opportunity and low-medium-high risk difficulty.
3.SWOT Analysis
efficiency)
It is rational to evaluate strength, weakness, opportunities and threats for a strategic option on technology innovation. There may be major and minor strengths and weaknesses. Strength indicates positive aspects, benefits and advantages of a strategic option. Weakness indicates negative aspects, limitations and disadvantages of that option. Opportunities indicate the areas of growth of market and industries from the perspective of profit. Threats are the risks or challenges posed by an unfavorable trend causing deterioration of profit or revenue and losses.
4. Technological life cycle analysis
No element in this universe exists eternally. Similarly, each technology emerges, grows to some level of maturity and then declines and eventually expires [19-25].
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Figure 1 3 : SWOT Analysis
Figure 1 4 : Technology life –cycle analysis
Some technologies may have relatively long technology life-cycle; others never reach a maturity stage. Emergence of new technologies follows a complex nonlinear process. It is hard to understand how technology life-cycles interact with other technologies, systems, cultures, enterprise activities and impacts on society. All technologies evolve from their parents; they interact with each other to form complex technological ecologies. The parents add their technological DNA which interacts to form the new development. A new technological development must be nurtured; many technologies perish before they are embedded in their environments. Next phase is growth; if a technology survives its early phases, it adapts and forwards to its intended environment with the emergence of competitors. This is a question of struggle for existence and survival for the fittest. Next phase is a stable maturity state with a set of incremental changes. At some point, all technologies reach a point of unstable maturity i.e. a strategic inflection point. The final stage is decline and phase out or expire; all technologies eventually decline and are phased out or expire at a substantial cost. TLC may have other different types of phases such as acquisition, utilization, and phase-out and disposal; preparation or initiation, implementation and operation; organization, directive, delegation, coordinate, collaborative, and dissolution; acquisition; emergence, diffusion, development, and maturity. How to manage evolution of technological innovation? The nature of innovation shifts markedly after a dominant design emerges. The pace of performance improvement utilizing a particular technological approach is expected to follow an S-curve pattern. The evolution of innovation are determined by intersecting trajectories of performance demanded in the market vs. performance supplied by technologies. Diffusion of innovations indicates how new technologies spread through a population of potential adopters. It is controlled by characteristics of innovation, characteristics of the adopters (e.g. innovators, early adopters, early majority, late majority and laggards) and characteristics of the social environment.
5. Technology management
5.1
Project Analytics
Classical models of resource constrained project scheduling problems are not adequate to solve real world problems due to increased complexities and uncertainties. Intelligent project analytics are essential for complex, fuzzy, stochastic, multi-mode, resource constrained project scheduling problems with multiple objectives. This work explores how to apply the concept of intelligent deep analytics for project management. Efficient project management requires
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
coordination and integration among seven elements associated with a project such as Scope (novelty, objectives, constraints), System (technology), Structure (complexity), Staff, Skill (innovation, design, SCM, ERP) &, Style (pace, leadership), Security (threat analysis, risk assessment and mitigation) and Strategy (shared vision, communication). It is essential to define the scope of a project correctly through feasibility study, priority and cost-benefit analysis. This work presents an algorithmic Project Analytics Mechanism (PAM) in terms of agents, input, output, strategic moves, case based planning algorithm, performance metrics, revelation principle, verification protocols for security intelligence and payment function. The intelligence of PAM is explored through a set of strategic moves such as case based planning, collaborative, security and collective intelligence. The complexity of the analytics is analyzed in terms of computational cost and security analysis. Traditionally, the computational burden of project planning depends on the efficiency of heuristic search algorithm to find out the critical path of a project. But, it may not capture the uncertainties, risks and complexities involved in a real world project. The computational complexity of PAM is associated with the efficiency of case based reasoning (CBR) i.e. case retrieval and case adaptation algorithms. Case based planning searches reference plan from a case base through efficient case retrieval and adaptation mechanism. 100% matching in case retrieval is a NP hard problem. Traditionally, many CBP algorithms have tried to find exact matching between the graphs of resource and time constrained project networks. It may be practically infeasible. The basic objective of K-Nearest Neighbor Search algorithm in PAM is to search for approximate matching among the neighbors. The project analytics monitor project performance and adjusts the reference plan. The revelation principle preserves the privacy of contracts and payment function through signcryption. This work also outlines the architecture of an intelligent project analytics in terms of computing, communication, data, application and security schema. The concept of deep project analytics and PAM has been applied to analyze three test cases – smart village project, smart city project and software project management. Here the key focus elements are project analytics, deep Analytics, performance metrics, case based planning, case retrieval, case adaptation.
5.1.1. Project Analytics Mechanism [PAM]
Agents : Project chain partners /* Service providers (S), Client (C) or service consumer, Supply chain partners, Mediator */; Objectives :
search an approximate efficient plan for a multi-objective, multi-mode, stochastic, time and resource constrained project;
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
manage uncertainty through improved coordination and integration; Input: demand plan of C, service plan of S, project parameters, project case base; Strategicmoves (S7):
call deep analytics model ‘ 7-S’; /* refer section 2.1, figure 1*/
casebasedplanningthrough case retrieval, case adaptation, learning and case base maintenance;
define projectintelligence pi = f(Cp, C’ p, Sp, Bp, Mp); Cp : collective intelligence, C’ p : collaborative intelligence, Sp : security intelligence, Bp : business intelligence; Mp : machine intelligence, f: secure verification function.
fix intelligent contracts in terms of price, discount, payment terms, mode, incentives and penalty clauses through negotiation;
Project portfolio rationalization through linear / proportional / priority based selective capital allocation.
Uncertainty management strategies :
o multidimensional view of intelligentreasoning (logical, analytical, probabilistic, perception, imaginative) to assess a project in terms of novelty, technology, complexity and pace;
o Estimate critical path based on S3,S4 and S5;
o Contingency plan
o Alternative solutions and parallel paths for resiliency
o Business case
o Prototype
o Simulation game
Algorithm : define project scope (q: aspiration point, reservation point, preferential thresholds) based on S1,S2 and S 3; do case based planning; case retrieval from case base through similarity search based on q; /* refer algorithm CRA, section 2.2 */ case adaptation based on q; estimate time, resources, activity and ordering constraints; /* refer algorithm CAA, section 2.3*/ set reference project plan P; call analytics (A) during project execution verify project intelligence pi = f(Cp, C’ p, Sp, Bp, Mp) based on S4,S5,S6 and S 7; adjust P P’ based on pi ; case evaluation and case base maintenance; Paymentfunction:
fix intelligent contracts (payment terms, payment mode, penalty clause, reward) through multi-party negotiation;
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
audit business intelligence in terms of incentives received by corrupted agents and adversaries. The honest agents compute penalty function and charge the corrupted agents.
Revelationprinciple:
S and C preserve privacy and confidentiality of signcrypted contracts and payment function; also ensure non-repudiation and data integrity.
verify security intelligence (S6) of the project analytics.
o call threatanalytics and assess risks of single or multiple attacks on the analytics such as false data injection and privacy attack, sense exception and give alerts.
Identify what is corrupted or compromised: agents, communication / data / application / computing schema?
time : what occurred? what is occuring? what will occur? assess probability of occurrence and impact.
insights : how and why did it occur? do cause-effect analysis.
recommend : what is the next best action?
predict : what is the best or worst that can happen?
o Measure and monitor optimal number of critical project performance metrics (refer section 2.4) :
Operations : scope creep, project completion stage, flexibility, quality, cost, time, inventory, customer satisfaction;
Finance : revenue growth rate, cost reduction, profitability, ROI, payback period, NPV;
Human Resources (HR) : performance, productivity, capacity utilization, skill;
audit fairness and correctness of project plan computation and adjustment as per exceptions based on rationality;
monitor authentication, authorization, correct identification, transparency and accountability in project planning, execution and control;
verify system performance in terms of reliability, consistency, resiliency, liveness, deadlock-freeness, reachability, synchronization and safety.
Output : data visualization checklist (dashboards, charts, alert), performance scorecard, time series analysis, insights analysis, risk analysis, cause-effects analysis, prediction and recommendation.
5.1.2 Case Retrieval Algorithm [CRA]
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Case Based Planning [1-6]:
Case based reasoning (CBR) is a methodology for solving problems by utilizing previous experience and saves time and effort in project planning. It involves retaining a memory of previous project problems and their solutions and solving new problems by referencing the past cases. An expert presents a new query case to CBR system. The system searches its memory of past cases stored in case base and attempts to find a case that has the same problem specification of the current case. If the system does not find an identical case in its case base, it will attempt to find the case or cases that match most closely to the current query case. There are two different types of search such as similarity search and neighborhood search. In case of similarity search, the solution of the retrieved case is directly used for the current problem. The system adapts the retrieved cases if the retrieved case is not identical to the current case. In a complex search, the system requires the access of multiple case bases which are located at various locations.
Case Retrieval Algorithm [CRA]
Agents : C, S;
Input: query (q);
Protocol:
Retrieve the most similar cases (c1,…,ck) k nearest neighbors w.r.t. q from the case base; /* Refer similarity search algorithm, section 2.2.1 */
Adapt the proposed solutions to a solution s(q) compute s(q) by combining the solutions sj of the cases cj. sj is weighted as per the differences between cj and q;
Learnafter applying s(q) to q in reality Store the new solution in the case base for solving q’ .
Evaluate performance: Rejection ratio = no. of unanswered queries / total no. of queries.
Output: Recommended solution;
CBR is selected for resource and time constrained project planning due to various reasons. The domain has an underlying model, the process is not random and the factors leading to the success or failure of a solution can be captured in a structured way. Cases recur in the domain though there may be exceptions and novel cases. The solutions can be improved through case retrieval and case adaptation. Relevant healthcare cases are available at different healthcare institutes; it is possible to obtain right data. Caseretrievalis the process of finding within the case base those cases that are the closest to the current case. There must be criteria that determine how a case is evaluated to be appropriate for retrieval
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
and a mechanism to control how the case base is searched. Most often, an entire case is searched. But, partial search is also possible if no full case exists. A case is a record of a previous experience or problem in terms of problem definition, project scope in terms of application domain, size, cost, novelty, complexity, technology, pace and risks, solution methodology, project network having initial and goat state, activities, time and resource estimation, edges and constraints (e.g. ordering, time, resource, data) and project plan. A case base also stores global best practices and project management standards. All these information must be coded. 100% matching in case retrieval is a NP hard problem. Data is stored based on domain knowledge and objectives of the reasoning system. The cases should be stored in a structured way to facilitate the retrieval of appropriate case when queried. It can be a flat or hierarchical structure. Case indexing assigns indices to the cases for retrieval and comparisons. There are different approaches of case retrieval. In case of nearest neighbor search, the case retrieved is chosen when the weighted sum of the features that match the query case is greater than the other cases in the case base. A case that matches the query case on n number of features is retrieved rather than a case which matches on k number of features where k < n; different features may be assigned with different weights. Inductive approach is driven by a reduced search space and requires reduced search time. This result reduced search time for the queries. Knowledge based approaches select an optimal set of features of case by using domain knowledge. The complexity of case retrieval depends on multiple factors: (a) number of cases to be searched, (b) domain knowledge, (c) estimation of the weights for different features and (d) case indexing strategy.
Similarity Search Algorithm
input : Training objects : D; Test object: Z (a vector of attribute values); output: k nearest neighbors of Z; Algorithm: for each object y D do compute d(z,y), the distance between y and z; option 1 : eucledian distance d(x,y) = √∑n k=1 (xk – yk)2 ; option 2 : manhattan distance d(x,y) = √∑n k=1 |(xk – yk)|; sort d(z,y); end;
The algorithm computes the distance or similarity between z and all the training objects to determine nearest neighbor list for given training set D and test object z which is a vector of attribute values. The storage complexity of KNN algorithm is
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
o(n) where n is the training objects. The time complexity is also o(n) since the distance needs to be computed between the target and each training object. There are several key elements of this approach : (a) a distance of similarity metric to compute the closeness of objects; (b) the value of k, number of nearest neighbors and (c) the method of distance measurement. KNN is a specific case of instance based learning such as CBR. The performance of KNN algorithm depends on the choice of k, an estimate of the best value for k that can be obtained by cross validation. If k is very small, the results can be sensitive to the noise points. If k is too large, then the neighborhood may include too many points from the classes.
5.1.3 Case Adaptation Algorithm
Caseadaptationis the process of translating the retrieved solution appropriate for the current problem; it adds intelligence to the recommendation process. There are various approaches of case adaptation. The retrieved case can be directly used as a solution to the current problem without any modification. Otherwise, the retrieved solution should be modified according to the current problem. The steps or processes of the previous solution can be reused or modified. The solution of the current case can be derived by combining knowledge of multiple retrieved cases. Case adaptation is a complex decision making task, it considers multiple factors: how close is the retrieved case to the query case? How many parameters are different between the retrieved and the query case? DMAs can apply common sense or a set of rules or heuristics for case adaptation. Making sense of the information found during an investigational search is a complex task of case based reasoning. Sense making is to find meaning in a situation; it is the cognitive act of understanding information. The system should support collaborative information search by providing several rich and interactive views of the search activities of a group. One of the problems facing HCI research today is the design of computer interfaces to enable sense making of the processed information. Sense making is not only important for individuals, but also for groups to achieve shared goals. Traditional sense making tools focus on data mining, provide better information representation, visualization and organization of search results. But, it is also required to support the collaboration and communication that occurs among the investigators when they make sense of information together.
Interactive search: The basic steps of interactive search algorithm which operates between a DMA and the MA are as follows:
1. MA computes an initial feasible solution.
2. MA interacts with the DMA.
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
3. MA obtains a (or a set of) new solution. If the new solution or one of the previous solutions is acceptable to the DMA, stop. Otherwise, go to step 2.
Here, the MA has the option of running interactive search session with the DMAs in sequence or in parallel depending on its resources available. Either case, the result is same, since the interaction with each DMA occurs independently of interaction with others. The design of interactive search methods depends on various issues:
The form through which the DMA gives information
The approach by which the multi-objective problem is transformed into a single objective optimization problem
The type of data used for interaction with DMA
Number of non-dominated points to be presented to the DMA (a single point or a sample of points) and
How the DMA evaluates a set of alternatives?
Here, we consider a specific interactive search procedure called LightBeamSearch (LBS) method [18]. The idea of light beam search is analogous to projecting a focused beam of light from the aspiration point onto the non-dominated frontier. The lighted part of the frontier changes if the aspirationpoint or the point of interest in the non-dominated set is changed. As already mentioned above, any interactive search (including LBS) occurs between a DMA and the MA. The mediator asks the DMA to specify its preference in the form of aspiration and reservation point and various types of preferential thresholds. At each iteration of LBS procedure, MA generates a sample of non-dominated points using this preferential information. The sample is composed of a middle point and a set of non-dominated points from its neighborhood. MA shows these points to the decision-making agent. Appendix 1 has defined several parameters related to light beam search such as aspiration point (PA), reservation point (PR), indifferent threshold (Ith), strong preference threshold (Sth), weak preference threshold (Wth), veto threshold (Vth), middle point (MP) and characteristic neighbors.
Case Adaptation Algorithm [CAA]
Agents : A decision-making agent (DMA) and the mediator agent (MA).
Input : The mediator holds the deterministic problem; The DMA holds its aspiration point, reservation point, indifferent threshold, strong and weak preference threshold and veto threshold.
Output: DMA knows a set of solutions; MA can not know the output.
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
1. MA requests the DMA to specify its preferential parameters ( PA,PR,Ith, Pth, Sth, Wth, Vth) based on S1 - S 7. /* refer figure1 */
2.The DMA generates a set of preferential parameters and sends to MA.
3. Repeat until the DMA is satisfied with a solution or concludes that no compromise point exists for the present constraints
a. MA computes a middle point (MP) along with characteristic neighbors for each set of preferential parameters.
b. The DMA gets back the results of middle points along with characteristic neighbors using; DMA scans the inner area of the current neighborhood and stores its preferred solutions in a list L1; it stores the invalid middle points in a list L 2.
c. Case
(i) TheDMAwantstodefineanewaspirationand/orreservationpointand/or updatespreferentialthresholds:
The DMA adds a set of new aspiration and/or reservation points and/or new preferential thresholds and sends the same to MA.
MA projects the aspiration points onto the non-dominated set and generates middle points with characteristic neighborhood.
The DMA gets back the result of desired middle point alongwith characteristics neighbors.
(ii) TheDMAwantsapointfromthecurrentneighborhoodtobethenewmiddle pointorwantstoreturntooneofthestoredpointsofL1 :
The DMA adds the desired middle point to the list L2 and sends L2 to MA;
MA generates neighborhood of the middle points.
The DMA gets back the result of desired middle point along with characteristics neighbors.
Aspiration point: The value of an objective function which is desirable or satisfactory to the decision maker is called aspiration point.
Reservation point: The value of an objective function that the decision maker wants to avoid is called reservation point.
Nondominatedsetorparetooptimalfrontier: A decision vector x*S is pareto optimal if there does not exist another decision vector xS such that fi(x) fi(x*) for all i =1,…,k and fj(x) <fj(x*) for at least one index j; fi is objective function and S is feasible space. An objective vector z*Z is pareto optimal if there does not exist another objective vector zZ such that zi zi* for all i =1,…,k and zj < zj* for at least one index j.
Indifferencethreshold: The decision maker should inform the mediator various preference thresholds in order to compare alternatives and to define outranking relations. There is an interval of preference wherein it is not possible for the
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
decision-making agent to distinguish between different alternatives due to imprecision and uncertainty of measurements and this corresponds to indifference threshold.
Preference threshold: Strict preference threshold is defined as minimal increase/decrease of any objective that makes the new alternative strictly preferred with respect to this objective. There exists an intermediate region between indifference and strict preference threshold where the decision-making agent hesitates to compare alternatives. This corresponds to weak preference threshold.
Veto threshold: It indicates that what is the minimal increase/decrease of any objective that makes the new alternative unacceptable regardless of the value of other objectives.
Middlepoint: In each computation phase of LBS procedure, a finite sample of nondominated points is generated by the mediator agent. The sample is composed of a middle point and a set of points within its neighborhood. The starting middle point is obtained by projecting aspiration point on the non-dominated set in the direction of reservation point.
Characteristic neighbors of the middle point: For a middle point, the neighborhood is defined as a set of non-dominated points that are not worse than the middle point. The neighborhood points from the sample indicate to what extent the values of particular objectives can be improved in relation to the middle point.
5.1.4 Project Performance : KPIs and Data Visualization
It is essential for an efficient project manager to understand critical metrics and key performance indicators (KPI) and how to identify, measure, analyze, report and manage for the success of a project. KPIs and metrics are displayed in dashboards, scorecards and reports. Project metric is generic but KPI is specific. KPIs give early warning signs of poor project performance if the problems are not addressed appropriately [19]. The project success is measured in terms of time, cost, performance and customer satisfaction [20]. It is difficult to measure and monitor too many project performance metrics. Therefore, it is essential to consider optimal number of performance metrics and KPIs. It is possible to classify the performance metrics and KPIs into four categories.
Category1[Operation]: scope creep, project completion stage, flexibility, quality, cost, time, inventory, customer satisfaction; this category is associated with project success and element S2 and S 3 .
Category 2 [Finance] : revenue growth rate, cost reduction, profitability, ROI, payback period, NPV; this category is associated with element S3 .
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Category 3 [Human Resources (HR)] : performance, productivity, capacity utilization, skill; this category is associated with element S3 . Category4[Securityintelligence]: It is essential to audit fairness and correctness (i.e. accuracy of estimate and measurement) of project plan computation and adjustment as per exceptions based on rationality; monitor authentication, authorization, correct identification, transparency and accountability in project planning, execution and control; system performance should be measured in terms of reliability, consistency, resiliency, liveness, deadlock-freeness, reachability, synchronization and safety. This category is associated with element S 2 and S 6 .
5.2 Resource Allocation & Investment Strategy Analysis
When the capacity of the client is more than the total demand of a set of projects, the client may like to allocate the required resources such as fund or capital to each project using resource allocation model. However, when the capacity is less than total demand, the client would have to find the combination of projects, which would fit the resource allocation model and give maximum benefit. There are three different types of resource allocation protocols – linear, proportional and selective allocation.
Linear allocation: It is an equal sharing of the pain i.e. shortage of capacity of capital among a set of projects. If that pain exceeds the actual demand of a project, then the project becomes passive. The project Pi is allocated qi = d i – (1/n) max (0,
d* i - C) where n is the number of active projects, C is the capacity of capital of the client.
Proportionalallocation : The project Pi is allocated qi = min {d*i, C.d*
Here, n is the number of active projects and C is the total capacity of capital of the client. If the demand is more, more capital will be allocated to that project proportionately.
Selective allocation : It is basically priority based portfolio rationalization where the capital is allocated as per the priority of a set of projects. It is an interesting problem to find the allocation of the projects while maximizing the utility of the client under capacity constraints. This is basically a knapsack problem. Let {(u1,d*1),(u2,d*2), ..,(un,d*n), C} be an instance of the knapsack problem – C is the knapsack capacity i.e. total capacity of capital of the client; (ui,d*i) are respectively the utility and demand of capital of the project i. The goal is to choose a subset of projects of maximum utility with total demand of capital at most C. According to this resource capacity allocation strategy, all the projects
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
are not treated equally. In case of any shortage of capacity, several projects may become infeasible. The projects are ranked based on utility and priority and the capital is allocated as per the rank of the projects.
Figure 1 5 : Financial Investment Framework
The business analysts should consider a financial investment framework for optimal resource allocation and project portfolio rationalization along two dimensions : strategic objective and technology scope. There are four cells: transformation, experiments, process improvements and renewal. Most of the technology innovation projects fall in transformation and experiments cells. The basic objectives of transformation projects are growing need of application integration, end-to-end business process re-engineering and improved support. It demands process change. But, during economic downturn, it may be a costly option. The expected benefits are efficient customer service, greater accuracy and long-term growth. The basic objectives of experiments are to adopt new business models using new technology; the expected benefits are change of organization structure, infrastructure and business process improvements. The basic objective of process improvement is to yield more profit from improved operational performance. The process owner or a functional unit realizes benefits such as short term profitability. The basic objective of renewal is to replace old shared technology with new cost effective powerful technology maintaining the existing infrastructure and keeping it cost effective. The expected benefits are improved maintainability, reduced support and efficient capacity utilization. Resource allocation and mobilization are two critical aspects of project management. It is possible to call different types of logic such as linear, proportional and selective resource allocation (as stated above) subject to shortage of capacity. Each strategic project defines a set of objectives, strategies and demand plans and then the resources are allocated to different projects
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
according to the demand plans. It is basically the principle of management by objectives (MBO) which commits the reservation of different types of financial, non-financial and human resources. The sick projects may need new investment for turnaround and renewal; the star projects may need additional fund for continuous strategic growth; the emerging projects may need capital for appropriate technology management and skill development. The old dead assets should be divested; wastage of energy, utilities, materials and products should be minimized and existing capacity should be utilized intelligently. Resources are generally allocated to different business units through various types of budgeting such as capital budgeting, performance budgeting, zero based budgeting and strategic budgeting. Capital budgeting is decided based on payback period, NPV, IRR and profitability index. Zero based budgeting evaluates the particular demand and need of each project. It involves identification of decisive projects, analysis of each decisive project, ranking of the demand of each project and then allocation of resources. Strategic budgeting asks a set of fundamental questions: What is the goal of a project in terms of performance and results? What are the key activities or tasks to be done to achieve the goal? The management should be cautious of the risk of resource allocation such as limited resource capacity, competition and past commitments.
5. Conclusion
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Figure 1.6 : Business Analytics - System Architecture
Figure 1.6 presents the system architecture of the projects analytics system in terms of five basic layers: computing, data, networking, security and application schema. The basic building block of the system is project analytics mechanism (PAM). The output of the project analytics is a set of data visualization objects like dashboards, charts, alert, prediction, recommendation, performance scorecard, time series analysis, insights analysis, risk analysis, performance scorecard. The application schema analyzes the role of strategic, operations, HR, marketing and finance analytics for project planning, execution, control and portfolio management. It also shows the importance of enterprise application integration i.e. the interface among project analytics, enterprise resource planning (ERP), supply chain management (SCM), knowledge management system (KMS), the information systems of supply chain partners and customers.
Reference
[1] A. Aamodt, E. Plaza, Case-based reasoning: foundational issues, methodological variations and system approaches, AI Communication. 7 (1) (March 1994), 39–59.
[2] T. Bylander, An average case analysis of planning, in: Proceedings of the Eleventh National Conference of the American Association for Artificial Intelligence (AAAI-93), AAAI Press / MIT Press, Washington, DC, 1993, pp. 480–485.
[3] D.B. Leake (Ed.), Case-Based Reasoning, MIT Press, Cambridge, MA, 1996.
[4] P. Liberatore, On the complexity of case-based planning, Journal of Experimental & Theoretical Artificial Intelligence 17 (3) (2005) 283–295.
[5] H. Muñoz-Avila, M. Cox, Case-based plan adaptation: An analysis and review, IEEE Intelligent Systems 23 (4), 2008, 75–81.
[6] F. Tonidandel, M. Rillo, The FAR-OFF system: A heuristic search case-based planning, in: M. Ghallab, J. Hertzberg, P. Traverso (Eds.), AIPS, AAAI, 2002, pp. 302–311.
[7] T.W.Malone, R.Laubacher and C.Dellarocas. The Collective Intelligence Genome. MIT Sloan Management Review. Spring 2010, volume 51, no. 3.
[8] S.L.Epstein. Wanted : Collaborative Intelligence. Artificial Intelligence 221, 2015, 36 - 45.
[9] I. Averbakh. Nash equilibria in competitive project scheduling. European Journal of Operational Research 205, 2010, 552–556.
[10] W.Herroelen and R.Leus. Project scheduling under uncertainty: survey and research potentials. European Journal of Operational Research, 165, 2005, 289–306.
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
[11] S.Chakraborty. A study of several privacy-preserving multi-party negotiation problems with applications to supply chain management. IIMC, 2007.
[12] A.J.Shenhar and D.Dovir. Managing R&D Defense Projects. Tel Aviv Institute for Business Research, Tel Aviv University and Ministry of Defense, Israel. 1993.
[13] A.J.Shenhar, D.Dovir, O.Levy and A.Maltz. Project Success: A Multidimensional Strategic Concept. Long Range Planning. 34, 2001: 699 -725.
[14] A.J.Shenhar. One size does not fit all projects : exploring classical contingency domains. Management Science. Vol. 47 no. 3, pp. 394 - 414. March 2001.
[15] P.W.Morris. The Management of Projects. Thomas Telford, London. 1997.
[16] A Guide to the Project Management Body of Knowledge. Project Management Institute. 2004.
[17] R.H.Waterman, T.J.Peters and J.R. Phillips. Structure is not organization. Business Horizons, June,1980.
[18] A.Jaszkiewicz and R.Slowinski. 1999. The light beam search approach – an overview of methodology and applications. European Journal of Operational Research, 113, 300-314.
[19] H.Kerzner and C. Belack.2010. Managing Complex Projects, John Wiley & Sons and the International Institute for Learning (IIL) Co-publishers.
[20] H.Kerzner, 2006. Project Management Best Practices; Achieving Global Excellence, Hoboken, NJ:John Wiley & Sons Publishers.
[21] Attewell, P. Technology diffusion and organizational learning: the case of business computing, Organ.Sci., 3(1), 1992.
[22] Basalla, G. TheEvolutionofTechnology, Cambridge University Press, New York, 1988.
[23] Rogers, E. M. DiffusionofInnovations, 4th ed., Free Press, New York, 1995.
[24] Cardullo, M. W. IntroductiontoManagingTechnology, Vol.4,J. A. Brandon, ed., Engineering Management Series, Research Studies Press, Taunton, England, 1996a.
[25] Cleland, D. I. and King, W. R. SystemsAnalysisandProjectManagement, McGraw-Hill, New York, 1983.
Appendix 1 :
Test Case 1 : Smart Village Project
Agents : Project chain partners /* Service providers (S), Client (C) or service consumer, Supply chain partners, mediator */;
Objectives :
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Search an approximate efficient plan for a multi-objective, multi-mode, stochastic, time and resource constrained project;
Manage uncertainty through improved coordination and integration; Input: demand plan of C, service plan of S, project parameters, project case base; Strategicmoves (S7sv):
Call deep analytics model ‘ 7-S’; define the demand plan of a smart village in terms of a portfolio of projects. /* refer figure 1, section 2.1*/
o Scope (S1sv) /* refer section 4.1.1*/
o System (S2sv) : civil, mechanical, electrical and electronics;
casebasedplanningthroughcase retrieval, case adaptation, learning and case base maintenance;
define projectintelligence pi = f(Cp, C’ p, Sp, Bp, Mp); Cp : collective intelligence, C’ p : collaborative intelligence, S p : security intelligence, Bp : business intelligence; Mp : machine intelligence, f: secure verification function.
fix intelligent contracts in terms of price, discount, payment terms, mode, incentives and penalty clauses through negotiation;
Project portfolio rationalization through linear / proportional / priority based selective capital allocation.
Uncertainty management strategies :
o multidimensional view of intelligentreasoning (logical, analytical, probabilistic, perception, imaginative) to assess a project in terms of novelty, technology, complexity and pace;
o Estimate critical path based on S3,S4 and S5;
o Contingency plan
o Alternative solutions and parallel paths for resiliency Algorithm : define project scope (q: aspiration point, reservation point, preferential thresholds) based on S1,S2 and S 3; do case based planning; case retrieval from case base through similarity search based on q; retrieve a case of globally benchmarked village; /* refer algorithm CRA*/ ; case adaptation based on q; estimate time, resources, activity and ordering constraints; /* refer algorithm CAA*/ set reference project plan P; call analytics (A) during project execution verify project intelligence pi = f(Cp, C’ p, Sp, Bp, Mp) based on S4,S5,S6 and S 7; adjust P P’ based on pi ; case evaluation and case base maintenance. Paymentfunction:
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
fix intelligent contracts (payment terms, payment mode, penalty clause, reward) through multi-party negotiation;
Audit business intelligence in terms of incentives received by corrupted agents and adversaries. The honest agents compute penalty function and charge the corrupted agents.
Revelationprinciple:
S and C preserve privacy and confidentiality of signcrypted contracts and payment function; also ensure non-repudiation and data integrity.
verify security intelligence (S6sv) of the project analytics.
o call threatanalytics and assess risks of single or multiple attacks on the analytics such as false data injection and privacy attack, sense exception and give alerts.
Identify what is corrupted or compromised: agents, communication / data / application / computing schema?
time : what occurred? what is occuring? what will occur? assess probability of occurrence and impact.
insights : how and why did it occur? do cause-effect analysis.
recommend : what is the next best action?
predict : what is the best or worst that can happen?
o Measure and monitor optimal number of critical project performance metrics (refer section 2.4) :
Operation : scope creep, project completion stage, flexibility, quality, cost, time, inventory, customer satisfaction;
Finance : revenue growth rate, cost reduction, profitability, ROI, payback period, NPV;
Human Resources (HR) : performance, productivity, capacity utilization, skill;
audit fairness and correctness of project plan computation and adjustment as per exceptions based on rationality;
monitor authentication, authorization, correct identification, transparency and accountability in project planning, execution and control;
verify project performance in terms of reliability, consistency, resiliency, liveness, deadlock-freeness, reachability, synchronization and safety.
Output : data visualization checklist (dashboards, charts, alert), performance scorecard, time series analysis, insights analysis, risk analysis, cause-effects analysis, prediction and recommendation.
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Scope for a smart village : Let us first explore the basic objectivesof a smart village project i.e. the basic needs of human life in today’s village.
Scope (S1sv):
- Agriculture (outlets for seeds, fertilizers and pest controllers, training centers), fisheries, poultry and dairy firms and warehouses; /* use of natural organic compost fertilizer from crop wastes, domestic waste and cowdung is a cost effective solution in agriculture */;
- Retail outlets and market : efficient supply chain management of nutritious food and beverages, low priced fashionable garments and consumer goods (FMCG);
- Accommodation : disaster proof nano-housing scheme protected from flood, storm, cyclone and earthquake, toilet, drainage system; resiliency against disaster management :
o Weather : heat wave, storm, flood, rainfall, earthquake, fire, snowfall;
o Security : anti-social activities;
o Environment : paste , snake, wild animals, mosquitoes, flies;
o Pollution control : air, water, soil;
- Infrastructure for logistics and transportation: roads, highways, bridges, bus stop, rail stations, ferry service, service centers;
- Energy : solar power, wind power; /* focus on renewable energy */
- Utility : adequate supply of pure drinking water, induction cooker, gas, oil;
- Healthcare : hospitals, clinics, medicine retail outlets;
- Education : Pre-primary, secondary and higher secondary schools, colleges, offline and online libraries, vocational training centers for skill development;
- Communication : mobile and landline, internet;
- Entertainment, religious and cultural affairs : smart TV and radio broadcast, marriage halls, temples, mosques, churches;
- Sports and games : play grounds, yoga and meditation centers;
- Banking and financial services (e.g. insurance, retirement planning);
- E-governance : tax, land management, e-panchayet, police stations, court.
Test Case 2 : Smart City Project
Agents : Project chain partners /* Service providers (S), Client (C) or service consumer, Supply chain partners, mediator */;
Objectives :
Search an approximate efficient plan for a multi-objective, multi-mode, stochastic, time and resource constrained project;
Manage uncertainty through improved coordination and integration;
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Input: demand plan of C, service plan of S, project parameters, project case base; Strategicmoves (S7sc):
Call deep analytics model ‘ 7-S’; /* refer figure 1, section 2.1*/ ; define the demand plan of a smart city in terms of a portfolio of projects on design, erection, construction, testing and commissioning.
o Scope (S1sc); /* refer section 4.2.1 */
o System (S2sc) : civil, mechanical, electrical, electronics, chemical, metallurgical;
casebasedplanningthrough case retrieval, case adaptation, learning and case base maintenance;
define projectintelligence pi = f(Cp, C’ p, Sp, Bp, Mp); Cp : collective intelligence, C’ p : collaborative intelligence, Sp : security intelligence, Bp : business intelligence; Mp : machine intelligence, f: secure verification function.
fix intelligent contracts in terms of price, discount, payment terms, mode, incentives and penalty clauses through negotiation;
Project portfolio rationalization through linear / proportional / priority based selective capital allocation.
Uncertainty management strategies :
o multidimensional view of intelligentreasoning (logical, analytical, probabilistic, perception, imaginative) to assess a project in terms of novelty, technology, complexity and pace;
o Estimate critical path based on S3,S4 and S 5;
o Contingency plan
o Alternative solutions and parallel paths for resiliency
Algorithm : define project scope (q: aspiration point, reservation point, preferential thresholds) based on S1,S2 and S 3; do case based planning; case retrieval from case base through similarity search based on q; /* refer algorithm CRA*/; retrieve a case of globally benchmarked smart city; case adaptation based on q; estimate time, resources, activity and ordering constraints; /* refer algorithm CAA*/ set reference project plan P; call analytics (A) during project execution verify project intelligence pi = f(Cp, C’ p, Sp, Bp, Mp) based on S4,S5,S6 and S 7; adjust P P’ based on pi ; case evaluation and case base maintenance;
Paymentfunction:
fix intelligent contracts (payment terms, payment mode, penalty clause, reward) through multi-party negotiation;
Reference:EBOOK/BAV1.0/24072018bySumitChakraborty
Another random document with no related content on Scribd:
clumsy Gestes ensemble la vie du preulx chevalier Bayard (1525) with the ideal of a knight “sans peur et sans reproche.” It will seize upon the death of Sir Philip Sidney as romantic. It will survive the allegorizing of Spenser. It is the refuge from the industrial age in Tennyson’s Idylls of the King. If fighting and politics remain as ugly as in the fifteenth century, Edwin Arlington Robinson will not be the last poet of romance.
Boiardo makes the contrast very explicit.
[The robber replied] “What I am doing every great lord does in your upper world. They make havoc of their enemies in war for aggrandizement and to cut a bigger figure. A single man like me makes trouble for seven, perhaps ten; they rage against ten thousand. And they do still worse than I in that they take what they do not need.” [Said Brandimarte] “It is indeed a sin to take from one’s neighbor as my world does; but when it is done only for the state, it is not evil; it is at least pardonable.” [The robber replied] “A man is more easily pardoned when he frames the charge himself. And I tell you, and make full confession, that I take what I can from any one who can less” (II xix 40)
O Fame, attendant of emperors, nymph so singing great deeds in sweet verse that thou bringest men honor even after death and makest them eternal, where art thou fallen? To sing ancient loves and tell the battles of giants. For the world of this thy time cares no longer for fame or for excellence (II. xxii. 2).
Then with choice rhymes and better verse shall I make combats and loves all of fire Not always shall the time be so out of joint as to drag my mind from its seat. But now my songs are lost. Of little avail to give them a thought while I feel Italy so full of woe that I cannot sing, and hardly can I sigh. To you, light lovers and damsels, who have at heart your noble loves, are written these fair stories flowering from courtesy and valor. They are not heard by those fell souls who make their wars for despite and rage (II. xxxi. 49-50).
The distinctive difference between the two fifteenth-century romancers is that Malory translates; Boiardo rewrites. Malory may contract or expand, adapt or add; but in general he follows what he calls “the French book.” Boiardo, finding the Italian versions vulgarized (tra villani, II. xii. 3), wishes to make the old tradition once more literary. To restore their dignity, he gives the paladins more than
verse. So romance throughout the Renaissance, as before and since, both survived and was changed. It was rehearsed, translated, printed in its medieval forms; and it was shaped to a distinctive Renaissance pattern.
2. SEPARATE ROMANCES
What the presses most readily carried on from the Middle Age was the separate romances that had not been merged in one of the cycles. The old fairy-mistress story told by Jean d’Arras as Mélusine (1387) was printed at Geneva in 1468. Pontus and the Fair Sidoine, translated into German in 1468, was printed in French at Lyon in 1484 and in German at Augsburg in 1548. Amadis of Gaul traveled from Brittany to Spain and Portugal and back to France. The French prose version printed in 1540 and again in 1548,[33] typically romantic in love, adventure, and chivalry, deserved its popularity also by narrative skill. With some lyric dilation of love and an occasional allusion to classical mythology, the style is generally restrained to the narrative purpose. Description, rarely dilated, is often cleverly inwoven. Dialogue adds not only liveliness, but some characterization. Though the simple transitions sound like Malory’s (“Now the author leaves this and returns to the treatment of the child”), this shifting is not frequent; often it does not really interrupt; and generally the composition has distinct narrative sequence. The knighting of Amadis is not merely a scene in a series; it is a situation, prepared and pointed as at once fulfillment and promise. So the complication of the rings (Chapter XI) is carried out before our eyes through suspense to solution.
In such separate romances the Middle Age had advanced the art of verse narrative. Not only in Chaucer’s masterpiece Troilus and Criseyde, but also in the contemporary Gawain and the Green Knight, the story is carried forward in consistent sequence to a distinct issue of character. The Renaissance, though generally it had other preoccupations, caught some of this vigor in the telling of single romances. One of these, loosely related to the Arthurian cycle without ever being embodied, is of Giron; and this was put into Italian
stanzas by Luigi Alamanni (1495-1556). An industrious and capable man of letters,[34] who spent much of his life in France, he was a convinced classicist. His Gyrone il cortese (Paris, 1548; Venice, 1549), though it was written long after Ariosto had secured fame by quite other methods, shows that he felt the obligation of a single story, distinct from a cycle, to keep narrative sequence.
Alamanni’s Dedicatory Epistle, after relating the Arthur stories to history, and mentioning some dozen of the Arthurian heroes, expounds tournaments and the quests of errant knights, and offers his Gyrone, as Caxton offers his Malory, to educate toward true valor. Confessing that he has not always followed his source in detail, he promises later “another new work of poetry made in the ancient style and order and imitating Homer and Vergil ” [This was fulfilled in the last years of his life by his Avarchide ]
The first five books proceed as follows. After preliminary adventures, Gyrone goes to his bosom friend Danain at Malahalto. On news of a tournament they decide to go disguised in black arms (II). But Danain’s wife languishes for love of Gyrone and gains permission to go thither also under escort while the two friends lodge with a hermit. After combats on the way, they arrive at the tournament as Sagramor is victorious in the first jousts The beauty of Danain’s wife arouses the jealousy of the other ladies and the passion of the Greek king Laco His ardor and his threats to seize her are overheard by Gyrone, who courteously rebukes him in a lecture
Further description of the combats at the tournament (III) leads to the final victory of Gyrone and Danain. While Laco still yearns (IV) toward Danain’s wife, a messenger arrives to conduct her to a neighboring castle. Laco parts from Meliadus, as Gyrone from Danain. Thus Laco and Gyrone are left sighing for the same lady in the same forest. They meet, express their admiration of each other, and sleep side by side in the wood
In the morning Laco routs single-handed the lady’s whole escort (V). The lady appealing eloquently to his honor in vain, Gyrone arrives and fells him. While the lady debates with herself whether to reveal her love, and Gyrone is torn by the conflict of his own with his loyalty to Danain, they are irresistibly drawn together. In a flowery mead by a spring they prepare for love. But Gyrone’s lance falling knocks his sword into the spring. When he has retrieved it, he reads as never before its inscription summoning to honor, and turns it in shame on himself. A peasant supervening betrays their sad plight to Laco
Thus the story is brought definitely to a situation of character Obvious as the Renaissance manipulation is in the space given to love, the handling makes this not merely lyric interlude, but story motive. Though Alamanni is unwilling, or unable, to carry this through the 28,000 lines of his twenty-four books, though he often fails to give that salience to critical situations which is evident here, he nevertheless achieves what the Renaissance cyclical romances generally ignored, narrative sequence. All he needed to make his Gyrone shorter, tighter, more compelling, was firmer control of fourteenth-century narrative art.
For a classicist Alamanni is remarkably sparing of the fashionable Renaissance allusions. He does, indeed, use that paganizing phrase which was satirized by Erasmus. “In the consecrated temples, devoutly about the sacrifices in accordance with true example, they listened, adored, and besought the immortal Father”; when Malory would say, “They heard their Mass and brake their fast.” But in spite of many Vergilian similes and of occasional orations to troops, Alamanni’s classicism is not intrusive. Apparently he thought it had small place in romance.
3. THE ARTHURIAN CYCLE IN MALORY
Among Renaissance romances presenting a traditional cycle in medieval form the most distinguished is Malory’s Morte d’Arthur. Caxton’s preface is a manifesto of romance; and his table of contents displays most of the stories that had gradually been brought together by the Middle Age at the Round Table. Between the “enfances” of the first book and the last great battle in the west are Balin and Tristan; that Percival, here called Gareth or Beaumains, who was reared apart in the wildwood; the mighty Lancelot, his mistress the Queen, and Elaine who died for love of him; the quest of the Grail; the traitor Modred. They are not composed in a narrative sequence. Balin, Gareth, Tristan, for instance, remain separate stories. For there is no real connection in Balin’s glimpse of an earlier Grail story not used in Malory’s Grail books, or in Gareth’s coming to Camelot and his knighting by Lancelot. But there is no
confusion. The separate stories are told straightforwardly; the main persons become familiar; and the exposure of Guinevere makes a crisis contributing to the subsequent ruin of the goodly fellowship. The Morte d’Arthur is not merely a series. But its distinction is in style. Malory’s prose follows that medieval habit which may be called pure narrative, the telling of a story singly for its story values. It was not the only medieval narrative habit; nor is he the only fifteenthcentury author to follow it; but it stands out both in contrast to classicism conceived as ornamental dilation and in his own quiet mastery Without parade, without pause for ornament, he maintains a grave simplicity that ranges from homeliness to eloquence. His rhythm—he has little other sentence art—lingers or quickens with the action, and answers the emotion.
Is that knyght that oweth this shelde your love? Yea truly, said she, my love he is God wold I were his love (XVIII xiv)
Than syr Bedwere cryed: “Ah! my lord Arthur, what shal become of me now ye go from me and leve me here allone emonge myn enemyes?” “Comfort thy self,” sayd the kyng, “and doo as wel as thou mayst; for in me is no truste for to truste in. For I wyl in to the vale of Avylyon to hele me of my grevous wounde” (XXI. v).
4. THE CAROLINGIAN CYCLE ON THE STREET
The other cycle, the Carolingian, was popularized in Italy by street story-tellers, who seem to have been, on the piazza before audiences of artisans and shopkeepers, somewhat like the medieval jongleurs before their successive groups of pilgrims. Their narrative art can be only guessed; for it was oral. But the guess is helped by the persistence, even to the present time, of the Carlomagno marionettes. The recital animated by these large puppets—for it is recital, not drama—is of a traditional version called I paladini di Francia, and goes on day after day by mere aggregation, and with many tirades.
5. PULCI
Italian literary manipulation of the Carolingian cycle in verse romances began with Luigi Pulci (1432-1484). His Morgante maggiore (1481, though largely written by 1470)[35] is selective. Though it bulks large with more than 30,000 lines of verse, it does not rehearse the deeds of the paladins by the serial method of installments. At the end of Canto 5 two main stories, Orlando’s and Rinaldo’s, are brought together. At the end of Canto IX, having meantime moved together, the two arrive, with the other persons whom they have picked up on the way, for the relief of Montauban and of Carlomagno at Paris. There is narrative progress from salience to salience. The dialogue is lively. Though it does not amount to characterization, it suffices for speed and for mood.
Said Rinaldo, “Wilt thou be so obtuse as not to look at that damsel? Thou wouldst not be acceptable as a lover ” Said Oliver, “Thou art ever for thy jokes Yonder is something more serious than word-play” (IV 61)
Oliver looked at Rinaldo, hardly able to hold his gaze for weeping, and said: “’Tis true that man cannot hide love and coughing. As thou seest, dear brother, love has caught me at last with his claws. I can no longer hide this desire. I know not what to do, what to decide. Cursed be the day on which I saw her. What am I to do? What dost thou advise?” Said Rinaldo, “Believe me, thou wilt leave this place. Leave the lady, marquis Oliver Our intention was not to yearn, but to find Orlando” (IV 88-90)
The naughty machinations of Malagigi are, indeed, comic interpolations; nothing comes of them; but the machinations of Gano have narrative function. There is hardly any separable description. Love laments are sketched, not dilated. Pulci is interested, and interests us, in his story as a story. To this end he takes a free hand with events. We have the usual paynim siege of Paris or of Montauban, but no attempt to include all the items of tradition. Pulci takes what he wishes and puts it where he wishes.
To call the poem a burlesque is misleading. The incidental farce, as in Boiardo and Ariosto, is rather appeal to Renaissance fondness for the grotesque as contrast. Though Pulci may have wished to pierce the inflation of the Carolingian street tirades, he was too clever to think of holding parody through 30,000 lines. Reducing the
medieval aggregation to an intelligible story, he also, with an art more delicate than burlesque, reduced the style. Turning from both medieval gravity and Renaissance luster, he brought romance down to earth. Oliver falls in love promptly, utterly, and successively. The humor of this in real life Pulci frankly seeks. When two knights dare each other, he renders their speech not as oration, but as homely flyting. He is irreverent in the way of fashionable conversation. But his main object and achievement, as it is not parody, so it is not satire. It is pleasant, often humorous story of familiar antiquated persons in traditional events and setting, but in daylight.
6. BOIARDO
Boiardo, indignant at the degradation of the Carolingian heroes among the vulgar—how did he proceed toward elevation?
Who will give me the voice and the words, and utterance magnanimous and profound? (I xxvii 1)
Till now my song has not ventured far from shore. Now I must enter upon the great deep, to open immeasurable war. All Africa lies beyond that sea; and all the world flashes with men in arms (II. xvii. 2).
The poet seems to nerve himself, as Vergil at the opening of Aeneid VII, for loftier diction. The average Renaissance poet of the next century would invoke the Muses for that “high style” which had come down from classical through medieval rhetoric. But the words of Boiardo’s invocation are not heightened thus; nor is his diction generally. He not only omits the Muses here; he is very sparing with classical allusion throughout.
Book I mentions the Cyclops, Circe, Thyestes, Medusa, the Centaurs, Vulcan, Atalanta, the dragon’s teeth, and Hercules; Book II, a faun, the easy descent to Hades, the god of love, Pasithea, Narcissus, the Laestrigonians and Anthropophagi, the goddess Fame, Arion, and the Sibyl. Hector’s arms are brought in as a piece of medieval derivation from Troy. Classical similes are inserted here and there, as if conscientiously: the meeting of two winds, fire in
grain, a boar or bear at bay, two bulls or two lions. Nor does Boiardo strive for other ornament. His heightening is rather the sheer extravagance of epic brag.
Their blows were heard nearly a mile in the wood (I. iii. 59).
They came with such a battle-cry as made earth tremble, and sky and sea (I. iv. 51).
The moat brimmed with the blood of the slain (I. xi. 32).
Fire came to his heart and his face, and flamed from his helmet. He ground his teeth. His knees so clamped Brigliadoro that the mighty steed sank in the path (I. xv. 19).
The grinding of his teeth could be heard more than a bow-shot (I xv 33)
Otherwise his words are usually as simple as Malory’s. So far, classicism has made no headway in Renaissance verse narrative.
Boiardo’s sentences, as Malory’s, are typically aggregative, sometimes even crude. Instead of tightening a sentence or a stanza, he remains frankly diffuse. Fluent, sometimes slack, he runs on as if orally. His verse is pleasantly varied. Though he hardly ever lets a line end with a down-beat, he freely begins either up or down, or shifts to a dactyl. A stanza rarely runs over; but it is often linked with the next by refrain, as in popular poetry. Here and there the closing line of his octave sounds like an experiment in the direction followed later by Spenser. Inferior in stanza control to Boccaccio and to Ariosto, diffuse, somewhat careless, he is always agreeably and sometimes charmingly fluent.
Yet description, which became a regular Renaissance cue for dilation, Boiardo handles economically. Even where he is conventional he does not dilate; and usually he is both distinct and concise.
That spring was all adorned with white and polished alabaster, and so richly with gold that it shone in the flowery mead (I. iii. 33).
A fair rich palace made of marble polished so smooth as to mirror the whole garden (I viii 2)
Secret gardens of fresh verdure are above on the roofs and hidden on the ground. Gems and gold pattern all these noble and joyous places. Clear springs unstintingly fresh are surrounded by shady thickets Above all, the place has an odor to give oppressed hearts their joy again (I viii 5)
His stories pictured on walls (depintura istoriata) whether fresco or mosaic, have a literary source. They are from the Troy stories pictured on the walls of Dido’s palace in Aeneid II. Boiardo’s briefer rendering may have been suggested by the “epigrams” of the Greek Anthology, or by survivals in southern Italy of such pictures with verse inscriptions. Certainly his palaces and gardens often recall the Norman-Arab art of Sicily. For it is art that he pauses to note oftener than scenery. In all this his classicism is both discreet and artistic. He does not borrow; he adapts.
The larger scene, the field of the traditional struggle of East with West, receives more definite geography. The haze over medieval Ermonie had been often pierced by merchant voyagers. Though there are still the Isole Felice or Lontane, we read now of Aragon and Barcelona, Granada, Toledo, Seville, Valencia, and Gibraltar; of Agrigentum and Mongibello as well as of Sicily at large; of Cyprus, Crete, and Rhodes; of Aigues Mortes, Bordeaux, Gascony, Languedoc, Perpignan, and Roussillon; of Damascus, Niniveh, Trebizond, and Tripoli.
The traditional chivalric equality of Saracen knights with Christian, as in Malory, is emphasized.
King Charles the Great with genial face had seated himself among his paladins at the round table Before him were also Saracens, who would not use chair nor bench, but lay like mastiffs on their rugs, scorning the usage of the Franks (I. i. 12-13).
The paynim king Balugante, divining Rinaldo’s irritation at some of his fellow Christians, sends him a courteous and discerning message Saracen knights are armed, titled, respected as are Christian, and mingle with them freely Their bravery is not merely admitted; Rodomonte is a legendary demon of force, and Ruggiero in his pagan days is a pattern of both force and courtesy
The traditional echoes of folklore are repeated. Feraguto’s strength revives when he touches earth. A child stolen in infancy is recognized. Herbs are gathered under a new moon. Ruggiero, as Percival, is brought up beyond sight of arms. There are waters of forgetfulness, a loathly lady waiting to be restored by a kiss, a magic steed, a white hart, a monster adversary transformed, and a retreat under water.
Grotesque interludes, barely touched by Malory, found occasionally elsewhere, and quite regular in Italian popular versions, are not only admitted by Boiardo; they are dwelt upon with evident relish. Thus Rinaldo fights with a giant.
Of no avail the furious assault; of no avail the baron’s nimble skill He could not reach so high Suddenly Rinaldo dismounted and with one bound leapt upon the giant’s croup when he was not looking He knocked his helm and his steel cap to pieces and, redoubling his strokes as if he were hammering iron at the furnace, he split the great head in two Fell the giant with a rush that made the earth shake (I. iv. 64-65).
Orlando leapt even higher, so that again and again he met his giant face to face—in the air. Angelica threw into a monster’s mouth a cake that stuck his teeth together, so that Rinaldo might safely, though with enormous effort, strangle him. Rodomonte bare-headed at sea hears his hair rattle with ice. Astolfo is beguiled to board a whale, and Rinaldo follows, both on horseback.
On Bayard he plunged into the sea after the great fish in desperation. That whale went slowly, slowly; for it is very large and by nature grave (II. xiii. 65).
Marfisa, the woman knight, had her horse stolen and pursued the thief long in vain. The conception is grotesque; the execution, pure farce.
A fortnight had she followed him, nor was fed meantime on aught but leaves The false thief, who was most astute, sped his flight with quite different food For he was so quick and so bold that every tavern he saw he would enter and fall to eating, and then flee without paying his shot
And although the taverners and their waiters were after him with their pitchers and jugs in their hands, off he was, wiping his mouth and grinning (II xv 68-69)
Love, announced in the title Orlando innamorato and frequently asserted, has little more scope in Boiardo than in Malory. Whether the title expresses an original intention abandoned, or an appeal to court ladies, or merely a certain period in the hero’s life, Boiardo’s interest was elsewhere.
Long time Morgan, Alcina, and their magic wiles have kept me waiting; nor have I shown you a good sword-stroke (II. xiv. 1).
The good sword-stroke is what he gives with both hands, even as Malory.
Since the main interest is single combat, and all the fighters, even the Saracens, are memorable, there is a long roll of persons. Of the hundred mentioned in the first five cantos about a third never reappear; few are characterized consecutively; none is consecutively in action for any considerable period. The long poem is frankly a series, not a sequence. Boiardo’s usual method is to carry one of his stories to a crisis, leave it to pick up another, and so on.
Let us now return to Astolfo, who remained, you know, alone at the fountain (I. ii. 17).[36]
As the Carolingian recitals on the piazza, or behind the marionettes, the Orlando innamorato may be entered at almost any point. What is there heard or read is interesting mainly for itself, very little as arising from previous action and characterization or as preparing for what follows. To say, then, that it fuses the two cycles is quite misleading. Boiardo brings in an Arthurian name, Tristan or Lancelot, as simply as he adds another Carolingian. He puts Merlin’s well in the forest of Ardenne. He interpolates Morgan le Fay among the Orlando stories. But fusion, whether of these Arthurians or of his own Carolingians, is not in all his thoughts. He is engaged not in composing the Carolingian story, but in rehearsing the Carolingian stories.
Boiardo’s Orlando, then, is a collection of heroes fighting in the struggle between East and West. Within that frame, as within the frame of Arthur’s Round Table, tradition had collected many stories. Boiardo finds room for them, and even for others quite unrelated. Those of the greatest knights, Orlando, Rinaldo, Oliver, and their ladies and friends, the obligatory stories, he can tell by installments because they are familiar and have been already connected. The others he inserts here and there for variety. Not only does he accept the medieval cyclical aggregation, he ignores the later medieval achievement of narrative sequence in smaller scope through characterization.
A certain Tisbina, who has nothing to do with Charlemagne, is in much the same dilemma as Chaucer’s Dorigen in the Franklin’s Tale; and her Iroldo’s response is much the same as that of Dorigen’s Arveragus. Chaucer’s solution is convincing, in spite of impossible marvels, because it is motivated by Dorigen’s character. Boiardo’s solution is inferior because it is quite extraneous and casual. His Tisbina is not characterized sufficiently to motivate the story toward any convincing issue.
Much less is Boiardo concerned to motivate his whole story. His Orlando in love is even removed for long stretches from the great struggle; and Boiardo interrupts both the love and the struggle to tell of Tisbina and Iroldo or insert a fabliau. [37] True, his poem remained unfinished; but evidently he had no idea of making it a coherent whole. His Latin and Greek did not suggest to him the shaping of verse narrative. Discernible in his style, though never intrusive, they do not move his composition; for composition was not his concern. Ignoring alike the medieval progress and Pulci’s narrative cleverness with his own material, he was content with abundant activity, variety, and fluency.
But in another use of the classics he forecast the Renaissance habit of encomium. Ruggiero, legendary ancestor of his Ferrara patrons, brought up in paganism and remote from deeds of arms, is sought by Agramante for his great expedition against the Christian West. Ruggiero’s aged tutor warns Agramante against the ultimate consequences of taking the marvelous youth into France.
Charlemagne, he says, may be defeated, and our pride and courage enhanced;
... but afterward the youth will become Christian, and ah! traitress house of Maganza, which heaven should not tolerate on earth in the end Ruggiero shall have through thee his death.
Would that were the final grief! But his descendants shall remain Christian, and come to honor as great as any the world knows today They shall keep all, all generosity, all courtesy, sweet love and joyous state in a house the flower of the world.
I see Hugo Alberto di Sansogna descend to the Paduan plain, expert in arms, in intellect, in all the ways of glory, generous, noble, and above all humane Hear, ye Italians: I warrant you He who comes with that standard in hand brings with him all your redemption Through him shall Italy be filled with prowess
I see Azzo I and Aldobrandino III, nor know which to call the greater; for the one has killed the traitor Anzolino; the other has broken the Emperor Henry. Behold another Rinaldo paladin. I say no more of him than Lord of Vicenza, of Treviso, of Verona, who strikes the crown from Frederick.
Nature shows forth her treasure Lo! the marquis who lacks no point of honor. Blest the age, and happy they who shall live in a world so free! In his time the golden lilies shall be joined to that white eagle whose home is in heaven; and his domain shall be the flower of Italy from the one fair seacoast to the other.
And if the other son of Amphitryon, who there appears in habit of a duke, has as much mind to seize dominion as he has to follow good and flee evil, all the birds not to say the men who act in this great play would flock to obey him But why should I gaze further into the future? Thou destroyest Africa, King Agramante (II xxi 54-59)
With reminiscence, perhaps, of Dante, this is obviously patterned on Vergil. The historical vision of the house of Este has its model in the vision of the Augustans. Encomium with Boiardo is neither so frequent nor so fulsome as it was to become with Ariosto and with Spenser. Was that, perhaps, one reason for his double eclipse? He was first superseded by Ariosto and then rewritten by Berni.
7. ARIOSTO
The brilliant Orlando furioso (1516) of Ludovico Ariosto (14741533) is one of the most typical verse narratives of the Renaissance, as it was the most popular. More accomplished than Boiardo in diction, verse, and composition, and more responsive to the Renaissance, Ariosto still follows the same serial plan. The two poets differ more in degree than in kind. Both were trained in the classics; both began by writing Latin; both offer romance as inspiring contrast with actuality.
O great hearts of those ancient knights! They were rivals; they differed in religion; they still felt the rude and wicked strokes aching throughout their bodies; and yet through dark woods and crooked paths they went together without distrust (i 22)
[War has been debased through the diabolical invention of artillery (xi. 22-25)]. How foundest thou ever place in human hearts, O invention criminal and ugly! Through thee military glory has been destroyed, through thee the craft of arms dishonored, through thee valor and prowess so diminished that oft the knave seems better than the good soldier.
And though Rinaldo was not very rich in cities or treasure, he was so affable and genial and so prompt to share with them whatever he had that not one of his meinie was drawn away by offer of more gold. A man of Montauban never forsakes it unless great need constrains him elsewhere (xxxi. 57).
O famished, deformed, fierce Harpies, who in blinded and misguided Italy, perhaps as punishment of ancient sins, bring to every table divine judgment! Innocent children and faithful mothers drop with hunger while they see one feast of these foul monsters devour what might have kept them alive [Italy cries] Is there no one of you to free your tables from the filth and the claws ... as the paladin freed that of the Ethiop king? (xxxvi. 1-3).
Encomium with Ariosto becomes pervasive. Animated of course by the personal ends of a court poet, it serves also the literary end of magnificence, the first aspect in which the Renaissance viewed epic.
Who will give me the voice and the words fit for a subject so noble? Who will lend me wings strong enough to attain my lofty conception? Far greater than its customery heat must be the poetic furor in my breast. For
this part I owe my lord, since it sings of the noble line from which he sprang.
Among the illustrious lords issued from heaven to govern the earth, never seest thou, O Phoebus who surveyest the wide earth, a race more glorious in peace or in war, nor any whose nobility has been kept longer, and shall be kept, if that prophetic light which inspires me errs not, so long as the heavens revolve about the pole (iii. 1-2).
The Vergilian vision of Augustan Rome is heard again in Merlin’s prophecy. Epic rolls of honor muster the English warriors, the women of Este, even the painters. Besides these are many incidental references, especially at canto openings.
Of courtesy, of nobility, examples among the ancient warriors were many, and few are there among the moderns But of impious ways enough was seen and heard in that war, Hippolito, whose captured standards thou hast used to adorn our temples, as thou broughtest to thy ancestral shores their captive galleys laden with prey (xxxvi. 2).
The mission of the poet to confer fame, proclaimed by Ariosto and repeated by Ronsard, is also seen in bitter contrast. The speaker is St John the Evangelist.
So worthy men are snatched from oblivion worse than death by poets. O intelligent and wise princes who follow the example of Augustus in making writers your friends, and thus need not fear the waves of Lethe. Poets, as singing swans, are rare, poets not unworthy of the name; for heaven prevents too great abundance of famous ones by the great fault of stingy lords, who by oppressing excellence and exalting vice banish the noble arts We may suppose that God has deprived these ignorant men of their wits and darkens their light of reason in making them shy of poetry, that death may quite consume them For wicked as their ways might be, if only they knew how to win the friendship of Apollo they might rise from their graves in sweeter odor than nard or myrrh.
Aeneas was not so pious, nor Achilles so mighty, as their fame, nor Hector so brave There have been thousands and thousands who might with truth have been put before them; but the palaces or great villas bestowed by their descendants have given them sublime honors without end at the honored hands of writers Augustus was neither so holy nor so benign as sounds the trumpet of Vergil; but his having good taste in poetry
brings him pardon for his unjust proscription. Nor would he who had against him earth and hell have the less fame, perhaps, if he knew how to keep the writers his friends
Homer made Agamemnon victorious, the Trojans cowardly and dull, and Penelope constant to her husband through the thousand persecutions of the suitors. If you wish to uncover the truth, convert the story to its contrary; that the Greeks were routed, Troy the victor, and Penelope a harlot. On the other hand hear how fame leaves Dido, whose heart was so chaste, to be reputed a baggage, only because Vergil was not her friend. Wonder not that I am oppressed thereat, and that I speak of it at such length. I love writers and pay them what I owe; for in your world I too was a writer (xxxv. 22-28).
This strange parenthesis of satire, dubious in its humor, shocking in its irreverence, sounds today like encomium reduced to advertisement. It sounds also like the bitter retort of realism to that fiction of the courtier which was to have literary vogue through Castiglione. Bitter and foul the actual wars of Italy in contrast to old chivalry; but bitter also the trade of those who sing them.
Though often oratorical, Ariosto rarely seeks his magnificence by elaboration of style. He has too much taste, too much concern for popular appeal. He even admits the appeal of the traditional epic brag. Rodomonte alone sacks a city; Grifone throws a knight over the wall; great rocks are hurled from ships; and—triumph of rodomontade—the fragments of a combat fly up to the sphere of fire and come down lighted.
There are a few reminiscences of the Aeneid, fewer of Horace, fewest of Dante and Petrarch. Classical allusion has become a common decoration. Aurora is already obligatory for dawn. Occasionally a classical periphrasis (“Hardly had the Licaonian seed turned her plow through the furrows of heaven” xx. 82) is obscure; or there is incongruity in combining Avernus and the Sibyl with Merlin’s grotto, or the Fates with Death, Nature, and St John. But allusion is neither paraded nor often intruded. Classical similes, much more frequent than with Boiardo, are evidently sought for decoration. They are one of the signs that Ariosto’s time thought of epic in terms of style. But they are used also for vividness; and they range widely.
Besides those drawn conventionally from beasts of prey or from storm, there are many quite sharply individual: wood steaming in a fire, grass ebbing and flowing in the wind, a pile-driver, and a mine cave-in. Ariosto’s decoration is rarely a hindrance, rarely even elaborate. He is easy to read. The bearing of a passage here and there may be dubious because of looseness in the narrative, but never its meaning. He has reconciled dignity with popularity. Instead of posing as literary, he makes his readers feel literary themselves. He puts them at ease in fine company.
In sentence and stanza movement Ariosto has made his poem easy to read by diffuse and various fluency. Writing for entertainment, he uses balance or other word-play only as occasional means of variety. He is neither sententious nor pretentious. His metrical skill, remarkable in range and control, is not put forward for exhibition; it is an accompaniment so flexible to mood as constantly to enhance the connotation. Rarely lengthening the final line of the stanza often using refrain to link stanzas, and sometimes within stanzas, he is most characteristic in making the ottava rima run on not only from line to line, but from stanza to stanza. This fluent ease is by no means impromptu spontaneity. It is the work of ten years. His diffuseness, then, is not carelessness; it is adjustment alike to the immediate audience of the court and to the increasing readers of the press.
Especially significant, therefore, is his handling of description. His landscape is often both brief and conventional.
Winsome thickets of pleasant laurel, of palms and gayest myrtle, of cedar, of orange with fruit and flowers woven in forms most various and all beautiful, make a refuge from the fervid heat of summer days with their thick parasols; and among these branches in safe flight nightingales go singing
Among the purpled roses and the white lilies, which the warm air keeps ever fresh, rabbits and hares are seen at peace, and deer, heads high and proud, without fear that any one may kill or take them, feed or chew their cud at rest. Swift and nimble leap the harts and goats that abound in those country places (vi. 21-2).
But he has a way of animating convention with a sharp word of his own.
When the trembling brooks (trepidi ruscelli) began to loosen the cold ice in their warm waves (xii. 72).
Architecture and decoration often remain generalized, or offer few details. Ampler is pageantry.
With triumphal pomp and great festivity they return together into the city, which is green with branches and garlands. All the streets are hung with tapestries. A shower of herbs and flowers spreads from above and falls upon and around the victors, cast in handfuls from loggias and fair windows by ladies and damsels
In various places where they turn a corner they find improvised arches and trophies displaying pictures of the ruins and fires of Biserta and other worthy deeds; elsewhere, balconies with divers games and spectacles and mimes and plays; and at every corner is inscribed the true title: To the Liberators of the Empire.
With sound of shrill trumpets and mellow clarinets, with harmony of every instrument, mid laughter and applause, joy and favor of the people, who could hardly come close enough, the great Emperor dismounted at the palace, where several days that company stayed to enjoy itself with tournaments, personnages and farces, dances and banquets (xliv. 32-34).
This is a preciously distinct picture of actual Renaissance pageantry. More vividly detailed is the funeral of Brandimarte. Even these, however, are not dilated. They are appropriate to their narrative function. The long descriptive summary of Astolfo’s journey through the Valley of the Moon is an interlude of satire; and Orlando’s battle with the monster Orc is pure grotesque. He rows into the Orc’s mouth, casts anchor there, and, when the monster plunges, tows him ashore. Moreover, both these are narrated; neither is a descriptive pause.
Ariosto does pause, however, to dilate description of the beauty of women. Seven stanzas enumerate the charms of the enchantress Alcina (vii. 10-16).
The fair palace excelled not so much in surpassing the richness of every other as in having the most delightful folk in the world and the noblest. Little did one differ from another in flowered age and in beauty; only Alcina was most beautiful of all, as the sun is more beautiful than any star
In person she was as well formed as the industry of painters can imagine: her blond hair long and tressed; gold is not more splendid and lustrous. Rose mingled with hawthorn white spread over her delicate cheek. Of polished ivory was her joyous forehead, and of just proportion.
Beneath two black and fine-spun brows are two black eyes, as two clear suns, sympathetic in gaze, frugal in movement, about which Love seems to sport and fly, and from which he empties his quiver and visibly steals hearts. Thence descends a nose in which Envy herself could find no fault.
Beneath this, as between two valleys, the mouth besprent with native cinnabar, wherein are two rows of choice pearls, enclosed or opened by fair, sweet lips, whence issues speech of courtesy fit to soften even a base heart, and whence rises the winsome laughter that brings paradise to its place on earth [and so on for three more stanzas]
This is the conventional description called by the Middle Age blason. It is used again for Olimpia bound to the rock, where the situation itself is conventional. A bit of very old folklore, and coming down also through classical mythology as Perseus and Andromeda, it was a commonplace for dilation.
But what Ariosto dilates oftenest is emotion. His characteristic pauses are lyric. Thus he interpolates the medieval compleint d’amour not only again and again, but for long exhalations. Bradamante alone utters a whole series of these laments. The second begins as follows:
Then shall it be true (said she) that I must seek him who flees me and hides? Then must I prize him who scorns me? Must I implore him who never answers me? Shall I endure to hold at heart him who hates me, who thinks his qualities so rare that an immortal goddess must descend from heaven to kindle his heart with love?
In his pride he knows that I love him, that I adore him; nor will he of me for lover nor for slave. In his cruelty he knows that I yearn and die for him; and he waits till after death to give me help. And lest I tell him of my martyrdom, fit to move even his stubborn will, he hides himself from me, as the asp who to keep her venom refuses to hear the charm.
Ah! Love, stay him who hastes so free beyond my slow running, or restore me to the state whence thou hast taken me, when I was subject neither to thee nor to any other Alas! how deceitful and foolish is my hope that ever prayers should move thee to pity! For thou delightest to draw streams of tears from our eyes; nay, thereon thou feedest and livest (xxxii 18-20)
Substantially the same is the famous madness of Orlando. Though Ariosto cleverly gives it narrative enough to relieve its prolongation through twenty-five stanzas, it is a dilated lyric interlude.
The art that dilates these lyrics is rhetoric. Thus they answer not only the learning, but the taste of the Renaissance. With Alcina’s charms and Olimpia’s, they were the favorite passages of the Pléiade. Ronsard, using them often, was especially fond of Orlando’s madness. Beyond the Pléiade, they open a long vista toward Italian opera. To look the other way, back to the Middle Age, is to meet the sharp contradiction of Dante. Paolo and Francesca, or Ugolino, is the poetic antithesis to Bradamante and Orlando.
Such dilated interludes would interrupt any progress of the whole story; and they are not the only interpolations. Traditionally the cyclical romances might pause to add incidental stories, usually told by errant damsels seeking help. Ariosto inserts these freely, and quite as freely others having even less relevance. The story of Ginevra, Ariodante, and Polinesso (Canto V), for instance, though it falls among Rinaldo’s adventures, has its own intrigue and motivation. Equally separable, the fabliau of Fiammetta is told for sex, and prolonged by appended dialogue and comment. “Ladies,” it begins, “and you who hold ladies in esteem, for heaven’s sake give no ear to this story.... Omit this canto; for my story needs it not and will be no less clear without it.” Evidently Ariosto has not planned his cantos as chapters.
A mere glance through the summaries prefixed to each canto will show that the many interruptions are not breaks in the sequence of the whole. There is no such sequence. The poem is a collection of parallel stories taken up in turn, and only thus combined, not
